Organizationala - Grup - 2010 Curseu

Volumul_10_nr_01.
qxd
4/18/2012
3:32 PM
Page 1
Asociaia de Psihologie
Industrial i Organizaional
Editura Comunicare.ro
Psihologia Resurselor Umane

Human Resources Psychology
Psychologie des Ressources Humaines
Volumul 10, nr. 1 / 2012
Asociaia de Psihologie Industrial i Organizaional

Centrul de Monitorizare Profesional
n Psihologia Muncii Organizaional,
Universitatea Babe-Bolyai, Cluj-Napoca
Volumul_10_nr_01.qxd
4/17/2012
1:06 PM
Page 2
Asociaia de Psihologie Industrial i Organizaional

Facultatea de Comunicare i Relaii Publice
Bucureti, Str. Povernei 6, 010416
Tel./fax: 021.310.0718
Adresa web: www.apio.ro
E-mail: office@apio.ro
Copyright 2012 Asociaia de Psihologie Industrial i Organizaional

i Editura Comunicare.ro
Publicarea de articole n Revista Psihologia Resurselor Umane

este avizat de doi recenzori.
Abonamente:
Pentru membrii APIO abonamentul este inclus n cotizaia anual.
Pentru nonmembri, abonamentul anual (2 numere) cost 40 lei (taxe potale incluse).
Pentru abonamente, v rugm s consultati site-ul www.apio.ro.
CONT IBAN: RO38 BTRL 0130 1205 9213 60XX.
ISSN 1583-7327
4/17/2012
1:06 PM
Page 3
Cuprins
Editorial
Petru Lucian Cureu
Reflecii asupra sinergiei cognitive n grupurile sociale restrnse (romn) / 6
Petru Lucian Cureu
Reflecii asupra sinergiei cognitive n grupurile sociale restrnse (englez) / 13
Studii i cercetri
Elena-Mdlina Iorga, Dan Florin Stnescu, Drago Iliescu
Relaia dintre munca emoional
i burnout n cazul reprezentanilor de vnzri directe un studiu pilot / 20
Silvia Rusu, Laureniu P. Maricuoiu, Irina Macsinga, Delia Vrg, Florin A. Sava
Evaluarea personalitii din perspectiva modelului Big Five. Date privind
adaptarea chestionarului IPIP-50 pe un eantion de studeni romni / 39
Smaranda Boro, Petru Lucian Cureu
A a fi sau a nu fi ... identificat.
Explorri ale (dez)identificrii studenilor ntr-o universitate romn / 57
Claudia L. Rus, Adriana Bban, Saul N. de Jesus
nvarea n echip i eficacitatea echipelor de munc. Sinteza analizei studiilor publicate / 70
Metodologie
Marian Popa
Restricia de amplitudine, o ameninare ascuns la adresa validitii de criteriu / 91
Paul Srbescu
Folosirea analizei factoriale exploratorii
n cazul itemilor. Aspecte specifice i recomandri / 102
Recenzii
Andrei Rusu
Ticu Constantin, Pregtirea i realizarea evalurii
psihologice individuale. Norme, metodologie i proceduri, Editura Polirom, Iai, 2012 / 116
Florin Zamfirache
Clive R. Boddy, Corporate Psychopaths.
Organisational Destroyers, Palgrave Macmillan Publishers, 2011 / 117
Andreea Butucescu
Anna Koch & Karl Westhoff, Task-Analysis-Tools (TAToo)
Step-by-Step Support for Successful Job and Work Analysis, Pabst Science Publishers, 2012 / 119
Smaranda Boro
Herman Van den Broeck & David Venter,
Beyonders. Transcending Average Leadership, LannooCampus, 2011 / 121
Evenimente
Ioana Breazu
Profesorul Robert A. Roe
n Romnia psihologia organizaional la timpul viitor, septembrie 2011, Bucureti / 123
Gabriel Fischmann
Psycho-Social Health in Organizations: Research and Interventions,
International OHP Workshop, 15-17 December 2011, Timioara / 124
In memoriam
Adrian Opre
Memento prof. univ. dr. Nicolae Jurcu
sau Despre profesionalism i omenie prof. univ. dr. Nicolae Jurcu / 126
Standarde de redactare
Editorii
Psihologia resurselor umane ghid pentru autori (romn) / 128
Editorii
Psihologia resurselor umane ghid pentru autori (englez) / 132
Editorii
Psihologia resurselor umane ghid pentru autori (francez) / 136
4/17/2012
1:06 PM
Page 4
Summary
Editorial
Petru Lucian Cureu
Reflections on group cognitive synergy (romanian) / 6
Petru Lucian Cureu
Reflections on group cognitive synergy (english) / 13
Studies and researches
The relationship between emotional labor
and burnout in direct sales representatives a pilot study / 20
Personality assessment in terms of the Big Five model. Data concerning
the adaptation of the IPIP-50 questionnaire on a sample of Romanian students / 39
To be or not to beidentified.
Explorations of students' (dis)identification in a Romanian university / 57
Team learning and work team effectiveness. Summary of published studies / 70
Methodology
Marian Popa
Range restriction, a hidden threat to the criterion validity / 91
Paul Srbescu
Using exploratory factor analysis
for items. Specific issues and recommendations / 102
Reviews
Andrei Rusu
Florin Zamfirache
Andreea Butucescu
Smaranda Boro
Scientific Events
Ioana Breazu
Gabriel Fischmann
In memoriam
Adrian Opre
Standards of publishing
The Editors
Human Resources Psychology guide for authors (romanian) / 128
The Editors
Human Resources Psychology guide for authors (english) / 132
The Editors
Human Resources Psychology guide for authors (french) / 136
4/17/2012
1:06 PM
Page 5
Sommaire
ditorial
Petru Lucian Cureu
Reflexions sur la sinergie cognitive des groups socials restreints (roumain) / 6
Petru Lucian Cureu
Reflexions sur la sinergie cognitive des groups socials restreints (anglais) / 13
tudes et recherches
La relation entre le travail motionnel
et l'puisement chez les reprsentants directs des ventes une tude pilote / 20
L'valuation de la personnalit en termes du modele Big Five. Des donnes
concernant l'adaptation du questionnaire IPIP-50 sur un chantillon d'tudiants roumains / 39
Etre ou ne pas etre ... identifi.
Explorations de la (ds)identification des leves dans une universit roumaine / 57
L'appredissage de l'equipe et l'effectivit de l'equipe du travail. La synthese des tudes publies / 70
Mthodologie
Marian Popa
Restriction d'amplitude, une menace cache pour la validit de critere / 91
Paul Srbescu
L'emploi d'une analyse factorielle
exploratoire des items. Questions spcifiques et des recommandations / 102
Recensions
Andrei Rusu
Florin Zamfirache
Andreea Butucescu
Smaranda Boro
vnements scientifiques
Ioana Breazu
Gabriel Fischmann
In memoriam
Adrian Opre
Standardes de rdaction
Les diteurs
Psychologie des ressources humaines guide pour les auteurs (roumain) / 128
Les diteurs
Psychologie des ressources humaines guide pour les auteurs (anglais) / 132
Les diteurs
Psychologie des ressources humaines guide pour les auteurs (francaise) / 136
4/17/2012
1:06 PM
Page 6
Reflecii asupra sinergiei cognitive n grupurile sociale restrnse

Petru Lucian Cureu1
Practica managerial sugereaz faptul c

grupurile sunt modaliti eficiente de organizare
a muncii deoarece integreaz resurse variate
(inclusiv cunotine) ale membrilor, pentru
obinerea performanelor superioare, precum i
pentru ndeplinirea obiectivelor organizaionale,
care sunt mult prea complexe pentru a putea fi
realizate individual de ctre membrii grupului.
Cu toate astea, cercetrile sugereaz c aceast
premis nu este ntotdeauna una realist,
deoarece grupurile organizaionale genereaz
adesea conflicte i tensiuni interpersonale (De
Dreu & Weingart, 2003), favoriznd evadarea
din sarcin i delsarea (Karau & Williams,
1993). Mai mult, analiza dezbaterilor de grup
arat c participarea membrilor la dezbaterile de
grup este inegal, un numr mic de membrii
contribuie intens la dezbateri, iar ceilali membrii ai grupului au numai cteva intervenii
(Cureu, 2006a). De asemenea, istoria abund n
exemple de grupuri compuse din indivizi competeni care au luat decizii gresite, uneori cu un
impact social profund, pentru c nu au utilizat
optim cunotinele i expertiza membrilor
grupului. Cteva exemple de eecurile atribuite
prelucrrii defectuoase a informaiilor n procesul decizional de grup sunt: dezastrul navetelor
spaiale Challenger (28 Ianuarie 1986) i
Columbia (1 Februarie 2003) sau decizia luat
de Ethyl Corporation (1925) de a aditiva benzina cu plumb toxic (fapt care a schimbat atmosfera Pmntului pentru totdeauna). De aceea,
sinergia de grup, dei se presupune c apare
automat n grupurile sociale restrnse este, n
practic, mult mai efemer. Unul din interesele
principale n cercetrile pe care le-am ntreprins
pe parcursul ultimilor ani s-a concentrat asupra
modalitii prin care se poate obine sinergia
cognitiv n cadrul grupurilor sociale restrnse.
1
Cu alte cuvinte, am ncercat s descopr cum

grupurile pot depi performana medie a membrilor, sau cum, prin intermediul grupurilor, se
pot realiza sarcini pe care nici un membru nu
le-ar putea realiza, n mod individual. Scopul
acestei lucrri este de a sintetiza o parte din
cercetrile focalizate pe descoperirea metodelor
prin care s poat fi facilitat sinergia cognitiv.
Voi ncepe prin a defini sinergia cognitiv de
grup pornind de conceptul general de sinergie
de grup (Hackman, 1987; Larson, 2007) i, apoi,
voi prezenta sintetic cteva studii empirice care
au explorat antecedentele (sau ncercrile de a
favoriza apariia) sinergiei cognitive n grupurile sociale restrnse.
Definirea conceptului
de sinergie cognitiv de grup
Hackman (1987) definete sinergia de grup
ca fiind fenomenul colectiv generat de interaciunile care au loc ntre membrii grupului i
influeneaz modul n care acetia gestioneaz
oportunitile i cerinele legate de sarcini
(Hackman, 1987, p. 335). Potrivit lui Larson
(2007, 2010), sinergia de grup se refer la ctigul obiectiv (care se atribuie interaciunii de
grup) n performana de grup, n raport cu performanele individuale, nsumate. Cu alte cuvinte, sinergia de grup apare atunci cnd performana colectiv depete performana obinut
prin combinaia simpl, pre-programat, de
eforturi independente ale membrilor grupului
(Larson, 2007, 2010). Larson (2007, 2010) clasific sinergia de grup n dou categorii: sinergia
puternic i sinergia slab. Grupurile realizeaz
un efect sinergic slab atunci cnd performana
Department of Organisation Studies, Tilburg University, NL.

Adresa de coresponde: p.l.curseu@uvt.nl
6
4/17/2012
1:06 PM
Page 7
colectiv este mai bun dect performana medie a membrilor grupului i realizeaz o sinergie
puternic atunci cnd performana colectiv
depete performanele celui mai bun individ
care face parte din grup (Larson, 2007, p. 415).
n conformitate cu aceste definiii generale ale
sinergiei de grup, am putea defini sinergia cognitiv de grup ca fiind structurile cognitive la
nivel de grup care emerg din interaciunile dintre membrii si i co-evoluia structuriilor cognitive individuale (Cureu, 2006a; Cureu,
Schruijer i Boro, 2007). Prin urmare, complexitatea cognitiv a unui grup, cum este definit
n Cureu, Schruijer i Boro (2007) i operaionalizat n Cureu, Schalk i Schruijer
(2010), reprezint o msur a sinergiei cognitive
de grup.
Calitatea muncii n echip

i sinergia cognitiv de grup
Calitatea muncii n echip reflect natura
interaciunilor din cadrul grupurilor i este un
factor esenial pentru emergena sinergiei n
grupuri. Pentru a explora asocierea dintre calitatea muncii n echip i sinergia cognitiv din
cadrul grupurilor, am utilizat datele care stau la
baza studiului prezentat n Cureu, Schruijer i

Boro (2007). n cadrul acestui studiu, 132 de
studeni au fost invitai s ia parte la un exerciiu
de realizare a unei hri cognitive a conceptului
de colaborare, nti individual i apoi n grupuri
de cte trei sau patru. Complexitatea hrii cognitive individuale i de grup a fost calculat prin
considerarea gradului de difereniere i integrare
a structurilor cognitive rezultate, procedur
prezentate pe larg n Cureu, Schalk i Schruijer
(2010). Am analizat, n continuare, datele raportate n Cureu, Schruijer i Boro (2007), pentru
a ilustra diferenele ce exist n calitatea lucrului
n echip i gradul de participare, raportat de
ctre membrii grupului care au fcut parte din
grupuri fr sinergie cognitiv (complexitate
cognitiv de grup fiind mai mic dect media
complexittii cognitive individuale i mai mic
dect cea mai ridicat complexitate cognitiv
individual), grupuri n care sinergia cognitiv
este slab (complexitate cognitiv de grup fiind
mai mare dect complexitatea medie individual
din cadrul grupurilor) i grupuri n care sinergia
cognitiv este puternic (complexitatea cognitiv de grup este mai mare dect cea mai nalt
complexitate cognitiv individual). Rezultatele
pentru cele dou variabile de interaciune interpersonal sunt prezentate n Figura 1.
Figura 1. Gradul de participare i calitatea lucrului n echip n funcie de sinergia de grup (pe baza
datelor raportate n Cureu, Schruijer i Boro, 2007)
4/17/2012
1:06 PM
Page 8
Aa cum se arat n Figura 1, grupurile n

care sinergia cognitiv este puternic raporteaz
cele mai nalte scoruri pentru calitatea a muncii
n echip i participare egal, acestea fiind
urmate de grupurile n care sinergia este slab.
Am utilizat, de asemenea, testul t pentru esantioane independente pentru a testa diferenele
dintre grupuri. Grupurile cu sinergie puternic
au raportat o calitate mai mare a muncii n
echip (M=4.43, SD=.47) dect grupurile cu
sinergie slab (M=3.96, SD=.38) t=2.48,
p<.02 precum i dect cele n care nu exist
sinergie cognitiv (M=3.71, SD=.60) t=3.19,
p<.003. Grupurile cu o puternic sinergie cognitiv au raportat scoruri mai ridicate la variabila
participare egal a membrilor (M=4.35,
SD=.51) n comparaie cu grupurile care prezentau sinergie cognitive slab (M=3.84, SD=.26)
t=2.89, p<.01 sau dect cele fr sinergie
(M=3.72, SD=.51) t=3.18, p<.003. Grupurile
cu sinergie cognitiv slab au raportat o calitate
sensibil mai mare a lucrului n echip i a participrii egale, comparativ cu grupurile n care
nu exist sinergie, totui, diferenele nefiind
semnificative din punct de vedere statistic. n
aceast etap, trebui s menionm c, felul n
care a fost raportat aici comparaia nu ar trebui
utilizat pentru a face inferene cauzale,
deoarece variabile precum: sinergia cognitiv,
calitatea lucrului n echipe sau participarea
egal a membrilor nu au fost manipulate n
studiu prezent. Rezultatele raportate aici ar trebui folosite pentru a concluziona c grupurile cu
sinergie cognitiv puternic raporteaz o calitate
a muncii in echip mai mare i o participare a
mai multor membrii ai grupului dect echipele
cu o sinergie cognitiv mai slab sau cele fr
sinergie. n termeni conceptuali, aa cum am
argumentat n seciunea precedent, este mai
probabil ca participarea i calitatea interaciunilor ntr-un grup s genereze sinergie cognitiv
de grup. n aceeai ordine de idei, ntr-un alt
studiu, am artat faptul c o nalt calitate a
lucrului n echip mediaz impactul diversitii
grupului asupra complexitii cognitive de grup
(Cureu & Pluut, n press). ntr-un alt studiu
asupra creativitii de grup (Cureu, 2010) am
ilustrat faptul c procesul de planificare i conflictul legat de sarcina de lucru mediaz
impactul dispersiei i varietii de grup asupra
8
creativitii i complexitii cognitive a acestuia.

Se pare c, mai ales conflictele relaionate cu
sarcina stimuleaz creativitatea de grup care, n
schimb, este benefic sinergiei cognitive de
grup (complexitatea cognitiv de grup) (Cureu,
2010). n mod similar, am artat ntr-un alt
studiu (Cureu, Janssen & Raab, n press) faptul
c un conflict relaionat cu sarcina este benefic,
n timp ce un conflict relaional este n detrimentul complexitii cognitive de grup. De asemenea, am evideniat c att conflictele
relaionale cu sarcina, ct i cele relaionale
depind de structura reelei de comunicare din
grupuri, reele de comunicare fragmentate stimulnd conflictul legat de sarcin iar reele de
comunicare dense reducnd conflictul relaional
(Cureu, Janssen & Raab, n press). Un mod
particular de a stimula sinergia cognitiv n
grupuri este prin mbuntirea calitii interaciunilor interpersonale dintre membrii acestora. Acest aspect este n acord cu cercetrile
asupra proceselor de grup, care arat c procesele de grup (n etapele de aciune, tranziie i cele
interpersonale), sunt ntotdeauna favorabile
pentru eficacitatea grupurilor (LePine, Piccolo,
Jackson, Mathieu & Saul, 2008).
Preferinele relaionale reciproce,

calitatea muncii n echip
i sinergia cognitiv
Pn n prezent, cercetrile au explorat eficacitatea diferitelor modaliti de design a
grupurilor organizaionale, ns, ntrebarea
Cum s alctuieti echipe eficiente? nu are
nc un rspuns clar (Hodgkinson & Healey,
2008). ntr-un studiu empiric raportat n Cureu,
Kenis, Raab i Brandes (2010) am testat eficacitatea echipelor formate pe baza preferinelor
relaionale reciproce (engl. team dating). Am
folosit perspective din literatura de specialitate
privind comportamentul decizional spontan
(Ambady & Rosenthal, 1992) i formarea primei impresii (Jones, 1990) pentru a argumenta
c primele opinii dezvoltate pe baza unei scurte
interaciuni sunt predictive pentru dezvoltarea
relaiilor interpersonal viitoare. Am utilizat perspectiva reelelor sociale ca procedur analitic,
4/17/2012
1:06 PM
Page 9
Figura 2. Comparaia dintre grupurile formate prin maximizarea diversitii i cele formate prin team
dating (variabilele dependente sunt procesele de grup i strile emergente) (bazat pe datele raportate
n Cureu, Kenis, Raab & Brandes, 2010)
astfel nct s maximizm densitatea reelei sociale produse de exprimarea preferinelor relaionale reciproce ntre potenialii membrii ai
unui grup i s reducem preferinele relaionale
reciproce ntre grupuri. Rezultatele cercetrii
noastre au artat c, n comparaie cu grupurile
formate prin maximizarea diversitii, grupurile
constituite prin team dating experieniaz o mai
mare calitate a muncii n echip (procese de
grup i stri emergente) i au o performan mai
ridicat n sarcini cognitive (avnd o mai mare
complexitate cognitiv de grup) (Cureu, Kenis
et al., 2010). Un rezumat al rezultatelor privind
procesele de grup i strile emergente (utilizate
pentru evaluarea calitii muncii n echipe) sunt
prezentate n Figura 2.
n alt studiu prezentat n Cureu, Kenis &
Raab (2009) am explorat msura n care diversitatea poate conduce la conflicte relaionale, att
n grupuri randomizate, ct i n cele formate pe
baza preferinelor relaionale reciproce. Rezultatele noastre arat c diversitatea n interiorul
grupului este mai puin probabil s genereze
conflicte relaionale n grupurile formate pe
baza preferinelor relaionale dect n grupurile
formate prin randomizare. Deoarece conflictele
relaionale sunt ntotdeauna n detrimentul performanei de grup (DeDreu & Weingart, 2003),
putem concluziona c formarea echipelor prin
team dating reduce pierderile procesuale. Astfel,
ambele studii empirice rezumate mai sus arat
c preferinele relaionale mutuale dintre membrii au efect sinergic asupra grupului (prin mbuntirea calitii muncii n echip i prin
reducerea conflictelor relaionale).
De aceea, un grup perfect format pe baza
preferinelor reciproce o sa fie un grup n care
fiecare membru va exprima preferine puternice
pentru toi ceilali membri. Un astfel de grup o
s experimenteze mai puine pierderi procesuale
i o calitate superioar a muncii n echip, n
consecin, fiind mai probabil ca membrii grupului s contribuie n mod unic cu expertiza i
cunostinele lor la activitaile grupului. Aa cum
am descries n Cureu, Kenis i colab. (2010),
grupurile formate prin team dating dezvolt
structuri cognitive colective mai complexe dect
grupurile formate prin maximizarea diversitii
(care se presupune ca este benefic pentru elaborarea cunotinelor relaionate cu sarcina, vezi
Cureu, Schruijer & Boro et al., 2007). Aceste
rezultate arat c preferinele relaionale reci-
4/17/2012
1:06 PM
Page 10
proce sunt antecedente ale sinergei cognitive de

grup. Reciprocitatea preferinelor relaionale
este, de asemenea, de natur s stimuleze dezbaterile ntr-un climat de securitate psihologic
(Cureu, Kenis & Raab, 2009), iar diversitatea
de grup este mult mai probabil s declaneze
conflicte relaionate cu sarcina n echipele n
care membrii exprim preferine puternice unii
fa de ceilali dect n cele randomizate. n concluzie, formarea echipelor prin optimizarea preferinelor relaionale are, de asemenea, potenialul s stimuleze participarea la deciziile de
grup i la dezbateri. Am argumentat anterior c
participarea n grupurile formate ah-hoc este
inegal i doar unii membrii particip n mod
activ i n concluzie sinergia cognitiv este mai
puin probabil s se produc dect n grupurile
n care toi membrii particip n mod egal la
dezbaterile de grup (vezi Figura 1).
Interveniile normative
i sinergia cognitiv de grup
n procesul de luare a deciziilor
Lucrrile de referin n domeniul deciziei
collective abund n tehnici care pot fi folosite
pentru mbuntirea procesului de luare a deciziilor n cadrul grupurilor. S-a artat c intervenii normative simple pot promova sinergia n
luarea deciziilor n cadrul grupurilor (Hall &
Watson, 1967). Aceste intervenii normative sau
metoda atingerii consensului n grup, aa cum a
fost numit iniial de ctre Hall i Watson
(1967), folosete un set de reguli de baz pentru
munca de grup, iar participanii sunt instruii s
urmeze urmtoarele reguli pe parcursul interaciunilor: 1. s evite disputele privind opiniile,
perspectivele sau preferinele exprimate iniial;
2. s evite disputele n termini de pierdere sau
ctig, 3. s nu se supun opiniei majoritii
pentru a evita conflicte i pentru a atinge armonia social; 4. s nu foloseasc tehnicile de
reducere a conflictelor (de exemplu: votul
majoritii, media preferinelor individuale etc.)
5. s priveasc diferenele de opinii ca fiind naturale i constructive; 6. s priveasc cu suspiciune acordul iniial rapid (Hall & Watson, 1967,
p. 304). Scopul acestor reguli de baz cu privire
10
la atingerea unui consens este acela de a ncuraja participarea egal n cadrul grupurilor i de a
promova, n cele din urm, sinergia de grup. n
cadrul unui studiu empiric prezentat n Cureu i
Schruijer (in press), am testat impactul acestor
intervenii normative asupra raionamentului
grupului i asupra complexitii cognitive a grupului. n primul studiu empiric, demonstrm c
interveniile normative conduc la reprezentri
cognitive colective mai complexe, iar efectul
este mult mai puternic n cadrul grupurilor cu
istorie (stabile) dect n cadrul grupurilor formate ad-hoc. n cel de-al doilea studiu empiric,
am testat impactul acelorai intervenii normative asupra raionalitii grupului. Definim
raionalitatea de grup ca fiind acel nivel de competen de grup ce rezult din competenele
individuale i procesul de interaciune din cadrul grupurilor. Studiul cvasi- experimental utilizeaz un set de zece sarcini de luare a deciziilor prezentate n Cureu (2006a) pentru a putea
evalua raionalitatea la nivel individual mai
nti, iar apoi, de grup, solicitndu-se participanilor s foloseasc aceleai situaii decizionale, n cadrul unor grupuri formate din 4 pn
la 7 membrii. Datorit design-ului, putem evalua att sinergia slab, ct i sinergia puternic
(aa cum au fost ele definite de ctre Larson,
2007). Astfel, putem estima n continuare impactul interveniilor normative pentru ambele
tipuri de sinergie cognitiv (aceaste analize nu
au fost raportate n lucrarea iniial). Sinergia
cognitiv slab a fost calculat prin scderea
scorului raionalitii de grup (rezultatul grupului la sarcina lurii deciziilor) din raionalitatea
individual medie, n timp ce sinergia cognitiv
de grup puternic a fost calculat scznd rezultatul grupului din cel mai mare scor individual
realizat. Rezultatul comparaiei dintre grupurile
crora le-au fost aplicate interveniile normative
i cele care nu au fost supuse interveniilor este
reprezentat n Figura 3. Aa cum rezult din
Figura 3, grupurile care au fost supuse interveniilor normative au scoruri mai ridicate ale
sinergiei cognitive slabe comparativ cu grupurile care nu au fost supuse niciunei intervenii
(t=3.96, p<.001). Mai mult dect att, diferena
dintre grupurile cu/ fr intervenii normative
este semnificativ i n cazul sinergiei cognitive
puternice (t=2.90, p<.006). Cu toate acestea,
4/17/2012
1:06 PM
Page 11
Figura 3. Comparaia dintre grupurile cu i fr intervenii normative (bazat pe datele raportate n

Cureu i Schruijer, in press)
ambele scoruri sunt negative, dup cum rezult

din Figura 3. Aceast nseamn, n termeni absolui, c interveniile normative genereaz sinergie slab deoarece cel mai bun individ din
cadrul grupului ntrece grupul n ambele situaii
(cu sau fr intervenii normative). Sinergia
puternic este realizat doar n cazul n care
grupul obine un scor mai bun dect cel mai bun
individ din cadrul grupului. Acest model al
rezultatelor deschide un nou domeniu de cercetare i anume, dezacordul minoritii ca metod
de obinere a sinergiei cognitive. Am artat faptul c, grupurile care experimenteaz dezacordul minoritar (sau influen minoritar un
membru al grupului exprim opinii diferite de
cele ale majoritii) dovedesc o complexitate
cognitiv mai mare dect cele n care acest
dezacord nu exist (Cureu, Schruijer & Boros,
in press). Dezacordul pare s ncurajeze activitatea cognitiv n cadrul grupurilor i acest fapt
poate constitui o strategie util (alturi de participarea egal) de a obine sinergia cognitiv
puternic n cadrul grupurilor. Membrii grupurilor (n special cei mai informai) ar trebui
stimulai s participe i s contribuie la dezbaterile grupurilor (ca opozani, critici) i aceasta
ar conduce, n cele din urm, la sinergie (puter-
nic) n cadrul grupurilor, dat fiind faptul c sinergia cognitiv puternic este obinut doar dac
grupul, ca ntreg, i surclaseaza cel mai valoros
membru.
Concluzii
Sinergia cognitiv nu e un dat n cadrul
grupurilor sociale restrnse. Performana superioar a grupurilor este, ns, rezultatul interaciunilor sinergice dintre membrii grupului. De
aceea, n abordarea prezentat n Larson (2010),
considerm c este esenial s nelegem
condiiile n care sinergia cognitiv se nate n
cadrul grupurilor mici. Aceast lucrare analizeaz bazele conceptuale ale sinergiei cognitive
de grup i prezint, pe scurt, cteva metode prin
care sinergia poate fi obinut n cadrul
grupurilor mici. n prima parte, am artat c sinergia de grup este asociat n mod pozitiv cu calitatea lucrului n echip i cu participarea egal
a membrilor grupului. n continuare, am argumentat c preferinele relaionale reciproce pot
fi folosite eficient pentru a spori calitatea muncii
n echip i, deci, pentru a obine sinergie cogni-
11
4/17/2012
1:06 PM
Page 12
tiv n cadrul grupului. Mai mult dect att, am

artat c interveniile normative sunt modaliti
simple i eficiente pentru a obine sinergia cognitiv n cadrul grupurilor (mai degrab sinergie
slab dect puternic). n concluzie, cercetrile
ulterioare ar trebui s analizeze interaciunea
dintre diferena de opinie i participarea egal,
ca antecedente ale sinergiei cognitive n cadrul
grupurilor.
Referine bibliografice
Ambady, N. & Rosenthal, R. (1992). Thin slices of
expressive behavior as predictors of interpersonal
consequences: A meta-analysis. Psychological
Bulletin, 111, 2, 256-274.
Cureu, P. L. (2006a). Emergent states in virtual teams:
A complex adaptive systems perspective. Journal
of Information Technology, 21, 249-261.
Cureu, P. L. (2006b). Need for cognition and rationality in decision-making. Studia Psychologica, 48,
2, 141-156.
Cureu, P.L., Schruijer, S.G.L. & Boro, S. (2007). The
effects of groups variety and disparity on groups
cognitive complexity. Group Dynamics: Theory,
Research and Practice, 11, 3, 187-206.
Cureu, P.L., Kenis, P. & Raab, J. (2009). Reciprocated
relational preferences and intra-team conflict,
Team Performance Management, 15, 1, 18-34.
Cureu, P.L., Schalk, R. & Schruijer, S.G.L. (2010). The
use of cognitive mapping in eliciting and evaluating group cognition. Journal of Applied Social
Psychology, 40, 1258-1291.
Cureu, P.L., Kenis, P., Raab, J. & Brandes, U. (2010).
Composing effective teams through team-dating.
Organization Studies, 31, 7, 873894.
Cureu, P. L. (2010). Team creativity in web site design.
An empirical test of a systemic model. Creativity
Research Journal, 22, 1, 98-107.
Cureu, P. L. & Schruijer, S. G. L. (in press). Normative
interventions, emergent cognition and decision
rationality in ad-hoc and established groups. Management Decision, 50, 6.
12
Cureu, P. L. & Schruijer, S. G. L. & Boro, S. (in

press). Socially rejected while cognitively successful: The impact of minority dissent on group
cognitive complexity. British Journal of Social
Psychology.
Cureu, P. L. & Pluut, H. (in press). Student groups as
learning entities: The effect of group diversity and
teamwork quality on groups cognitive complexity. Studies in Higher Education.
Cureu, P. L., Janssen, S. & Raab, J. (in press). Connecting the dots: Social network structure, conflict, and
Group Cognitive Complexity. Higher Education
De Dreu, C. K. W. & Weingart, L. R. (2003). Task versus relationship conflict, team performance, and
team member satisfaction: A meta-analysis. Journal of Applied Psychology, 88: 741-749.
Hackman, J. R. (1987) The design of work teams, in J.
W. Lorsch (Ed.) Handbook of Organizational
Behavior, Englewood Cliffs: Prentice Hall, pp.
315-342.
Hall, J. & Watson, W. H. (1967). The effects of a normative intervention on group decision-making
performance. Human Relations, 23, 299-317.
Hodgkinson, G. P. & Healey, M. P. (2008). Toward a
(pragmatic) science of strategic intervention:
Design propositions for scenario planning. Organization Studies, 29, 435-457.
Jones, E. E. (1990). Interpersonal perception. New
York: W. H. Freeman and Co.
Karau, S. J., & Williams, K. D. (1993). Social loafing:
A meta-analytic review and theoretical integration.
Journal of Personality and Social Psychology, 65,
681-706.
Larson, J. R., Jr. (2007). Deep diversity and strong synergy: Modeling the impact of variability in members problem-solving strategies on group problem-solving performance. Small Group Research,
38, 413-436.
Larson, J. R. Jr. (2010). In search of synergy in small
group performance. Psychology Press, New York,
US.
LePine, J. A., Piccolo, R. F., Jackson, C. L., Mathieu, J.
E., & Saul, J. R. (2008). A meta-analysis of teamwork processes: Tests of a multidimensional
model and relationships with team effectiveness
criteria. Personnel Psychology, 61, 273307.
4/17/2012
1:06 PM
Page 13
Reflections on group cognitive synergy

Petru Lucian Cureu1
Common managerial wisdom suggests that

groups are effective ways of organizing work
because they integrate various resources
(including knowledge) of their members in
order to achieve superior performance and
accomplish organizational goals that are too difficult for their individual members. However,
research to date, suggests that this assumption
might not always be realistic as organizational
groups are hotbeds for conflict and turmoil (De
Dreu & Weingart, 2003), are conducive for free
riding behaviors and social loafing (Karau &
Williams, 1993) and individual participation in
small group settings is often skewed with only a
few members contributing a lot and the majority of the group members having only few interventions during group debates (Cureu, 2006a).
Moreover, historical evidence abounds with
examples of groups composed of bright individuals that made (collectively) faulty decisions,
sometimes with pervasive social impact because
they failed to make use of group members
knowledge and expertise. Some of the failures
attributed to defective information processing in
decision-making groups include the Challenger
(28 January 1986) and Columbia (1 February
2003) Space Shuttle disasters, and the decision
made by the Ethyl Corporation (1925) to add
toxic lead to gasoline (which changed Earths
atmosphere forever). Therefore, group synergy,
although assumed to emerge automatically in
small groups, is in practice, more difficult to
achieve. One of my core research interests was
to uncover how cognitive synergy can be
achieved in small group settings. In other words,
I tried to explore how groups can surpass the
average performance of their members or
achieve things that no individual member alone
could achieve. The aim of this paper is to sum-
marize some of my research attempts to find out

ways in which cognitive synergy can be facilitated. I begin by defining group cognitive synergy starting from the general group synergy concept (Hackman, 1987; Larson, 2007) and then
summarize a few empirical studies that explored
the antecedents (or attempts to foster the emergence) of cognitive synergy in small groups.
Group cognitive synergy defined

Hackman (1987) defines group synergy as
group level phenomena that emerges from the
interactions among members and affect how
well a group deals with the task-related demands
and opportunities (Hackman, 1987 p. 335).
According to Larson (2007, 2010), group synergy refers to an objective gain in group performance as compared to summed individual performances that is attributable to group
interaction. In other words, group synergy is
achieved when the collective performance of
interacting individuals exceeds the performance
achieved by simple, preprogrammed combination of standalone group member efforts (Larson, 2007, 2010). Larson (2007, 2010) further
differentiates between strong and weak synergy.
Groups achieve weak synergy when collective
performance is better than the average performance of group members and strong synergy,
when collective performance exceeds the performance of the best performing individual in the
group (Larson, 2007, p. 415). In line with these
general definitions of group synergy, I define
group cognitive synergy as group level cognitive
structures that emerge from the interactions
among members and the co-evolution of their
Department of Organisation Studies, Tilburg University, NL.

13
4/17/2012
1:07 PM
Page 14
individual cognitive structures (Cureu, 2006a;

Cureu, Schruijer and Boro, 2007). Therefore,
group cognitive complexity as defined in
Cureu, Schruijer and Boro (2007) and operationalized in Cureu, Schalk and Schruijer
(2010) is a measure of group cognitive synergy.
Teamwork quality
and group cognitive synergy
Teamwork quality reflects the nature of
interpersonal interactions within groups and it is
closely related to group synergy. I explore further on the association between teamwork quality and cognitive synergy in groups using the data
reported in Cureu, Schruijer and Boro (2007).
In this study, one hundred and thirty two students
were asked to take part in a cognitive mapping
exercise first individually and then in groups of
three or four. The complexity of individual and
group cognitive maps was computed using the
integrative cognitive complexity framework
extensively presented in Cureu, Schalk and
Schruijer (2010). I have further analyzed the data
reported in Cureu, Schruijer and Boro (2007)
to illustrate differences in teamwork quality and
equal participation as reported by group mem-
bers as they were part of groups with no synergy

(group cognitive complexity is lower than the
average individual cognitive complexity and
lower than the highest individual cognitive complexity), weak cognitive synergy (group cognitive complexity is higher than the average individual complexity within groups) and strong
cognitive synergy (GCC is higher than the highest individual cognitive complexity). The results
for the two interpersonal interaction variables
are presented in Figure 1.
As shown in Figure 1, groups with strong
cognitive synergy report the highest levels of
teamwork quality and equal participation, followed by groups with weak synergy. I also used
the t test for independent samples to test the
paired differences. Groups with strong synergy
reported better teamwork quality (M=4.43,
SD=.47) than the groups with weak cognitive
synergy (M=3.96, SD=.38) t=2.48, p<.02
and no cognitive synergy (M=3.71, SD=.60)
t=3.19, p<.003. Moreover, groups with strong
synergy reported higher scores for equal participation (M=4.35, SD=.51) as compared to
groups with weak cognitive synergy (M=3.84,
SD=.26) t=2.89, p<.01 and groups with no
synergy (M=3.72, SD=.51) t=3.18, p<.003.
Finally, groups with weak cognitive synergy
reported better teamwork quality and equal par-
Figure 1. Equal participation and teamwork quality as a function of group synergy (based on data
reported in Cureu, Schruijer and Boro, 2007)
14
4/17/2012
1:07 PM
Page 15
ticipation as compared to groups with no cognitive synergy, yet the difference was not statistically significant. It must be stated at this stage
that the way the comparison is reported here
should not be used to make causal inferences, in
that neither cognitive synergy nor teamwork
quality or equal participation were manipulated
in this study. The results as reported here should
be used to simply draw the conclusion that
groups with strong cognitive synergy report better teamwork quality and equal participation
when compared to groups with weak or no cognitive synergy. In conceptual terms, as argued in
the previous section, it is more likely that participation and quality of interpersonal interactions
(teamwork quality) generate cognitive synergy
in groups. In line with this argument, we show
in another paper that teamwork quality mediates
the impact of different forms of group diversity
on group cognitive complexity (Cureu & Pluut,
in press). Moreover, in a study on group creativity (Cureu, 2010) I show that task conflict and
planning processes mediate the impact of group
disparity and group variety on group creativity
and group cognitive complexity. Task conflict in
particular seems to foster group creativity,
which in turn is beneficial for group cognitive
synergy (group cognitive complexity) (Cureu,
2010). Similarly we show in another study

(Cureu, Janssen & Raab, in press) that task
conflict is beneficial, while relationship conflict
is detrimental for group cognitive complexity.
We also show that both task and relationship
conflict depend on the structure of the communication network within groups, with network
fragmentation stimulating task conflict and network density reducing relationship conflict
(Cureu, Janssen & Raab, in press). A particular
way of stimulating cognitive synergy in groups
is by improving the quality of interpersonal
interactions in groups. This particular claim is
also in line with research on group processes,
showing that action phase, transition phase and
interpersonal processes are all conducive for
group effectiveness (LePine, Piccolo, Jackson,
Mathieu & Saul, 2008).
Reciprocated relational
preferences, teamwork quality
and cognitive synergy
Research to date explored the effectiveness
of various design strategies for organizational
groups, yet the question of how to compose
Figure 2. Comparison between groups formed by maximizing within group diversity and team dating
teams (dependent variables are group processes and emergent states) (based on the data reported in
Cureu, Kenis, Raab & Brandes, 2010)
15
4/17/2012
1:07 PM
Page 16
effective groups has no unequivocal or definite

answer (Hodgkinson & Healey, 2008). In an
empirical study reported in Cureu, Kenis, Raab
and Brandes (2010) we test the effectiveness of
team dating (a team formation strategy based on
optimizing reciprocated relational preferences)
as a team design strategy. We used insights from
the thin-slices of behaviour literature (Ambady
and Rosenthal 1992) and first impression formation (Jones 1990) to argue that first opinions
developed based on short interpersonal interaction are predictive for the development of interpersonal relations in the future. We use a social
network analytic framework to optimize
expressed relational preferences among potential group members in such a way that reciprocated preferences are maximized within groups
and minimized between groups. Our results
show that as compared to groups formed by
maximizing diversity, group composed based on
expressed relational preferences (team dating
groups) experience higher teamwork quality
(group processes and emergent states) and perform better in cognitive tasks (have higher
group cognitive complexity) (Cureu, Kenis et
al., 2010). A summary of the results for group
processes and emergent states (used to evaluate
teamwork quality) is presented in Figure 2.
In another study reported in Cureu, Kenis
and Raab (2009) we explore the extent to which
diversity leads to relationship conflict in groups
formed by randomly assigning group members
to groups and groups formed through team dating. Our results show that group diversity is less
likely to generate relationship conflict in team
dating teams than in randomized groups. As
relationship conflict is always detrimental for
group performance (DeDreu & Weingart, 2003)
we can conclude that team dating reduces process losses in groups. All in all, both empirical
studies (on reciprocated relational preferences)
summarized above show that maximizing reciprocated relational preferences in groups has synergetic effects (improves teamwork quality and
reduces the emergence of relationship conflict).
The basic principle of team dating is that it uses
a social network framework to maximize the
reciprocated relational preferences within
groups. Therefore, a perfect group formed
using team dating would be a group in which
16
each member expresses strong relational preferences for all the other members in the group.
Such a group is likely to experience few process
losses and a higher quality of teamwork interactions and as a consequence group members are
more likely to contribute with their unique
knowledge and specific expertise to the group
task. As reported in Cureu, Kenis et al (2010),
team dating groups develop more complex collective cognitive structures as compared to
groups formed by maximizing diversity. These
results support the claim that reciprocated relational preferences are antecedents of cognitive
synergy in groups. Reciprocated relational preferences are also likely to stimulate debate
(Cureu, Kenis & Raab, 2009), as group diversity is more likely to trigger task conflict in team
dating rather than randomized groups. This
means that team dating has also the potential to
stimulate participation to group discussion and
group debates. I argued that participation in adhoc groups is skewed and in such groups in
which only a few members participate, cognitive synergy is less likely to occur than in groups
in which all members participate equally to
group debates (see Figure 1).
Normative interventions
and group cognitive synergy
in decision-making tasks
Literature on group decision making
abounds with techniques that can be used to
improve decision quality in groups. Simple normative interventions have been shown to foster
synergy in decision-making groups (Hall &
Watson, 1967). These normative interventions
or the group consensus method as initially
labeled by Hall and Watson (1967) uses a set of
ground rules for group work and participants are
instructed to: (1) avoid arguing about initial
opinions, views and preferences, (2) avoid
win-lose statements, (3) avoid conforming to
a majority in order to reduce conflict and reach
social harmony, (4) avoid conflict-reducing
techniques (e.g., majority vote, average individual preferences, etc), (5) view differences of
4/17/2012
1:07 PM
Page 17
opinion as natural and helpful, (6) view initial

agreement as suspect (Hall & Watson, 1967,
p.304). The aim of these ground rules for
achieving consensus is to reinforce equal participation in groups and eventually foster group
synergy. In an empirical study reported in
Cureu and Schruijer (in press) we test the
impact of these normative interventions on
group rationality and group cognitive complexity. In the first empirical study we show that normative interventions lead to more complex collective cognitive representations and the effect
is stronger in established rather than ad-hoc
groups. In the second empirical study we test the
impact of the same normative interventions on
group rationality. We define group rationality as
an emergent group level competence that results
from individual competencies and the interaction processes within groups. The quasi experimental study uses a set of ten decision tasks
described in Cureu (2006a) to first evaluate
individual rationality and then participants are
asked to perform the same decision tasks in
established groups having 4 to 7 members. Due
to the design we can evaluate both weak as well
as strong synergy (as defined by Larson, 2007).
As such we can further estimate the impact of
normative interventions on both measures of
cognitive synergy (these analyses were not
reported in the original paper). Weak cognitive

synergy was computed by subtracting the score
of group rationality (the results of the group
decision task) from the average individual rationality, while strong group synergy was computed by subtracting the group score from the highest individual score. The results of the
comparison between the groups that received
the normative intervention and those that did not
have any intervention, is presented in Figure 3.
As depicted in Figure 3, groups that
received normative intervention have higher
scores on weak synergy as compared to groups
that did not receive normative intervention
(t=3.96, p<.001). Moreover, the difference
between groups with and groups without normative intervention is significant also for strong
synergy (t=2.90, p<.006), nevertheless, both
scores are negative as depicted in Figure 3. This
means that in absolute terms, normative interventions generate weak but not strong synergy
because the best individual in the group outperforms the group in both conditions (with and
without normative interventions). Unless the
group does better than the best performing individual in the group, strong synergy is not
achieved. This pattern of results opens a new
venue for research namely minority dissent as a
way to achieve group cognitive synergy. We
Figure 3. The comparison between groups with and groups without normative interventions (based on
the data reported in Cureu and Schruijer, in press)
17
4/17/2012
1:07 PM
Page 18
showed that groups experiencing minority dissent exhibit greater cognitive complexity when
compared to groups without minority dissent
(Cureu, Schruijer & Boro, in press). Dissent
seems to foster cognitive activity in groups and
it could be a useful strategy (in addition to equal
participation) to achieve strong cognitive synergy in groups. Group members (especially the
knowledgeable ones) should be stimulated to
participate and contribute to the group debates
(as dissenters) and this would ultimately lead to
(strong) cognitive synergy in groups, as strong
cognitive synergy is achieved only if the group
as a whole outperforms its worthiest member.
Conclusions
Cognitive synergy is not a given in small
group settings. Superior group performance is
nevertheless the result of synergetic interactions
among group members. Therefore, in line with
Larson (2010), I argue that it is essential to
understand the conditions under which cognitive synergy occurs in small groups. This paper
explored the conceptual foundations of group
cognitive synergy and shortly reviewed a few
ways in which cognitive synergy can be
achieved in small groups. I show that group cognitive synergy is positively associated with
teamwork quality and equal participation in
groups. Further on I argue that reciprocated relational preferences can effectively be used to foster teamwork quality and ultimately cognitive
synergy in groups. Moreover, I show that normative interventions are simple and effective
ways of achieving weak (rather than strong)
cognitive synergy in groups. I conclude that further research should explore the interplay
between dissent and equal participation as
antecedents of cognitive synergy in groups.
References
Ambady, N. & Rosenthal, R. (1992). Thin slices of
expressive behavior as predictors of interpersonal
consequences: A meta-analysis. Psychological
Bulletin, 111, 2, 256-274.
18
Cureu, P. L. (2006a). Emergent states in virtual teams:

A complex adaptive systems perspective. Journal
of Information Technology, 21, 249-261.
Cureu, P. L. (2006b). Need for cognition and rationality in decision-making. Studia Psychologica, 48,
2, 141-156.
Cureu, P.L., Schruijer, S.G.L. & Boro, S. (2007). The
effects of groups variety and disparity on groups
cognitive complexity. Group Dynamics: Theory,
Research and Practice, 11, 3, 187-206.
Cureu, P.L., Kenis, P. & Raab, J. (2009). Reciprocated
relational preferences and intra-team conflict,
Team Performance Management, 15, 1, 18-34.
Cureu, P.L., Schalk, R. & Schruijer, S.G.L. (2010). The
use of cognitive mapping in eliciting and evaluating group cognition. Journal of Applied Social
Psychology, 40, 1258-1291.
Cureu, P.L., Kenis, P., Raab, J. & Brandes, U. (2010).
Composing effective teams through team-dating.
Organization Studies, 31, 7, 873894.
Cureu, P. L. (2010). Team creativity in web site design.
An empirical test of a systemic model. Creativity
Research Journal, 22, 1, 98-107.
Cureu, P. L. & Schruijer, S. G. L. (in press). Normative
interventions, emergent cognition and decision
rationality in ad-hoc and established groups. Management Decision, 50, 6.
Cureu, P. L. & Schruijer, S. G. L. & Boro, S. (in
press). Socially rejected while cognitively successful: The impact of minority dissent on group
cognitive complexity. British Journal of Social
Psychology.
Cureu, P. L. & Pluut, H. (in press). Student groups as
learning entities: The effect of group diversity and
teamwork quality on groups cognitive complexity. Studies in Higher Education.
Cureu, P. L., Janssen, S. & Raab, J. (in press). Connecting the dots: Social network structure, conflict,
and Group Cognitive Complexity. Higher Education
De Dreu, C. K. W. & Weingart, L. R. (2003). Task versus relationship conflict, team performance, and
team member satisfaction: A meta-analysis. Journal of Applied Psychology, 88: 741-749.
Hackman, J. R. (1987) The design of work teams, in J.
W. Lorsch (Ed.) Handbook of Organizational
Behavior, Englewood Cliffs: Prentice Hall, pp.
315-342.
Hall, J. & Watson, W. H. (1967). The effects of a normative intervention on group decision-making
performance. Human Relations, 23, 299-317.
Hodgkinson, G. P. & Healey, M. P. (2008). Toward a
(pragmatic) science of strategic intervention:
Design propositions for scenario planning. Organization Studies, 29, 435-457.
Jones, E. E. (1990). Interpersonal perception. New
York: W. H. Freeman and Co.
4/17/2012
1:07 PM
Page 19
Karau, S. J., & Williams, K. D. (1993). Social loafing:

A meta-analytic review and theoretical integration.
Journal of Personality and Social Psychology, 65,
681-706.
Larson, J. R., Jr. (2007). Deep diversity and strong synergy: Modeling the impact of variability in members problem-solving strategies on group problem-solving performance. Small Group Research,
38, 413-436.
Larson, J. R. Jr. (2010). In search of synergy in small

group performance. Psychology Press, New York,
US.
LePine, J. A., Piccolo, R. F., Jackson, C. L., Mathieu, J.
E., & Saul, J. R. (2008). A meta-analysis of teamwork processes: Tests of a multidimensional
model and relationships with team effectiveness
criteria. Personnel Psychology, 61, 273307.
19
4/17/2012
1:07 PM
Page 20
The relationship between emotional labor and burnout

in direct sales representatives a pilot study
Elena-Mdlina Iorga1, Dan Florin Stnescu2, Drago Iliescu3
Abstract
Special attention has been paid lately to the emotional component of the job, especially concerning people
work jobs, such as health care, social services work, teaching or sales. On the other hand, affective outcomes
have been linked to burnout. Therefore, the current study seeks to investigate the extent to which emotional labor
strategies result in burnout for employees who work in direct sales. The study was conducted using a survey
method of 107 bank tellers. The results confirmed the fact that expressing organizationally desired emotions
while interacting with customers is emotionally taxing. Significant correlations were found not only concerning
the scores for emotional labor strategies and core burnout, but also between the respective subscales. The regression analysis indicated the intensity of emotions being regulated in service encounters as the best predictor for
burnout in the case of the sample in discussion. Likewise, in a second model, the variety of emotions required in
work settings as part of the work role together with the intensity subscale account for approximately 20% of the
variance in burnout. Performance did not show the expected relationship with burnout and emotional labor. These
findings represent a basis for new antecedents on the implications of emotional labor in the Romanian workplace.
Keywords: emotional labor, burnout, COR, emotional dissonance, emotional intelligence, people work, performance
Rsum
La dimension motionnelle du lieu de travail a t rcemment en premier plan, notamment en termes de professions qui impliquent des interactions directes avec dautres personnes, comme ces du domaine mdical, de
lassistance sociale, de lducation et des ventes. Dautre part, les effets affectifs ont t associs avec le burnout.
A partir de ces questions, la prsente tude essaye de dterminer comme lutilisation des stratgies du travail motionnel peut conduire le burnout. Ltude a t accomplie par enqute base de questionnaire pour 107
employs de la banque. Les rsultats ont confirm qui laffichage des motions dsires par lorganisation dans
linteraction avec les clients est motionnellement coteux. Ont trouv des corrlations significatives dans les
scores pour les deux stratgies du travail motionnel et burnout et aussi pour les sous-chelles.
Lintensit des motions affiches au travail a t le meilleur prdicteur du burnout, comme il t rvl par
lanalyse de rgression. Tout, la varit des motions avec lintensit des motions qui sont demandes au travail
expliquent environ 20% de la variance pour burnout. Performance na pas montr la relation attendue avec le
burnout et le travail motionnel. Ces rsultats reprsentent un point de dpart pour des nouveaux antcdents sur
les implications de lutilisation du travail motionnel dans lenvironnement organisationnel roumaine.
Mots-cls : travail motionnel, burnout, la thorie de conservation des ressources, dissonance motionnelle,
intelligence motionnelle, professions impliquant des interactions directes avec dautres personnes, performance
College of Communication and Public Relations, NSPSAS.

Adresa de coresponden: iorgaem@yahoo.com
2
Adresa de coresponden: dan.stanescu@comunicare.ro
3
Adresa de coresponden: dragos.iliescu@comunicare.ro
20
4/17/2012
1:07 PM
Page 21
The relationship between emotional labor and burnout in direct sales representatives a pilot study
Rezumat
Componenta emoional a locului de munc a fost n prim-plan n ultima perioad, n special n ceea ce
privete ocupaiile care presupun interaciuni directe cu alte persoane, precum cele din domeniul medical, al asistenei sociale, educaional sau al vnzrilor. Pe de alt parte, efectele de ordin afectiv au fost asociate cu burnoutul. Pornind de la aceste aspecte, studiul de fa ncearc s determine msura n care utilizarea strategiilor de
munc emoional poate duce la burnout n cazul celor care lucreaz n vnzri directe. Studiul a fost realizat utiliznd ancheta pe baz de chestionar n cazul a 107 de agenti bancari. Rezultatele au confirmat faptul c afiarea
emoiilor dorite de organizaie n cadrul interaciunii cu clienii este costisitoare din punct de vedere afectiv. S-au
gsit corelaii semnificative nu numai la nivelul scorurilor pentru strategiile de munc emoional i burnout ci i
la nivelul subscalelor. Analiza de regresie a indicat intensitatea emoiilor reglate n cazul interaciunilor de la locul
de munc drept cel mai bun predictor pentru burnout n cazul eantionului analizat. n aceeai msur, n cadrul
unui al doilea model, varietatea emoiilor cerute la locul de munc mpreun cu intensitatea acestora explic
aproximativ 20% din varian. n cazul performanei nu s-au nregistrat corelaii semnifictative cu burnoutul i
munca emoional. Aceste rezultate reprezint un punct de plecare pentru noi antecedente privind implicaiile pe
care utilizarea muncii emoionale le are n mediul organizaional romnesc.
Cuvinte cheie: munc emoional, burnout, teoria conservrii resurselor, disonan emoional, inteligen
emoional, ocupaii care presupun interaciuni directe cu alte persoane, perfoman
Introduction
Emotions in work settings represented a
recurrent issue in the 1930s, but from this point
forward the interest towards the emotional
dynamics in work and organizations diminished.
Despite that, Rafaeli, Semmer and Tschan (n.d.)
note that there were two exceptions from the
main rationale-cognitive approach of organizational behavior: the study of job satisfaction,
which was seen as an affective reaction and the
research on stress at work, which was primarily
related to negative emotions (Lazarus & CohenCharash, 2001). The end of the 20th century was
marked by such a surge in the academic interest
in emotions in the workplace, that Barsade,
Brief and Spataro proclaimed the affective revolution (Rafaeli et al., n.d.).
One of the pillars of this revolution was the
concept of emotional labor, which was first conceptualized by the American sociologist Arlie
Hochschild (1979, 2003). Once launched in the
academic market it became a privileged issue.
Researchers seemed to try and determine
whether emotional labor was to open new directions in understanding organizational behavior
or it was just another buzz word (Hunter &
Smith, 2007). Therefore, beginning with The
Managed Hearth (Hochschild, 1983) a wide
range of studies were conducted in order to
describe the concept of emotional labor and its
outcomes, among which the most frequently

approached were burnout or job satisfaction and
performance (Ashforth & Humphrey, 1993;
Morris & Feldman, 1996; Grandey, 2000, 2003;
Brotheridge & Grandey, 2002; Brotheridge &
Lee, 2003; Pugliesi, 1999 etc).
Emotional labor was initially studied in the
service context, which was seen as the prototypical emotional labor work (Brotheridge &
Grandey, 2002). Adopting Goffmans dramaturgical perspective, Hochschild considered service encounters as a performance directed by
the organization (Grandey, 2003). This performance implies impression management in order
to achieve organizational goals, such as client
satisfaction and return business. Since the interpersonal aspect of the job is taken into consideration (the customer-client relationship) and as
emotions are an important part of communication, the management of feeling to create a
publicly observable facial and bodily display
urges (Hochschild, 2003, p. 7). The main difference between emotion work the term proposed
by Hochschild in a first stage of her study
(1979) and emotional labor is the fact that the
later has exchange value, since it is sold for a
wage, while the former is used in private life
and therefore has use value.
Furthermore, Hochschild proposed two
main ways of managing emotions in the workplace in order to meet the specified display
21
4/17/2012
1:07 PM
Page 22
rules: surface acting (regulating the emotional

expressions) and deep acting (modifying inner
feelings) (Hochschild, 2003). These processes
are effortful and, when organizations get to
control something as personal as employees
emotions, the commoditization of emotions
intervenes and it can lead to unpleasant consequences (Hochschild, 1979).
As one can see from the discussion above,
Hochschilds conceptualization of emotional
labor mainly refers to the management of feelings. Ashforth and Humphrey (1993), in
exchange, are more concerned about the expressions. They define emotional labor as the act of
displaying appropriate emotions with the goal to
engage in a form of impression management for
the organization (Ashforth & Humphrey, 1993,
p. 89). Hence, the focus moves on the observable behavior.
Morris and Feldman (1996) moved even
further and approached the concept of emotional labor taking into account the characteristics of
the job. From an interactional point of view,
they described emotional labor as the effort,
the planning and control needed to express organizationally desired emotions during interpersonal transactions (Morris & Feldman, 1996, p.
987). Emotional labor was seen as consisting of
four dimensions: the frequency of interactions,
attentiveness (intensity of emotions and duration of interaction), variety of emotions required
representing the organizational expectations
for employees in their interactions with clients,
and emotional dissonance (an internal state of
tension due to the demand of performing emotional labor regardless of what the employee
truly feels).
Morris and Feldmans description of the
dimensional structure of emotional labor was
contended by Kruml and Geddes (2000), who
claimed that some dimensions, such as the frequency or the duration of service interactions
are rather job characteristics and not emotional
labor as such. Instead, Kruml and Geddes supported in their study (2000) a two-dimensional
structure of the construct, viewed as emotive
dissonance and emotive effort. In fact, the multidimensional structure of emotional labor was
posited even earlier, by Wharton (1993). Brotheridge and Lee, in their conceptualization and
22
scale development (1998), viewed emotional

labor as consisting in six dimensions: duration
and frequency of service encounters, variety and
intensity of emotions displayed and deep and
surface acting as emotion regulation strategies.
Considering all the theories on emotional
labor presented up to now, Grandey (2000)
notes that, despite the differences in defining the
process, there is a commonality: the assumption
that emotional labor involves regulating ones
emotions and emotional expressions in order to
be consistent with organizational display rules.
On this basis, Grandey (2000) formulated an
integrated model of emotional labor, as the process of regulating both feelings and expressions
in order to meet the organizational goals. In
doing so, she started from Grosss model of
emotion regulation (1998b).
Since in most theories on emotion, the term
is usually associated with a physiological
arousal and cognitive appraisal of the situation,
Gross (1998b) identifies two strategies of emotion regulation: antecedent-focused emotion
regulation and response-focused emotion regulation. The first strategy implies that individuals
can regulate the antecedents of emotion, such as
the situation or the appraisal. But, as in service
jobs there are no options such as choosing or
modifying the situation, Grandey suggests that
changing the focus of personal thoughts and
changing external appraisals of the situation
(attention deployment and cognitive change, in
Grosss terms) are antecedent-focused emotion
regulation types that are similar to Hochschilds
deep acting. In the same manner, responsefocused emotion regulation, understood as modifying expression either by faking or by enhancing nonverbal signs of emotion, corresponds to
Hochschilds surface acting.
Grandey (2000) also emphasizes that the
perception itself of the organizational expectations regarding certain emotional displays will
lead to a higher degree of management of emotions by the employee. And there are three distinct types of emotional work requirements, as
shown by Jones and Best (1995) and Wharton
and Erickson (1993): integrative emotional
work requirements (it is the case of front-line
workers who have to service with a smile);
differentiating emotional work requirements
4/17/2012
1:07 PM
Page 23
(bill collectors or law enforcement officers, for

example, have to display negative emotions);
masking emotional work requirements (controlling emotions such as in the case of judges or
therapists).
From the very beginning, there was a clear
distinction between genuinely felt emotions and
surface acting and deep acting as emotional
labor strategies. Hochschild (2003, p. 36) suggested that feelings do not erupt spontaneously
or automatically in either deep acting or surface
acting. In both cases the actor has learned to
intervene either in creating the inner shape of a
feeling or in shaping the outward appearance of
one. That is why, later on, Rafaeli and Sutton
(1987) characterized the two processes as faking mechanisms. There is though a difference
in terms of faith between deep acting (seen as
faking in good faith, since the intention of the
actor is to seem authentic) and surface acting
(described as faking in bad faith, since what
motivates the actor to conform to display rules
is not helping the customers or the organization,
but rather the personal goal of keeping the job).
Furthermore, Zapf (2002) referred to a third way
of performing emotional labor, which he named
automatic regulation (when the employee automatically displays an organizationally desired
emotion deriving from an emotion that is spontaneously felt). The expression of naturally felt
emotions as a distinct strategy for displaying
emotions at work was also acknowledged by
Diefendorff, Croyle and Gosserand (2005).
Totterdell and Holman argued that not all
emotion regulation at work is done in the service of emotional labor (2003, p. 71), but as
Grandey and Brotheridge posited (2002)
whether one uses job-focused emotional labor
or employee-focused emotional labor, the effort
of regulating emotions is still meeting workplace demands.
Further, theorists were also interested in
determining the direction of emotion regulation.
From this point of view, emotion amplification
and emotion suppression were identified as the
core paths in managing emotional displays in
order to be consistent with organizational
display rules (Hochschild, 2003; Levenson,
1994b). While emotion amplification consists in
initiating or enhancing public displays of emo-
tion, the suppression implies reducing or eliminating the displays of emotion. However, Diefferndorf and Greguras (2009) offered a critical
view of this approach, contextualizing emotional display rules on the basis of the assumption
that these are more complex and nuanced than it
has been considered in prior research. Their
findings supported the hypothesis that most
employees characterize their display rules as
involving something other than complete
expression or suppression, such as qualifying,
masking or amplifying (which differ in terms of
intensity of the emotion expressed).
Regardless of the strategy used or the direction of this strategy it is certain that employees
invest a certain amount of emotional effort in
their jobs trying to meet the requirements of displaying the appropriate emotions. Therefore, the
expression of emotion, once a personal decision,
has become a marketplace commodity with
standards and rules dictating how and when
emotion should be expressed (Morris & Feldman, 1996).
Up to this point one can see that emotional
labor cannot be discussed as a dichotomous
variable (in terms of presence/absence), but it
has to be considered in its multiple dimensions
and with focus on the different strategies of
emotion regulation available, the direction of
the emotional expression or even the specific
emotions that are being regulated. These issues
influence, further on, the study of the consequences of emotional labor, since the abovementioned aspects (taken together or not) may
have distinct impacts on workers (Pugliesi,
1999). Considering the individual variable,
there are people who self-select occupations that
require particular types of emotional labor
(Shuler & Sypher, 2000).
In analyzing the possible outcomes of emotional labor, we shall infer, together with Tolich
(1993) that emotional labor can be both alienating and liberating. Hence, regulating emotions
in work settings in order to meet specific organizational display rules can have either negative
or positive consequences for employees.
The major tendency in research was to determine the extent to which emotional labor is taxing in people work. A comparison between the
emotions generally experienced in the workplace
23
4/17/2012
1:07 PM
Page 24
and those experienced in private life revealed that

people tend to feel more often negative emotions
at work than at home (Rafaeli et al., n.d.). Consistent with these findings, emotional labor was
mainly related to negative attitudinal, psychological and behavioral outcomes for employees.
That is why special attention has been paid to the
relationship between emotional labor and
burnout, as a specific form of job stress.
The association of the two variables was
first suggested by Hochschild (2003). Further
on, Hochschilds view that organizational control of emotional displays is inherently stressful
has received mixed support in both qualitative
and quantitative studies (Morris & Feldman,
1996; Pugliesi, 1999; Erickson & Wharton,
1997; Kruml & Geddes, 2000; Brotheridge &
Grandey, 2002; Grandey, Fisk & Steiner, 2005).
Morris and Feldman (1996) similarly found that
emotional labor and emotional exhaustion correlate positively among debt collectors, military
recruiters and nurses. In another study, Pugliesi
(1999) showed that emotional labor was significantly and positively related to job stress and
increased psychological distress. Therefore, the
main hypothesis of the current study is:
HI. Emotional labor, namely the two emotion regulation mechanisms that it implies (surface acting and deep acting) will positively correlate with core burnout in the case of a sample
of Romanian bank tellers.
Kruml and Geddes (2000) showed that
individuals experience stress when they fake
emotion, rather than genuinely expressing what
they feel, since there appears the so-called emotional dissonance process. In fact, emotional
dissonance was associated from the very beginning with emotional labor, more precisely with
surface acting, being described as an aversive
psychological state which implies experiencing
a sense of discrepancy between the real self
and the socially presented self (Hochschild,
2003, p. 90).
Whether it is seen as a discrete dimension
of emotional labor (Kruml & Geddes, 2000) or
an independent process resulting from the clash
between authentic and required feelings
(Hochschild, 2003; Zapf, 2002), emotional dissonance represents a key factor in exploring the
relationship between emotional labor and
24
burnout. In this respect, Bakker and Heuven

(2006) argued that the equivocal evidence for
the relationship between emotional job demands
and burnout might be attributable to the fact that
emotional dissonance plays a mediating role in
this relationship. Their findings showed that as a
result of the emotionally charged interactions
with patients and civilians, nurses and police
officers experience a discrepancy between felt
and displayed emotions which, in turn, leads to
emotional exhaustion and cynicism.
In field studies emotional dissonance was
correlated with surface acting, but not with deep
acting (Glomb & Tews, 2004; Holman,
Chissick, & Totterdell, 2002). Further on, this
means that emotional dissonance predicts a positive association between surface acting and
strain and, accordingly, no association between
deep acting and strain. Self-reports of emotional dissonance were related to burnout, anxiety,
depression, and job dissatisfaction, as shown by
Abraham (1999), Glomb and Tews (2004), Holman et al. (2002), Kruml and Geddes (2000),
Morris and Feldman (1996), Zapf et al. (1999),
but also job involvement (Kruml & Geddes,
2000) and personal accomplishment (Zapf et al.,
1999).
Morris and Feldman (1996) supported that
emotional dissonance is also determined by positive and negative affectivity. That is, when the
organizationally desirable emotions conflict
with employees affectivity (positive or negative), emotional dissonance may occur. Brotheridge and Lee (1998) argued that affectivity
influences emotional labor through the range
and intensity of emotions displayed, meaning
that individuals with high levels of affectivity
may find it more difficult to surface act or deep
act, compared to low-affect intense individuals.
Later on, Johnson (2004) acknowledged that
positive affectivity moderates the relationship
between surface acting and emotional exhaustion, meaning that individuals with high scores
at positive affectivity tend to experience emotional exhaustion to a greater extent when
engaging more in surface acting.
But Johnsons study examined the influence of affectivity in the case of service jobs that
required only the expression of positive emotions (integrative emotional work requirements,
4/17/2012
1:07 PM
Page 25
in Wharton & Ericksons (1993) terms). In a

series of studies (Diener & Larsen; Higginson;
Larsen & Kettalar as cited in Elfenbein, 2007) it
was reported that individuals high in positive
affectivity are more focused externally on promoting positive outcomes, whereas those high
in negative affectivity are more focused internally on preventing negative outcomes. Hence,
the prevalence of positive, respectively negative
affectivity determines the regulatory strategy
used (promotion or prevention).
Mikolajczak, Menil and Luminet (2007)
introduced the construct of negative consonance
to describe another form of discrepancy that
between felt and expressed emotions and organizational display rules. To put it otherwise,
there are situations when employees choose to
ignore the emotional work requirements and
express their true feelings. Furthermore, the
authors found evidence that negative consonance and burnout are positively associated, the
same trend being specific to the relationship
between negative consonance and somatic complaints (as physiological components of jobrelated stress).
Other researchers used moderating variables such as job resources (de Jonge, le Blanc,
Peeters & Noordam 2008), personal control
(Grandey, Fisk & Steiner, 2005) or gender and
emotional intelligence (Johnson, 2004) in
exploring the relationship between emotional
labor and burnout. So, burnout (in terms of its
core dimensions emotional exhaustion and
depersonalization/cynicism, as acknowledged in
a series of studies: Bakker, 2009; Bussing &
Glaser, 2000; Demerouti & Bakker, 2008; Shirom, 2009; Taris, Le Blanc, Schaufeli &
Schreurs, 2005) was shown to be a response to
emotionally demanding tasks; the results supported a stronger relationship when particular
job resources such as emotional support were
absent (de Jong et al., 2008). In fact, these findings are consistent with Hobfolls theory of conservation of resources (1998, 2002).
According to this approach, people have a
basic motivation to obtain, retain and protect
that which they value. Things that people value
were generically called resources and they were
described as material, social and energetic
resources. Hobfoll (1989, 1998) claimed that
stress at work is due to threats with resource

loss, loss of resources or failure to regain
resources after they were invested. From this
point of view, energetic resources are specifically taken into account, such as emotional robustness, cognitive ability and physical vigor (Hobfoll & Shirom, 2001; Shirom, 2003). Moving
further, burnout appears as the end state of a
long-term process of resource loss that gradually develops over time depleting energetic
resources (Hobfoll & Freddy, 1993, p. 117).
When the employee experiences a large discrepancy between felt emotions and demanded
displays, the possibility for spontaneously
expressing what one feels reduces and the
necessity to act (whether deep or surface)
increases consequently. Hence, emotion regulation will deplete employees emotional
resources, which, in turn will result in more
burnout complaints (physical fatigue, emotional
exhaustion, cognitive weariness) as supported
by Bakker and Heuven (2006). Therefore, individuals who lack resources are the most vulnerable to further losses, whereas those with greater
resources are less vulnerable to resource loss
and more capable of resource gain (Gorgievsky
& Hobfoll, n.d.). Forth, this variation can be
partly explained by the fact that people generally experience differently loss and gain, as they
are more concerned about avoiding loss than
achieving more resources.
Up to now we can conclude that in most of
the studies the findings supported the claim
made by Richards and Gross (2000) that from
the two core emotional labor strategies, surface
acting seems to be more taxing, since it implies
both the effortful suppression of emotions and
the production of appropriate, organizationally
desirable ones and this is the reason why the
first secondary hypothesis of the present study
entails that:
HI.I. Surface acting will positively correlate with emotional exhaustion in the case of a
sample of Romanian bank tellers.
In the same manner, the next secondary
hypothesis seeks to explore the relationship
between surface acting and depersonalization, a
positive correlation being expected between the
two subscales, as well.
25
4/17/2012
1:07 PM
Page 26
HI.II. Surface acting will be positively

related to depersonalization in the case of a
sample of Romanian bank tellers.
Of course that, as Ashkanasy, Hartel and
Daus (2002) point out, the findings of the studies previously mentioned should be taken into
account considering the individual characteristics of employees, where some may be better
equipped or skilled when it comes to performing
emotional labor effectively and without adverse
personal consequences. And this observation
leads to the question whether managing the
heart in the workplace must be necessarily
stressful or not.
Although the approach of positive emotional labor outcomes was underexplored, there
were some studies that acknowledged emotional labor as a social dimension which connects
employees and makes their work more enjoyable (Schuler & Sypher, 2000). In the same
manner, in examining the relationship between
emotion regulation and work strain, Ct (2005)
formulated an interpersonal model of emotion
regulation, suggesting that the extent to which
managing emotions in the workplace can result
in strain depends upon the receivers response to
the senders emotion regulation and display, the
form of emotion regulation and the emotion
being regulated. Even Hochschild (2003) admitted that deep acting, as an emotional labor strategy, may have potential benefits for employee
outcomes, but warned of the commoditization
of employees feelings by the organizations.
There are also differences in the affective experience concerning the reaction to workplace
events: individuals high in positive affect are
more reactive to positive workplace events and
less reactive to negative events (Miner, Glomb
& Hulin, 2005).
Taken into account those inferred above,
mainly the fact that emotional labor can also
lead to positive outcomes, especially when
employees are acting in good faith, a positive
relationship is expected between deep acting
and personal accomplishment (the self-evaluation component of burnout):
HI.III. Deep acting will show a positive
association with the dimension of positive
accomplishment from the burnout scale.
26
As the main interest in the current study

was to explore the relationship between emotional labor and burnout in direct sales representatives, the conceptualization of the later term
urges. Job burnout was seen as a psychological
syndrome in response to chronic interpersonal
stressors on the job (Maslach, Schaufeli &
Leiter, 2001, p. 399). The research on this syndrome is rooted in caring and service occupations, since these kinds of jobs are based on the
relationship between provider and recipient
(Maslach et al., 2001). Therefore, due to this
interpersonal context of the job, burnout was
studied in terms of an individuals transactions
in the workplace rather than as an individual
stress response. A specific service occupation is
that of bank tellers. They are considered the
front line in the banking business, as they promote the financial products of the institution
they work for. Thus they deal directly with customers and have to service with a smile,
regardless of their true feelings, which may lead
them to experiencing burnout.
But how the experienced burnout can be
measured? It was described as a tri-dimensional
response, consisting in emotional exhaustion
(the individual dimension), depersonalization or
cynicism (the interpersonal dimension) and
reduced efficacy or accomplishment (the selfevaluation dimension) (Maslach & Jackson,
1981). In the initial research and scale development, involvement was proposed as a fourth factor, but subsequently it became optional.
Theoretically it was inferred that emotional
exhaustion and cynicism represented the core
dimensions of burnout (Green, Walkey & Taylor, 1991; Schaufeli, 2003; Schaufeli & Enzman, 1998; Schaufeli & Taris, 2005), and this
assumption is supported by a series of empirical
findings (Lee & Ashforth, 1996; Toppinen-Tanner, Kalimo, & Mutanen, 2002; van Dierendonck, Schaufeli, & Buunk, 2001). Despite this,
Boersma and Lindblom pointed out that very
little is actually known about how and when
these variables interact over time in the process
towards burnout (2009, p. 266). Therefore, two
studies examined the hypothesis that there are
different pathways to burnout for different individuals.
4/17/2012
1:07 PM
Page 27
Demerouti, Verbeke and Bakker (2005)

identified five possible burnout configurations,
based on the necessity to simultaneously take
into account the three basic symptoms of
burnout syndrome, using cluster analysis: a nonburnout cluster, a burnout cluster and three clusters with what they called, incomplete
burnout, characterized respectively by exhaustion only, cynicism only and reduced efficacy
only.
Maslach and Leiter (2008) continued the
work of Demerouti et al. (2005) by exploring
the trajectories of change over time for individuals with different scoring patterns. They used
the two core dimensions of burnout (emotional
exhaustion and cynicism) and created two subgroups with congruent patterns (burnout: high
exhaustion, high cynicism and engaged: low
exhaustion, low cynicism), respectively incongruent patterns (exhaustion only: high exhaustion, low cynicism and cynicism only: low
exhaustion, high cynicism).
The factors leading to burnout were widely
categorized as situational and individual. Job
characteristics both quantitative (such as
workload or time pressure) and qualitative (such
as role conflict or role ambiguity) and organizational characteristics (such as organizational
values) represent typical situational factors that
affect employees well-being. On the other
hand, demographic aspects (age, gender, and
education), personality traits or job attitudes are
individual factors that relate to burnout
(Maslach et al., 2001).
Since the current study was conducted in
the Romanian service work context, it is important to note that comparative studies emphasize
a difference in terms of intensity of experienced
burnout between European and North American
emotional laborers (Maslach et al., 2001).
Important findings regarding the extent to
which emotional labor can set employees in
people work jobs on fire and make them burn
out were reported by Brotheridge and Grandey
(2002). Focusing on the quality of experiences
in service encounters, the authors established
that surface acting or faking emotional expressions at work was related to feeling exhausted
and detached, whereas deeper emotion work
was related positively to personal accomplishment.

When it comes to the intensity of the
provider-recipient interactions, it seems that
employees who had less emotionally charged
interactions with clients reported less emotional
exhaustion than did those whose interactions
were more intense (Maslach, 1978a). In another
study, the variables of duration, frequency and
intensity of interactions were not confirmed as
predictors for employee burnout (Cordes,
Dougherty & Blum, 1997).
Job performance emerges as a constant
variable in the study of both emotional labor and
burnout. As mentioned from the beginning of
the current study, job performance was identified as one of the most common outcomes of
emotional labor (Grandey, 2000, 2003; Brotheridge & Grandey, 2002; Pugliesi, 1999).
Although for the individual emotional labor can
be taxing, as seen in the previous section, in
terms of customer service and task effectiveness
emotional labor was proven to have a positive
impact (Ashforth & Humphrey, 1993;
Hochschild, 2003). However, as there are different emotional labor strategies, some may be
more effective when it comes to improving job
performance: for instance, when employees
engage in deep acting they are more likely to be
perceived by customers as authentic, than when
they surface act, since they act in bad faith
(Grandey, 2000).
On the other hand, Wright and Bonett
(1997) supported in a longitudinal study the
negative relationship between emotional
exhaustion and job performance. This relationship was proven to be the opposite when job
performance was determined through supervisory ratings (Wright & Cropanzano, 1998). In fact,
Keijsers, Schaufeli, Le Blanc, Zwerts and
Miranda (1995) had already acknowledged that
burnout has different effects on performance
depending on how one operationalizes the later,
the tendency being that self-reports of job performance are negatively associated with
burnout, while objective assessments of job performance positively relate to burnout. Irrespective of the directionality of the associations,
emotional exhaustion was constantly identified
as the only burnout dimension that correlated
27
4/17/2012
1:07 PM
Page 28
with job performance (Wright & Bonett, 1997;

Wright & Cropanzano, 1998; Parker & Kulik,
1995).
As for the possible relationships between
emotional labor, burnout and performance,
Bakker, Demerouti and Verbeke (2004) found
that the relationship between job demands
(including emotional demands) and in-role performance is mediated by feelings of exhaustion.
This relation was further supported by Bakker
and Heuven (2006), who showed that emotionally demanding interactions with recipients
leads to emotional dissonance, which, in turn,
leads to job burnout and impaired performance.
Equally, displaying positive emotions at work in
emotional demanding interpersonal interactions
was proven to be beneficial for the organization,
since the frequency of emotional labor was positively associated with client satisfaction and
return of business (Tsai, 2001).
Based on these associations, the last secondary hypothesis of the present research is:
HI.IV. Performance will negatively correlate with the experienced burnout and positively
correlate with emotional labor in the case of the
sample investigated.
Although there is an increasing interest in
emotional labor and burnout, as they emerged as
core concepts in recent organizational behavior
studies, these concepts are still in their pioneer
state in the Romanian context. That is why the
objective of the present research was to determine whether there is a relationship between
emotional labor, more precisely between the
two emotion regulation strategies (surface acting and deep acting) and burnout in the case of
Romanian direct sales representatives (bank
tellers), as they represent prototypical emotional laborers. Furthermore, we aimed to identify
those emotional labor subscales which can best
predict burnout, as well as the effects that regulating emotions in order to be consistent with
emotional work requirements and/or experiencing burnout can have on the job performance of
bank tellers from our sample.
28
Method
Participants
The instruments used for assessing the two
constructs were administered on a sample consisting of 107 bank tellers, of which 87 were
females and 17 were males. The questionnaires
were administered to employees (N = 107) of a
financial institution from Bucharest. They were
all with a bachelors and/or masters degree and
they had between 2 and 12 years of service in
the banking industry (M = 6.8 years, SD = 1.56);
ages were between 23 and 47 years (M = 32.15,
SD = 5.32).
Procedure
The instruments were administered online,
via e-mail to all employees from six agencies of
a financial institution from Bucharest. Among
these bank tellers, 107 of them responded to the
questionnaires, generating a 75% response rate.
The collection of data lasted three weeks. In
order to ensure anonymity of the respondents,
they were not asked to provide their names,
although data regarding the samples demographic features were collected separately.
Measures
Emotional Labor. The Brotheridge & Lee
Emotional Labor Scale (Brotheridge & Lee,
1998) consists of 14 items, scoring in six subscales, measured on a five-point frequency scale
(1 = never, 5 = always). The six subscales,
which determine distinct dimensions of emotional labor, are: the duration of customer interaction (a single free response question, yielding
a number); the frequency subscale (three items;
e.g. on an average day at work, how frequently
do you interact with customers) addresses the
frequency of the display of organizationally prescribed emotions; the intensity subscale (two
items; e.g. on an average day at work, how frequently do you express intense emotions)
describes how often employees express strong
or intense emotions at work; the variety of emotional displays at work (three items; e.g. on an
average day at work, how frequently do you display many different kinds of emotions); the
deep acting subscale (three items; e.g. on an
average day at work, how frequently do you try
4/17/2012
1:07 PM
Page 29
burnout (Green et al., 1991; Lee & Ashforth,

1996; Schaufeli, 2003; Schaufeli & Enzman,
1998; Schaufeli & Taris, 2005; Toppinen-Tanner, Kalimo, & Mutanen, 2002; van Dierendonck, Schaufeli, & Buunk, 2001). The fourth subscale, Involvement with people, did not obtain
high scores in the factorial analysis. However,
as it moderately correlated with Emotional
exhaustion, this scale was retained as an optional factor of the MBI. The authors (Maslach &
Jackson, 1981) report good reliability for the
MBI subscales: alpha=.89 (frequency) and
alpha=.86 (intensity) for Emotional Exhaustion, alpha=.74 (frequency) and alpha=.74
(intensity) for Personal Accomplishment,
alpha=.77 (frequency) and alpha=.72 (intensity) for Depersonalization and alpha=.59 (frequency) and alpha=.57 (intensity) for Involvement.
Performance. The performance scores used
in the current study were provided by the
human resources department according to
the annual performance evaluation sheets, consisting in performance evaluation scales with
behavioral anchors designed according to internal criteria and procedures.
to actually experience the emotions that you

must show) determines how much an employee has to modify his inner feelings in order to be
consistent with the organizational display rules
and the surface acting subscale (three items;
e.g. on an average day at work, how frequently
do you pretend you have emotions that you
dont really have) measures the extent to which
one has to express emotions that are not really
felt and therefore modifies only the outward
appearance of the emotion. The authors (Brotheridge & Lee, 2002) report good combined
alpha coefficients for both role characteristics
(frequency, intensity and variety) subscales
(alpha = .71), and deep acting and surface acting subscales (alpha = .89, alpha =.86).
Burnout. The Maslach Burnout Inventory
(Maslach & Jackson, 1981) is grouped in four
subscales which are measured on a seven-point
Likert scale in terms of intensity (from 1 = very
mild, barely noticeable, to 7 = very strong,
major) and frequency (from 1 = never, to 7 =
every day). The subscales are: Emotional
Exhaustion subscale (nine items; e.g. I feel
emotionally drained from my work) which
describe feelings of being emotionally overextended and exhausted by the work; Depersonalization subscale (five items; e.g. I dont really
care what happens to my recipients) refers to
unfeeling and impersonal response towards the
recipients of ones care or service; Personal
Accomplishment subscale (eight items; e.g. I
have accomplished many worthwhile things in
this job) taps feelings of competence and successful achievement in people work. While
high scores on the first two subscales indicate
high degrees of experienced burnout, the latest
is negatively related to experienced burnout.
That is why, subsequently a series of studies
have acknowledged Emotional Exhaustion and
Depersonalization as core components of
Results
Descriptive Statistics
Table 1 presents descriptive statistics for
the variables included in the study. Means and
standard deviations for the core burnout and the
two emotion regulation strategies are important
indicators, since, as advanced by Wharton
(1993), it is not emotional labor itself that
results in burnout, but how it is performed.
Table 1. Mean and standard deviation for core burnout, and emotional labor strategies
M
SD
Core burnout
3.37
.48
Deep acting
2.75
.70
Surface acting
2.94
.68
29
4/17/2012
1:07 PM
Page 30
Table 2. Mean and standard deviation for the ELS

Emotional labor scales
SD
Frequency
3.98
.81
Intensity
2.82
.78
Variety
3.00
.78
Deep acting
2.75
.70
Surface acting
2.94
.68
Table 3. Mean and standard deviation for the MBI scales

Burnout scales
SD
Emotional exhaustion frequency
4.03
.61
Emotional exhaustion intensity
3.74
.55
Personal accomplishment frequency
5.09
.49
Personal accomplishment intensity
4.71
.43
Depersonalization frequency
2.72
.50
Depersonalization intensity
2.86
.49
Involvement frequency
3.04
.70
Involvement intensity
2.95
.73
From the two emotional labor strategies

surface acting is more frequently experienced
by bank tellers from our study (M=2.94,
SD=.68).
Further on, taking a closer look to all the
subscales of the two processes it can be seen
that, in terms of means, the values are not homogeneous (Table 2).
Thus, the items that compose the frequency
subscale in the ELS (how often do you adopt certain emotions as part of your job) were reported
to have the highest mean value (M=3.98,
SD=.81) in the work settings taken into account,
compared to the items in the remaining composite scales two expressing rather role characteristics and the other two emotional labor mechanisms (intensity and variety, respectively, deep
acting and surface acting).
The same tendency can be noticed in the
means distribution for the other investigated
process (Table 3). Personal accomplishment
30
(PA) seems to be the most frequently and

intensely experienced burnout subscale
(Mf=5.09, SDf=.49; Mi=4.71, SDi=.43), while
the lowest means in terms of both frequency and
intensity were reported for Depersonalization
(DP) (Mf=2.72, SDf=.50; Mi=2.86, SDi=.49).
Tests of the Hypotheses
In order to meet the objectives assumed for
the current study and test the hypotheses formulated in the previous section the correlations
between emotional labor and burnout scales and
subscales, as well as the job performance scores
were further determined. The findings offer relevant insights. As it can be noticed from the correlations matrix (Table 4), emotional labor (in
terms of its two emotion regulation strategies:
surface acting and deep acting) and burnout
(core burnout, calculated by determining the
mean between emotional exhaustion and depersonalization in terms of frequency) significantly
4/17/2012
1:07 PM
Page 31
Table 4. Bivariate correlations of emotional labor subscales and core burnout

core burnout
Emotional Labor Scale frequency
.38**
Emotional Labor Scale intensity
.41**
Emotional Labor Scale variety
.36**
Emotional Labor Scale deep acting
.38**
Emotional Labor Scale surface acting
.28**
Note: * p < .05, ** p < .01.
correlate in the case of the sample of Romanian

bank tellers examined and thus, the main hypothesis of the current study (HI.) is confirmed.
As was mentioned earlier, bank tellers represent a specific emotional labor as well as
burnout occupation. Working in a bank implies
having face-to-face and voice-to-voice interactions with the customer, meaning that bank
tellers engage in job-relevant interactions which
demand them to feel, or at least project the
appearance of positive emotions (thus to service with a smile). Furthermore, their emotional expressions are somewhat controlled by the
employer, for instance through training, policies
or supervision. And the notion of prescribed and
supervised emotions is what distinguishes emotional laborers from other employees, as noted
by Hochschild (2003).
Emotions appear to be important facets of
the products the service industry sells to clients
(Bakker & Heuven, 2006, p. 425) that is smiling
bank tellers emerge as assets of great importance for the publics perception of the profession and the client satisfaction. As emotional
laborers, bank tellers invest a certain amount of
emotional effort in their jobs in order to meet the
organizationally display rules. This emotional
effort requires both the physical and mental
management of ones emotions and it may result
in strain.
From the two emotional labor strategies
(deep acting, respectively surface acting) the
attempt to modify the inner shape of the feelings
in order to be consistent with the organizational
display rules seems to be more taxing for bank
tellers from our sample (r=.38, p<.01) than having to shape only the outward appearance of
feelings (r=.28, p<.01). From this point of view,

a series of previous studies support Richards
and Grosss claim (2000) that from the two core
emotional labor strategies, surface acting seems
to have a greater impact, since it implies both
the effortful suppression of emotions and the
production of appropriate, organizationally
desirable ones. In the case of the sample analyzed, the findings do not support this claim,
although, as shown in the descriptive statistics
section respondents report using the two emotional labor strategies with approximately the
same frequency in service encounters (deep acting M=2.75, SD=.70, surface acting
M=2.94, SD=.68).
A possible explanation of the fact that trying to actually feel the emotions required leads
bank tellers from our sample to experience
burnout to a greater extent than having to fake
the outward appearance of feelings could be the
relationship between deep acting and the intention to seem authentic. In other words, when
bank tellers try to reduce the clash between their
spontaneous feelings and those required, as part
of their work-role, they choose to reduce this
dissonance by spontaneously feeling and
expressing what they are required to express in
the given interpersonal service transactions as
they want to seem authentic. Therefore, trying
to be authentic in terms of emotionality seems to
have a more negative impact on the employees
wellbeing, especially when the service encounters are intensely charged. In fact, in an
exploratory analysis we found that, from the
five emotional labor subscales assessed by the
ELS, the intensity of emotions being regulated
showed the most significant positive relation31
4/17/2012
1:07 PM
Page 32
Table 5. Bivariate correlations between the subscales of emotional labor and the subscales of burnout
EEf
PAf
DPf
INf
EEi
PAi
DPi
INi
ELS_DA
.30**
.38**
.24*
.46**
.43**
.33**
.31**
.32**
ELS_SA
.16
.31**
.26**
.38**
.37**
.13
.12
.17
ELS_intensity
.28**
.49**
.32**
.43**
.55**
.28**
.37**
.23*
ELS_variety
.24**
.34**
.28**
.44**
.46**
.34**
.31**
.50**
Note: * p < .05, ** p < .01.

EEf = emotional exhaustion frequency;
PAf = personal accomplishment frequency;
DPf = depersonalization frequency;
INf = involvement frequency;
EEi = emotional exhaustion intensity;
PAi = personal accomplishment intensity;
DPi = depersonalization intensity;
INi = involvement intensity.
ship with core burnout (r=.41, p<.01). These

findings are consistent with those of Maslach
(1978), who supported that employees who had
less emotionally charged interactions with
clients reported less emotional exhaustion than
did those whose interactions were more intense.
Moving further, these relationships could be
better explored by taking into account in future
research dispositional factors, such as emotional
intelligence or dispositional affect (positive
affectivity, respectively negative affectivity).
In order to better understand the findings
presented up to this point and as well as to test
the secondary hypotheses formulated, the
bivariate correlations between the subscales of
both emotional labor and burnout were calculated. The results are reported in Table 5.
Concerning the first secondary hypothesis
(HI.I.), which implied that surface acting positively correlates with emotional exhaustion, it
was also acknowledged by the results reported
in Table 5 (r = .37, p < .01), the findings being
supported by previous research (Brotheridge &
Grandey, 2002). This means that, the more bank
tellers try to show positive emotions at work, the
more overextended and exhausted they feel.
And that is because painting on a smile when
not really feeling like smiling can be quite emotionally taxing, in terms of intensity (since surface acting is positively related only to the emotional exhaustion intensity subscale).
32
As for the next secondary hypothesis (HI.II),

regarding the relationship between surface acting
and depersonalization, it was also validated by
the findings, which show that the two subscales
are positively related, but only in terms of frequency (r=.26, p > .01). Thus, the more a bank
teller tries to change the outwardly emotional display, the more he or she will treat customers like
objects, showing a detached attitude.
However, consistent with the results presented above, regarding the relationship between
core burnout and the two core emotional labor
strategies, deep acting significantly correlates
with all the subscales of the burnout syndrome.
Although trying to actually feel what one should
feel as part of the job role was previously proven
to diminish the tendency of unfeeling (detaching oneself not only from his/her own feelings,
but also from others feelings) (Maslach & Jackson, 1981), the current findings suggest the
opposite relationship (rf=.24, p<.05, ri=.31,
p<.01). Grandey claimed that it is more likely
that payoffs of deep acting, such as reduced
emotional dissonance or positive feedback from
customers may restore the employees emotional
resources (2003, p. 93), but a possible explanation for the reverse effect that resulted in our
analysis could be the fact that energetic
resources that are used to a greater extent in deep
acting cannot be easily restored. That is why a
closer look to the individuals capacity to over-
4/17/2012
1:07 PM
Page 33
come the loss of resources, as well as to engage

in regaining them could further tone this association, since people experience differently gain
and loss (as a general tendency, people are usually more focused on avoiding loss, than on
achieving more resources).
As expected according to the third secondary
hypothesis (HI.III.), the dimension of personal
accomplishment is also positively related with
deep acting (rf=.38, p>.01, ri=.33, p<.01) and
this relationship has already been demonstrated
by Brotheridge and Grandey (2002).
When it comes to determining the impact of
experiencing burnout on performance, which
emerged in field studies as an important organizational outcome, no associations were found.
Instead, regarding the positive association
between emotional labor and performance as
entailed by the last secondary hypothesis
(HI.IV.), the intensity of the emotions required
in service interactions in order to be consistent
with emotional work requirements determines a
decrease in the level of performance of bank
tellers from the sample analyzed (r=-.27,
p<.01).
An exploratory analysis was also undergone,
which revealed other interesting findings (Table
5). For instance, the intensity subscale from the
ELS seems to have an important role when it
comes to strongly feeling worn out, depleted,
fatigued (r=.55, p<.01), whereas the intensity of
emotions being regulated positively correlates
with the depersonalization subscale (r=.37,
p<.01) in fact, from the five subscales of the
ELS, the intensity subscale shows the most significant relationship with experiencing negative
or inappropriate attitudes towards clients.
On the other hand, having to display many
different kinds of emotions (item 6) and doing
it frequently, also influences the extent to which
bank tellers from our sample experience each of
the four subscales used in assessing burnout, as
shown in Table 5.
The results presented up to now are considerable evidence that emotional labor can be
stressful in the case of bank tellers, since
expressing socially appropriate emotions as a
form of role demand undermines employees
wellbeing. As it can be noticed, emotional labor
and burnout significantly correlate through
almost all of the respective subscales, the same

tendency not being identified in the case of performance in terms of its relationship with emotional labor and burnout (performance is significantly related only with the intensity of
emotions being regulated in service interactions). Thus, in order to meet the objective of the
present research the next step of the analysis
consisted in identifying the exact set of variables that can best predict burnout in the case of
the sample under consideration thorough the
multiple regression analysis.
As de Vaus (2002) pointed out, in doing
multiple regression analysis the general principle of explanation is to seek the simplest powerful model (parsimony), and to avoid including
variables that add virtually nothing to the predictive value of the model. Therefore, one must
limit the number of variables included in the
analysis, since increasing the number of variables artificially inflates the R2 especially in
small samples. Another important aspect in
deciding how many variables to include in the
model is the size of the sample. The variable to
case ratio helps identify the sample size required
to accommodate the number of variables one
wishes to include. When all variables are
entered into the model in a single block the ratio
of cases to variable should be at least 20:1, and
the minimum suggested sample to variable ratio
is 5:1 (de Vaus, 2002).
Taking into consideration the above mentioned conditions, it was decided to include in
the analysis only the variables which correlated
significantly with the core burnout and the
respective subscales. The analysis was undergone using the hierarchical strategy, that is starting with the variable with the highest correlations (ELS intensity), then adding one by one
the all the other subscales of the ELS. Table 6
presents only the models that accounted the
most in predicting burnout.
The first model shows that the intensity of
emotions that bank tellers are required to display at work accounts for 17% of the variance
for the level of burnout that they are experiencing (R2=.171). The model is significant with
F(1,105)=21.59 (p<.01). When adding the second predictor the variety of emotions being
regulated in Model 2, R2 slightly increased
33
4/17/2012
1:07 PM
Page 34
Table 6. Summary of hierarchical regression analysis for variables predicting burnout (n = 107)
B
Sig.
.216
.413
4.64
.000
ELS_intensity
.165
.316
3.24
.002
ELS_variety
.114
.219
2.25
.026
Variable
R2
Adjusted R2
Sig.
Model 1
.171
.163
21.59
.000
ELS_intensity
Model 2
.209
.194
(R2=.209) and not just as an effect of the integration of another variable, since the adjusted R2
increased as well (from .163 to .194). This second model is also significant with
F(1,104)=5.07 (p<.05), which means that having to display many different kinds of emotions
during emotionally charged interactions in work
settings leads bank tellers from the sample
under consideration to experience burnout to a
greater extent.
In sum, the attempt to determine the set of
variables which could best predict burnout in
the case of the sample analyzed was valuable,
the results acknowledging the intensity of emotions that bank tellers are required to display as
the key element in assessing the negative outcomes of serving with a smile, as it emerged
from the very first level of the analysis. In addition, bank tellers from the sample included in
the current study are more likely to feel worn
out at work (not only physically, but mostly
emotionally) if they have to display different
intense emotions (as shown by model 2).
Furthermore, it was considered appropriate
to include background variables (age and gender) in the models that validated the best predictors for burnout for the participants in the present research, in order to obtain new insights
regarding the causal relationships revealed by
the regression analysis. This final step entails,
since women were proven to be more interpersonally sensitive than men. In other words,
women seem to be more accurate in perceiving
the personal, interpersonal and social environment (Knapp & Hall, 1997), ability that
34
5.07
.026
emerged in the literature both as a trait and as a

more specific skill, which refers to judging nonverbal cues (Briton & Hall, 1995; Spence,
Helmreich & Stapp, 1975 as cited in Hall &
Schmid, 2008). In the same manner, studies
have identified women as the more emotional
sex (Grossman & Wood, 1993), meaning that
they report more intense experience and expression of emotions, as well as a greater emotional
responsiveness than men and these differences
are not just due to normative expectations that
result from the social role interpretation, but are
also proven by physiology.
On the other hand, age differences in emotional experience, expression and control were
also investigated. For instance, Gross et al.
(1997) found across four studies a consistent
pattern of age differences: control of emotions
increases with age, while emotional expressivity decreases and it seems that aging is positively related to experiencing to a greater extent
positive emotions.
Therefore, taking into account the role of
age and gender in affective processes an analysis was undergone that sought to investigate
whether these two categorical variables moderate the relationship between burnout, as a
dependent variable and ELS-intensity, respectively ELS-variety as independent variables.
The results of this investigative approach did
not confirm the fact that gender or age further
tone the relationship between the dependent and
independent variables identified by the regression analysis.
4/17/2012
1:07 PM
Page 35
Discussion and conclusions

The current study has implications for the
research on the emerging role of emotions in the
workplace. Not only do the findings acknowledge the existence of emotion regulation strategies in the Romanian work settings, but they also
explore the impact of emotional labor on typical
service agents in terms of outcomes such as
burnout and performance. Of course, that the
present results should be discussed in terms of
strengths and limitations of the study. As a first
limitation, we note that the present research is an
explorative pilot study. Hence, the results cannot
be seen as generally valid in the Romanian
workplace, as they express the relationship
between the two organizational aspects in a specific form of people work that of bank tellers
and on a distinct sample (employees of a financial institution from Bucharest). Another limitation is that the sample used in the conducted survey was quite reduced (N=107) and it was quite
imbalanced regarding the gender of the participants (mainly females). Furthermore, though the
present results show that emotional laborers
experience negative affective outcomes future
research should examine the extent to which
emotional job demands can lead to positive outcomes. And that is because the results regarding
the subscales of emotional labor that correlate
with the personal accomplishment dimension
from the MBI support this antecedent.
This research also provides preliminary
support for using emotional labor as an independent variable in explaining other aspects of
organizational life, such as job satisfaction,
counterproductive behaviors at work, engagement or health. Future work should replicate
these findings in the case of organizational roles
beyond service roles, since as already stated by
Ashforth & Humphrey (1993, p. 109) it is difficult to imagine an organizational role to which
display rules would not apply at various points.
Likewise, taking a closer look to dispositional factors (emotional intelligence, positive
or negative affectivity, interpersonal sensitivity,
knowledge of nonverbal cues) and their role in
performing emotional labor could further tone
the associations of this variable with other
important constructs in the study of organizational behavior.

As for the practical implications of the current findings, it is important for organizations to
acknowledge the role of emotional labor in the
workplace, since understanding the social relations in work settings allows service based companies to develop and implement policies that
meet the needs of both internal and external customers. That is because emotional labor was
proven a key determinant of quality of service
and of organizational performance (Hsieh and
Guy, 2009). On the other hand, special attention
should be paid to the psychological effects that
portraying emotions has on the actor.
To conclude, there are still many questions
to be answered regarding the story behind service with a smile (Johnson, 2004), but the current study represents a milestone in exploring
the phenomenon in the Romanian context.
References
Abraham, R. (1999). The impact of emotional dissonance on organizational commitment and intention
to turnover. Journal of Psychology, 133, 441-455.
Allen, J., Pugh, D., Grandey, A. & Groth, M. (2010)
Following Display Rules in Good or Bad Faith?
Customer Orientation as a Moderator of the Display RuleEmotional Labor Relationship. Human
Performance, 23(2), 101-115.
Ashforth, B. & Humphrey, R. (1993). Emotional Labor
in Service Roles: The Influence of Identity. Academy of Management Review, 18(1), 88-115.
Ashkanasy, N.M., Hartel, C. & Daus, C. (2002). Diversity and emotion: the new frontiers in organizational behavior research. Journal of Management,
28(3), 307-338.
Bakker, A.B. (2009). The crossover of burnout and its
relation to partner health. Stress and Health, 25,
343-353.
Bakker, A.B., Demerouti, E., & Verbeke, W. (2004).
Using the Job DemandsResources model to predict burnout and performance. Human Resource
Management, 43, 83-104.
Bakker, A.B., Heuven, E. (2006). Emotional dissonance, burnout and in-role performance among
nurses and police officers. International Journal of
Stress Management, 13, 423-440.
Boersma, K. & Lindblom, K. (2009). Stability and
change in burnout profiles over time: A prospec-
35
4/17/2012
1:07 PM
Page 36
tive study in the working population. Work &

Stress, 23, 264-283.
Brotheridge, C. & Grandey, A. (2002). Emotional Labor
and Burnout: Comparing Two Perspectives of
People Work. Journal of Vocational Behavior,
60, 17-39.
Brotheridge, C.M. & Lee, R.T. (2003). Development
and validation of the Emotional Labor Scale. Journal of Occupational and Organizational Psychology, 76(3), 365-379.
Brotheridge, C. M., & Lee, R. T. (1998). On the dimensionality of emotional labour: Development and
validation of the Emotional Labour Scale. Paper
presented at the First Conference on Emotions in
Organizational Life, San Diego.
Bussing, A. & Glaser, J. (2000). Four-stage process
model of the core factors of burnout: The role of
work stressors and work-related resources. Work
and Stress, 14, 329-346.
Ct, S. (2005). A social interaction model of the effects
of emotion regulation on work strain. Academy of
Management Review, 30, 509-530.
Cordes, C.L., Dougherty, T.W. & Blum, M. (1997). Patterns of burnout among managers and professionals: A comparison of models. Journal of Organizational Behavior,18, 685-701.
De Jonge, J., Le Blanc, P., Peeters, M. & Noordam, H.
(2008). Emotional job demands and the role of
matching job resources: A cross/sectional survey
study among health care workers. International
Journal of Nursing Studies, 45, 1460-1469.
De Vaus, D. (2002). Analyzing Social Science Data: 50
Key Problems in Data Analysis. London: Sage
Publications.
Demerouti, E., & Bakker, A.B. (2008). The Oldenburg
Burnout Inventory: A good alternative to measure
burnout and engagement. In J. Halbesleben (Ed.),
Stress and burnout in health care. Nova Sciences.
Demerouti, E., Verbeke, W. & Bakker, A. (2005).
Exploring the relationship between a multidimensional and multifaceted burnout concept and selfrate performance. Journal of Management, 31(2),
186-209.
Diefendorff, J.M., Croyle, M. & Gosserand, R. (2005).
The dimensionality and antecedents of emotional
labor strategies. Journal of Vocational Behavior,
66, 339-357.
Diefendorff, J.M. & Greguras, G.J. (2009). Contextualizing Emotional Display Rules: Examining the
roles of Targets and Discrete Emotions in Shaping
Display Rules. Journal of Management, 35,
880-898.
Elfenbein, H.A. (2007). Emotions in organizations. A
review in stages. Working Paper Series, Institute
for Research on Labor and Employment, Berkley.
Erickson, R.J., Wharton, A.S. (1997). Inauthenticity and
depression: Assessing the consequences of interac-
36
tive service work. Work and Occupations, 24,

188-213.
Erickson, R.J. & Wharton, A.S. (1993). Managing emotions on the job and at home: U n d e r s t a n d i n g
the consequences of multiple emotional roles.
Academy of Management Review, 18(3), 457-486.
Glomb, T.M., & Tews, M.J. (2004). Emotional labor: A
conceptualization and scale development. Journal
of Vocational Psychology, 64, 1-23.
Gorgievsky, M. J. & Hobfoll, S. E. (n.d.). Work can
burn us out or fire us up: Conservation of
resources in burnout and engagement.
Grandey, A. (2000). Emotion Regulation in the Workplace: A New Way to Conceptualize Emotional
Labor. Journal of Occupational Health Psychology, 5(1), 95-110.
Grandey, A. (2003). When The Show Must Go On:
Surface Acting and Deep Acting as Determinants
of Emotional Exhaustion and Peer-Rated Service
Delivery. Academy of Management Journal,
46(1), 86-96.
Grandey, A., Fisk, G. & Steiner, D. (2005). Must Service with a Smile Be Stressful? The Moderating
Role of Personal control for U.S. and French
Employees. Journal of Applied Psychology, 90(5),
893-904.
Green, D. E., Walkey, F. H. & Taylor, A. J. W (1991).
The three-factor structure of the Maslach Burnout
Inventory. Journal of Social Behavior and Personality, 6, 453472.
Gross, J. (1998b). The emerging field of emotion regulation: An integrative review. Review of General
Psychology, 2, 271-299.
Gross, J., Carstensen, L., Hsu, A., Pasupathi, M., Skorpen, C.G. & Tsai, J. (1997). Emotion and aging:
experience, expression and control. Psychology
and Aging, 12(4), 590-599.
Grossman, M. & Wood, W. (1993). Sex differences in
intensity of emotional experience: A social role
interpretation. Journal of Personality and Social
Psychology, 65(5), 1010-1022.
Hall, J.A. & Schmid Mast, M. (2008). Are women
always more interpersonally sensitive than men?
Impact of goals and contact domain. Personality
and Social Psychology Bulletin, 34, 144-155.
Hobfoll, S. E. (1989). Conservation of resources: A new
attempt at conceptualizing stress. American Psychologist, 44, 513-524.
Hobfoll, S.E. (1998). Stress, Culture and Community.
The psychology and phylosophy of stress. New
York: Plenum.
Hobfoll, S.E. (2002). Social and psychological
resources and adaptation. Review of general psychology, 6, 307-324.
Hobfoll, S.E. & Freddy, J. (1993). Conservation of
resources: A general stress theory applied to
burnout. In W. B. Schaufeli, C. Maslach and T.
Marek (Eds.), Professional burnout: Recent devel-
4/17/2012
1:07 PM
Page 37
opments in theory and practice (pp. 115-133).

Washington, D.C.: Taylor & Francis.
Hobfoll, S.E. & Shirom, A. (2001). Conservation of
resources theory. In R. Golembiewski (Ed.),
Handbook of Organizational Behavior (pp.
57-80). New York: Dekker.
Hochschild, A. R. (1979). Emotion Work, Feeling
Rules, and Social Structure. American Journal of
Sociology, 85(3), 551-575.
Hochschild, A.R. (2003). The Managed Hearth. Commercialization of Human Feeling (2nd ed.). London: University of California Press.
Holman, D., Chissick, C., Totterdell, P. (2002). The
effects of performance monitoring on emotional
labor and wellbeing in call centers. Motivation and
Emotion, 26, 57-81.
Hseih, C.W. & Guy, M. (2009). Performance Outcomes:
The Relationship Between Managing the Heart
and Managing Client Satisfaction. Review of
Public Personnel Administration, 291, 41-57
Hunter, B. & Smith, P. (2007). Emotional labour: just
another buzz word? International Journal of Nursing Studies, 44 (6), 859-861.
Johnson, H.A. (2004). The Story behind Service with a
Smile: The Effects of Emotional Labor on Job Satisfaction, Emotional Exhaustion, and Affective
Well-Being
Jones, R.G. & Best, R.G. (1995). An examination of the
impact of emotional work requirements on individual and organization. Paper presented at the Annual Convention of the Academy of Management,
Vancouver, British Columbia, Canada.
Keijsers, G. J., Schaufeli, W. B., Le Blanc, P. M.,
Zwerts, C., & Miranda, D. R. (1995). Performance
and burnout in intensive care units. Work & Stress,
9, 513-527.
Knapp, M.L., & Hall, J.A. (1997). Nonverbal communication in human interaction (4th ed.), Belmont,
California: Wadsworth, Thompson Learning.
Kruml, S.M, Geddes, D. (2000). Exploring the dimensions of emotional labor: The heart of
Hochschilds work. Management Communication
Quarterly, 14, 8-49.
Lazarus, R.S., Cohen-Charash, Y. (2001). Discrete emotions in organizational life. In R. L. Payne and G.
L. Cooper (Eds.), Emotions at work: Theory,
research and applications for management (pp.
45-8). Chichester: John Wiley & Sons.
Lee, L.T. & Ashforth, B.E. (1996). A meta-analytic
examination of the correlates of the three dimension of job burnout. Journal of Applied Psychology, 81, 123-133.
Levenson, R.W. (1994b). Emotional control: Variations
and consequences. In P. Ekman and R. J. Davidson
(Eds.), The nature of emotion (pp. 273-279). New
York: Oxford University Press.
Lin, W.B. (2008). Antecedents of employee involvement with the comparative model. Quality and
Quantity, 44, 459-482.
Maslach, C. (1978a). Job burn-out: How people cope.
Public Welfare, 36, 56-58.
Masclach, C. & Leiter, M. (2008). Early predictors of
job burnout and engagement. Journal of Applied
Psychology, 93(3), 498-512.
Maslach, C. & Jackson, S. (1981). The measurement of
experienced burnout. Journal of Occupational
Behavior, 2, 99-113.
Maslach, C., Schaufeli, W. & Leiter, M. (2001). Job
burnout. Annual Review Psychology, 52, 397-422.
Mikolajczak, M., Menil, C., & Luminet, O. (2007).
Explaining the protective effect of trait emotional
intelligence regarding occupational stress: exploration of emotional labour processes. Journal of
Research in Personality, 41, 1107-1117.
Miner, A.G., Glomb & T.M., & Hulin, C. (2005). Experience sampling mood and its correlates at work.
Journal of Occupational and Organizational Psychology, 78, 171-193.
Morris, A., & Feldman, D. (1996). The Dimensions,
Antecedents and Consequences of Emotional
Labor. Academy of Management Review, 21(4),
906-1010.
Parker, P. A., & Kulik, J. A. (1995). Burnout, self and
supervisor-related job performance, and absenteeism among nurses. Journal of Behavioral
Medicine, 18, 581599.
Pugh, D. (2001). Service with a smile: emotional contagion in the workplace. Academy of Management
Journal, 44, 5, 1018-1027.
Pugliesi, K. (1999). The Consequences of Emotional
Labor: Effects on Work Stress, Job Satisfaction
and Well-being. Motivation and Emotion, 23(2),
125-154.
Rafaeli, A., Semmer, N. & Tschan, F. (n.d.). Emotion in
Work Settings. Oxford Companion to the Affective
Sciences, Oxford University Press.
Rafaeli, A. & Sutton, R. (1987). Expression of Emotion
as Part of the Work Role. Academy of Management
Review, 12, 1, 23-37
Richards, J.M. & Gross, J.J. (2000). Emotion regulation
and memory: The cognitive cots f keeping ones
cool. Journal of Personality and Social Psychology, 79, 410-424.
Shirom, A. (2003). Burnout in work organizations. In C.
L. Cooper and I. Robertson (Eds.), International
Review of Industrial and Organizational Psychology (pp. 25-48). New York: Whiley.
Shirom, A. (2009). Epilogue: mapping future research
on burnout and health. Stress and Health, 25,
375-380.
Schaufeli, W.B. (2003). Past performance and future
perspectives of burnout research. Sotuh African
Journal of Industrial Psychology, 29(4), 1-15.
37
4/17/2012
1:07 PM
Page 38
Schaufeli, W.B. & Enzmann, D. (1998). The burnout

companion to study and practice: A critical analysis. London: Taylor & Francis.
Schaufeli, W.B. & Taris, T.W. (2005). The conceptualization and measurement of burnout: Common
ground and worlds apart. Work and Stress, 19,
256-262.
Shuler, S. & Sypher, S. (2000). Seeking emotional
labor: When Managing the Heart Enhances the
Work Experience. Management Communication
Quarterly, 14(1), 50-89.
Taris, T. W., Le Blanc, P. M., Schaufeli, W. B., &
Schreurs, P. J. G. (2005). Are there causal relationships between the dimensions of the Maslach
Burnout Inventory? A review and two longitudinal
tests. Work & Stress, 19, 241-258.
Tolich, M. (1993). Alienating and liberating emotions at
work: Supermarket Clerks Performance of Customer Service. Journal of Contemporary Ethnography, 22, 361-381.
Toppinen-Tanner, S., Kalimo, R. & Mutanen, P. (2002).
The process of burnout in white-collar and bluecollar jobs. Eight year prospective study of
exhaustion. Journal of Organizational Behavior,
23, 555-570.
Totterdell, P. & Holman, D. (2003). Emotion regulation
in customer service roles: Testing a model of emotional labor. Journal of Occupational Health Psychology, 8, 55-73.
38
van Dierendonck, D., Schaufeli, W. & Buunk, B.

(2001). Toward a process model of burnout:
Results from a secondary analysis. European
Journal of Work and Organizational Psychology,
10(1), 41-52.
Tsai, W.C. (2001). Determinants and consequences of
employee displayed positive emotions. Journal of
Management, 27, 497-512.
Zapf, D. (2002). Emotion work and psychological wellbeing. A review of the literature and some conceptual considerations. Human Resources Management Review, 12, 237-268
Zapf, D., Vogt, C., Seifert, C. & Mertini H., Isic, A.
(1999). Emotion work as a source of work stress:
The concept and development of an instrument.
European Journal of Work and Organizational
Psychology, 8, 371-400
Wharton, A.S. (1993). The affective consequences of
service work. Work and Occupations, 20, 205-232.
Wright, T.A. & Bonett, D.G. (1997) The role of pleasantness and activation-based wellbeing in performance prediction. Journal of Occupational Health
Psychology 2, 212-219.
Wright, T. A., & Cropanzano, R. (1998). Emotional
exhaustion as a predictor of job performance and
voluntary turnover. Journal of Applied Psychology, 83, 486-493.
4/17/2012
1:07 PM
Page 39
Evaluarea personalitii din perspectiva modelului Big Five. Date privind

adaptarea chestionarului IPIP-50 pe un eantion de studeni romni
Silvia Rusu1, Laureniu P. Maricuoiu, Irina Macsinga, Delia Vrg, Florin A. Sava
Abstract
International Personality Item Pool (IPIP) is a project aiming to develop measures of individual differences
as part of the public domain. This project emerged as an alternative to personality assessment instruments, which
are protected by copyright. IPIP-50 is an instrument developed through the IPIP project, which measures the five
dimensions of personality: Openness, Extraversion, Emotional Stability, Conscientiousness and Agreeableness.
The aim of the present study is to test the validity of IPIP-50 on a Romanian sample of students. The factor analysis revealed that the model with five correlated factors is the best suited to describe the structure of IPIP-50. The
convergent validity was assessed through the correlations between IPIP-50, DECAS (Sava, 2008) and NEO-FFI
(Costa & McCrae, 1992/2008); except for Agreeability, all the other factors registered high correlations (ranging
from .73 to .84). The predictive validity of IPIP-50 was assessed through its correlations with behavioral indicators identified in the literature as being relevant to the five dimensions. Using the Marlowe-Crown Social Desirability scale, we developed a social desirability indicator composed of 10 IPIP-50 items. The results reported in
this study suggest that IPIP-50 can be used successfully in research on Romanian samples, being an authentic
support for the psychological community in Romania.
Keywords: Five-Factor Model, factorial structure, validity, reliability, dezirability
Rsum
International Personality Item Pool (IPIP) est un projet visant dvelopper des outils dvaluation des diffrences interindividuelles dans le domaine public. Ce projet a merg comme une alternative aux instruments
traditionnels dvaluation de la personnalit protgs par copyright. IPIP-50 est un instrument developp dans le
projet lIPIP et conu pour valuer les cinq dimensions de la personnalit: lOuverture, lExtraversion, la Stabilit motionnelle, la Conscienscieux et lAgreabilit. La prsente tude vise adapter l chelle IPIP-50 sur un
chantillon dtudiants roumains. Lanalyse de la structure factorielle a rvl que, parmi les modles tests, le
modle cinq facteurs corrls dcrit le mieux la structure de lIPIP-50. La validit convergente a t value en
corrlant IPIP-50 avec DECAS (Sava, 2008) et NEO-FFI (Costa & McCrae, 1992 / 2008); sauf lAgreabilit,
toutes les autres chelles ont montr des corrlations leves (entre 0.38 et 0.70). En ce qui concerne la consistence interne, les chelles IPIP-50 enregistrent des valeurs acceptable (entre 0.73 et 0.84). La validit prdictive
de lIPIP-50 a t valu par rfrence des indicateurs comportementals pertinents pour chaque dimension,
identifis dans la littrature de specialit. En utilisant l chelle Marlowe-Crowne (dsirabilit sociale) a t construit un indicateur de dsirabilit sociale compos de 10 items IPIP-50. Les rsultats rapports dans cette tude
recommande lutilisation de lIPIP-50 dans la recherche en utilisant des chantillons roumaine, ce qui constitue
un vritable soutien pour la communaut psychologique de Roumanie.
Mots-cls: Modle des Cinq Facteurs, structure factorielle, validit, fidlit, desirabilit
Universitatea de Vest din Timioara.

Adresa de coresponden: rusu.silvia@ymail.com
39
4/17/2012
1:07 PM
Page 40
Evaluarea personalitii din perspectiva modelului Big Five. Date privind adaptarea chestionarului IPIP-50
Rezumat
International Personality Item Pool (IPIP) este un proiect care i propune dezvoltarea unor instrumente de
evaluare a diferenelor interindividuale ca parte a domeniului public. Acest proiect a aprut ca o alternativ la
instrumentele clasice de evaluare a personalitii protejate de drepturile de autor. IPIP-50 este un instrument dezvoltat n cadrul proiectului IPIP care evalueaz cele cinci dimensiuni ale personalitii: Deschidere, Extraversiune, Stabilitate emoional, Contiinciozitate i Agreabilitate. Studiul de fa i propune adaptarea scalei IPIP-50
pe un eantion de studeni romni. Analiza structurii factoriale a scos n eviden c din modelele testate, modelul celor cinci factori corelai descrie cel mai bine structura IPIP-50. Validitatea convergent a fost evaluat prin
corelarea IPIP-50 cu DECAS (Sava, 2008) i NEO-FFI (Costa & McCrae, 1992/2008); cu excepia factorului
Agreabilitate, toate celelalte scale au nregistrat corelaii ridicate (ntre .38 i .70). Referitor la consistena intern,
scalele IPIP-50 nregistreaz valori acceptabile (ntre .73 i .84). Validitatea predictiv a IPIP-50 a fost evaluat,
prin raportarea la indicatori comportamentali relevani pentru fiecare dimensiune, identificai n literatura de specialitate. Cu ajutorul scalei de dezirabilitate social Marlowe-Crowne (Crowne & Marlowe, 1960), a fost construit un indicator al dezirabilitii sociale format din 10 itemi IPIP-50. Rezultatele raportate n acest studiu recomand folosirea cu succes a IPIP-50 n cercetri care utilizeaz eantioane romneti, constituind un sprijin
autentic pentru comunitatea psihologic din Romnia.
Cuvinte cheie: modelul celor cinci factori, structur factorial, fidelitate, validitate, dezirabilitate
Introducere
Dup dezvoltarea a numeroase teorii cu privire la personalitate i decenii de cercetri n
domeniu, comunitatea tiinific internaional
se apropie de un consens cu privire la o teorie
general asupra trsturilor de personalitate:
Modelul Celor Cinci Factori (Five Factor
Model / Big Five). Sinteza realizat de John,
Naumann i Soto (2008) asupra Modelului celor
Cinci Factori i a istoriei acestuia scoate n eviden un interes tot mai crescut al comunitii
tiinifice din ultimii 30 de ani, n defavoarea
modelelor dezvoltate de Catell i Eysenk.
Popularitatea Modelului celor Cinci Factori
a dus la dezvoltarea unui numr mare de instrumente pentru evaluarea psihologic. n acest
moment n Romnia exist o serie de intrumente
pentru evaluarea psihologic din perspectiva
acestui model atestate de Colegiul Psihologilor.
Dei instrumentele la care s-a fcut referire
respect standarde ridicate privind validitatea i
fidelitatea, accesul cercettorilor sau a studenilor la acestea este condiionat de drepturile
de autor i de aspectele economice ce in de utilizarea lor n scop de cercetare.
O alternativ la metodele consacrate de
evaluare psihologic este reprezentat de proiectul International Personality Item Pool (Goldberg, 1999). Acest proiect i propune dezvoltarea de itemi pentru evaluarea psihologic a unei
game foarte largi de variabile specifice studiului
40
personalitii. IPIP-50 este una din scalele care

poate fi descrcat gratuit. Cei 50 de itemi sunt
prezentai sub forma unor afirmaii scurte (cte
10 itemi pentru fiecare din cele cinci dimensiuni). IPIP-50 este construit pentru identificarea
markerilor de personalitate aa cum au fost ei
prezentai de Goldberg n 1992, fr a surprinde
diferenele interindividuale la nivel de faete.
Dei utilizarea IPIP-50 este gratuit, transferul acestuia n cultura romneasc rmne ns
un demers nc nerealizat. Studiul de fa i
propune verificarea validitii factoriale i
fidelitii, a validitii convergente i predictive
a IPIP-50 pe un eantion de studeni romni. Un
obiectiv secundar al cercetrii este de a dezvolta un indicator de dezirabilitate social pentru
protocoalele IPIP-50. Prin atingerea acestor
obiective ne propunem s punem la dispoziia
comunitii tiinifice un instrument de evaluare
recunoscut la nivel internaional, echivalent din
punct de vedere al prestigiului i calitilor psihometrice cu intrumente distribuite n acest
moment sub licen n Romnia.
Evaluarea personalitii
din perspectiva Modelului Big Five
Modelul celor Cinci Factori a fost descoperit pe baz empiric de Tupes i Christal (1961)
prin reanalizarea datelor lui Catell cuprinznd
seturi de adjective bipolare. Cei cinci mari factori identificai au fost denumii Extraversiune,
Contiinciozitate, Stabilitate Emoional (Neu-
4/17/2012
1:07 PM
Page 41
roticism), Agreabilitate i Cultur (Intelect /

Deschidere). Dup descoperirea celor cinci factori, numeroase cercetri au confirmat identificarea acestora n seturi de date din cele mai
diverse (pentru o sintez, vezi John i colab.,
2008). Aceast abordare lexical este cunoscut sub numele de Modelul Big Five. Unii
autori fac deosebirea ntre modelul Big Five i
Modelul Celor Cinci Factori, acesta din urm
fiind atribuit abordrii utilizate n construcia
inventarelor de personalitate (Saucier & Golberg, 1996). Pe de alt parte, ali autori, consider c nu exist diferene calitative importante
ntre cele dou denumiri (Hendriks, Hofstee &
De Raad, 2002).
n prezent, unul din cele mai utilizate
instrumente de evaluare a personalitii din perspectiva Modelului Celor Cinci Factori este cel
dezvoltat de Costa i McCrae (1992). Pornind
de la datele lui Catell, acetia au dezvoltat un
chestionar (NEO Personality Inventory) care s
evalueze trei dimensiuni ale personalitii: neuroticismul, extraversiunea i deschiderea.
Observnd c acesta se suprapune cu trei din
cele cinci dimensiuni ale modelului Big Five,
cei doi autori au nceput revizuirea chestionarului pentru a cuprinde i evaluri ale celorlalte
dou dimensiuni: agreabilitate i contiinciozitate. Astfel, n 1992, Costa i McCrae au publicat NEO Personality Inventory Revised (NEOPI-R). Acest inventar de personalitate cuprinde
240 de itemi i evalueaz cei cinci mari factori
i cte ase faete pentru fiecare din acetia.
ncepnd cu 1990, Goldberg a dezvoltat un
set de liste cu adjective att unipolare, ct i
bipolare, cu scopul de a evalua personalitatea
din perspectiva Modelului Big Five. Pe parcursul mai multor studii aceste liste de adjective au
demonstrat o structur factorial congruent cu
Modelul Celor Cinci Factori (Goldberg, 1990,
1992). De asemenea, studiile au artat corelaii
ridicate ntre listele adjectivale dezvoltate de
Goldberg, NEO PI, Inventarul Hogan (Goldberg, 1992), dar i NEO- PI- R (McCrae &
Costa, 1997).
John i colab. (2008) realizeaz o comparaie interesant ntre trei din cele mai cunoscute
instrumente axate pe Modelul Celor Cinci Factori, pe baza studiilor desfurate de John i Soto
(2007) i Soto i John (2009). Folosind un
eantion de 827 de studeni, acetia au putut compara trei instrumente care evalueaz personalitatea din perspectiva Modelului Celor Cinci Factori: NEO FFI (forma scurt a NEO-PI-R),
varianta lui Saucier (1994) de 40 de itemi a listei
adjectivale construit de Goldberg n 1992 i Big
Five Inventory (BFI), un instrument care conine
44 de itemi sub forma unor propoziii scurte
(John, Donahue, & Kentle, 1991). Analiza realizat de John i colab. (2008) la nivelul coeficienilor de consisten intern (alfa Cronbach a
artat valori medii de .84 pentru lista de adjective,
.83 pentru BFI i .81 pentru NEO-FFI). La nivelul celor trei chestionare, scalele cu cei mai buni
coeficieni de fidelitate au fost Extraversiunea,
Neuroticismul i Contiinciozitatea, n timp ce
Agreabilitatea i Deschiderea au obinut indicatori mai puin satisfctori. Cu privire la validitatea convergent, s-au constatat diferene n
modul n care cele trei instrumente definesc
scalele Extraversiune i Deschidere. Astfel, faeta
cldur (n englez, warmth), inclus n NEOPI-R n scala Extraversiune coreleaz, de asemenea, cu scala de Agreabilitate. Pe de alt parte,
Goldberg (1992) a artat c adjectivele legate de
cldur coreleaz mai degrab cu Agreabilitatea dect cu Extraversiunea. Ca urmare, s-a
constatat o discrepan mai mare ntre scala de
Extraversiune a NEO-FFI i cele ale BFI i a Listei de adjective. La nivelul dimensiunii
Deschidere s-au remarcat diferene datorate modului de conceptualizare a acestui factor. Astfel,
Goldberg (1992) adopt o versiune a Deschiderii
bazat pe Intelect sau Imaginaie (asemntor cu
versiunea BFI), n timp ce scala Deschidere din
cadrul NEO-FFI abordeaz o perspectiv mai
larg, incluznd i faete legate de valori i comportamente flexibile. Ca atare, scala Deschidere
din NEO-FFI a demonstrat n cadrul acestei analize comparative o convergen mai redus cu
celelalte dou instrumente (John i colab., 2008).
Studiile desfurate de Goldberg (1990) au
condus spre o operaionalizare a celor cinci mari
factori sub forma a 100 de adjective. Analizele
derulate de Goldberg au grupat cele 100 de
adjective n mai multe categorii (clusteri) pentru
polul pozitiv i negativ al fiecruia din cei cinci
factori. n Tabelul 1 sunt prezentate categoriile
descriptive pentru cei cinci factori, conform viziunii lui Goldberg (1990, 1992).
41
4/17/2012
1:07 PM
Page 42
Tabelul 1. Modelul celor Cinci Factori, din perspectiva lui Goldberg (1990, 1992)
Dimensiune
Adjective
Polul pozitiv: vioiciune, comunicativ, jucu, expresivitate, spontaneitate, dezinhibiie, nivel
energetic, vorbre, afirmare, animaie, curaj, stim de sine, sinceritate, umor, ambiie, optimism.
Extraversiune
Polul negativ: distant, linitit, rezervat, timid, inhibiie, lips de agresivitate, pasivitate, letargie,
pesimism.
Polul pozitiv: cooperare, amabilitate, empatie, ngduin, politee, generozitate, flexibilitate,
modestie, moralitate, cldur, realism, naturalee.
Agreabilitate
Polul negativ: conflictual, critic excesiv, autoritarism, impolitee, cruzime, infatuare, iritabilitate,
ngmfare, ncpnare, nencredere, egoism, insensibilitate, ursuz, iretenie, prejudeci,
neprietenos, impredictibil, lipsa generozitii, nelciune.
Polul pozitiv: organizare, eficien, demn de ncredere, precizie, perseveren, precauie,
punctualitate, fermitate, demnitate, predictibilitate, cumptare, convenionalism, logic.
Contiinciozitate
Polul negativ: dezorganizare, neglijen, inconsisten, neatenie, nechibzuin, fr scop, lene,
indecizie, frivolitate, nonconformism.
Stabilitate emoional
(Neuroticism sczut)
Polul pozitiv: calm, independen.
Intelect
(Cultur/Deschidere)
Polul pozitiv: intelectualitate, profunzime, intuiie, inteligen, creativitate, curiozitate, sofisticat.
Polul negativ: insecuritate, fric, instabilitate, emoionalitate, invidie, naivitate, caracter bgre.
Polul negativ: superficialitate, lips de imaginaie, lipsa spiritului de observaie, prostie.
Proiectul International
Personality Item Pool
n 1996, Goldberg a pus bazele unui proiect
de colaborare internaional care s permit utilizarea gratuit a unor instrumente de evaluare a
personalitii i, n acelai timp, dezvoltarea lor
permanent prin eforturile reunite ale comunitii tiinifice internaionale (Goldberg,
1999). Proiectul a luat numele de International
Personality Item Pool (IPIP) i s-a materializat
sub
forma
unei
platforme
online
(http://ipip.ori.org/ ) n care sunt publicai itemi
pentru evaluarea diferenelor interindividuale.
Ulterior, au fost adugate traduceri ale acestor
itemi n diverse limbi, rezultate ale cercetrilor
i informaii cu privire la validitatea scalelor.
Proiectul IPIP s-a concretizat printr-un un efort
internaional de a dezvolta i a rafina n permanen un set de scale de personalitate, toate fiind
parte a domeniului public, disponibile att pentru utilizarea n scopuri tiinifice ct i n scopuri comerciale (Goldberg i colab., 2006, p.87).
n 2006, la 10 ani de la debutul proiectului,
Goldberg i colab. constat o cretere semnificativ a utilizrii IPIP. Scalele au fost traduse n
peste 25 de limbi i se nregistreaz un numr
42
tot mai mare de publicaii care utilizeaz instrumente IPIP. Goldberg i colab. (2006) consider
c aceast cretere n popularitate a IPIP se
datoreaz mai multor factori, dup cum
urmeaz: este absolut gratuit; itemii pot fi obinui foarte rapid de pe internet; include un numr
de peste 2000 de itemi, toi accesibili pentru
analiz; modalitatea de calculare a scorului este
disponibil; itemii pot fi prezentai n orice
ordine, reformulai, tradui n orice limb i
adminstrai prin intermendiul internetului fr a
cere permisiunea nimnui.
Metodologia cercetrii
Pentru validarea IPIP-50 pe un eantion
romnesc s-au folosit date secundare obinute n
cadrul mai multor studii desfurate ntre mai
2009 i mai 2010.
Participani
Pentru analiza validitii factoriale i a
fidelitii au fost utilizate datele reunite din mai
multe studii desfurate ntre mai 2009 i mai
4/17/2012
1:07 PM
Page 43
2010 (N=747). Acest eantion e format din studeni (70,83% de gen feminin), vrsta medie
fiind de 21,03 ani.
n ceea ce privete validitatea convergent,
aceasta a fost testat asupra unui eantion de
351 de studeni, reprezentnd datele reunite din
dou studii separate n cadrul crora participanii au completat IPIP-50 alturi de NEO-FFI
(Costa i McCrae 1992/2008) i DECAS (Sava,
2008). n cadrul acestui eantion, 76,84% din
participani sunt de gen feminin, media de vrst
fiind de 20,54 ani.
Validitatea predictiv a IPIP-50 a fost analizat folosind date secundare dintr-o serie de
studii n care au fost nregistrate comportamente
ale participanilor (att auto-declarate, ct i
observate obiectiv), specifice celor cinci dimensiuni urmrite. Deoarece datele au fost colectate
n studii diferite, numrul de participani variaz
de la un indicator comportamental la altul, ntre
46 i 234 de participani / indicator.
Pentru dezvoltarea indicatorului de

dezirabilitate social a fost folosit un eantion
de 520 de studeni (datele reunite dintr-o serie
de studii diferite) care au completat att IPIP-50,
ct i proba Marlowe-Crowne (Crowne & Marlowe, 1960) de evaluare a dezirabilitii sociale.
Eantionul este format din 73,67% participani
de gen feminin, vrsta medie fiind de 21,15 ani.
Instrumente
IPIP-50. Este un instrument dezvoltat n
cadrul proiectului International Personality Item
Pool i evalueaz markerii de personalitate
descrii de Goldberg (1992). Proba conine 50 de
itemi sub forma unor afirmaii scurte referitoare
la caracteristici i comportamente pe care participanii le evalueaz folosind o scal Likert de la
1 la 5, n funcie de ct de mult i caracterizeaz.
Cei 50 de itemi au fost tradui prin retroversiune
de ctre doi traductori independeni. Itemii n
limba romn sunt prezentai n Anexa 1.
Tabel 2. Indicatorii comportamentali

Indicatori comportamentali auto-declarai
Frecventarea bisericii
Spovedania n momente dificile
Spovedania de srbtori
Saroglou & Jaspard, 2000; Taylor

& MacDonald, 1999
Indicele de mas corporal

Consumul de tutun
Numrul de igri fumate pe zi
Consumul de alcool
Frecventarea de baruri/cluburi / discoteci
Apelul la consiliere / terapie
Exerciiile fizice practicate n mod regulat
Paunonen, 1998; Paunonen &

Ashton, 2001
Participarea la aciuni, programe de voluntariat
Carlo, Okun, Knight, Rosario &

Guzman, 2005
Numrul de persoane cu care a avut ntlniri romantice n afara unei relaii stabile
Noapte alb pentru coal n timpul sesiunii
Auto-evaluarea performanei academice
(sunt in primii ... % din anul meu de studiu)
Timpul de studiu din sptmna anterioar
Paunonen, 1998; Paunonen &

Ashton, 2001
Prezena la orele de curs
Conard, 2006
Indicatori comportamentali obiectivi
Performana academic
Conard, 2006; Poropat, 2009
Punctualitate n participarea la experiment

Sarcin experimental de persisten (volumul, corectitudinea, exactitatea n sarcin)
Back, Schmukle & Egloff, 2006
Evaluarea unei sarcini de scriere n oglind (plcut / neplcut)
Gormly, 1984, de Jonge & Slaets,

2005; Paunonen, 2003
43
4/17/2012
1:07 PM
Page 44
Inventarul de personalitate DECAS (Sava,

2008) cuprinde 97 de itemi cu rspuns dihotomic (adevrat/fals). Itemii sunt distribuii n
cele cinci scale: deschidere, extraversiune,
contiinciozitate, agreabilitate i stabilitate
emoional (scorul sczut indic neuroticismul).
Consistena intern a scalelor ntr-un eantion
romnesc reprezentativ variaz ntre .70 pentru
contiinciozitate i .75 pentru stabilitate
emoional.
NEO-FFI (Costa & McCrae, 1992/2008)
este varianta scurt a NEO-PI-R i evalueaz
cele cinci dimensiuni ale Modelului Celor Cinci
Factori: Extraversiune, Deschidere, Neuroticism, Contiinciozitate i Agreabilitate. NEOFFI cuprinde 60 de itemi evaluai pe o scal de
la 1 la 5, i ofer doar scorurile pentru cei cinci
factori, nu i pentru faetele acestora. Consistena intern a scalelor pentru Romnia variaz
ntre .73 pentru Deschidere i .82 pentru
Contiinciozitate.
Scala Marlowe-Crowne (Crowne & Marlowe, 1960) are 33 de itemi i evalueaz tendinele participanilor spre dezirabilitate social.
Scala nu distinge ntre cele dou faete ale
dezirabilitii sociale (Paulhus, 1984): tendine
de auto-decepie (se refer la cunotine
incorecte despre sine, care afecteaz auto-evalurile) i managementul impresiei (se refer la
intenia de a se prezenta n faa celorlali ntr-o
lumin mai bun dect n realitate).
Indicatori comportamentali. Analiza literaturii de specialitate a relevat mai muli indicatori
comportamentali specifici fiecrei dimensiuni.
Aceti indicatori pot fi mprii n dou mari
categorii: indicatori auto-declarai (de tip selfreport) sau indicatori obiectivi (observaii sau
nregistrri experimentale ale comportamentului). n tabelul de mai jos se regsete lista comportamentelor auto-declarate utilizate pentru
analiza validitii predictive IPIP-50, precum i
studiile care au artat relevana lor pentru
dimensiunile Big Five.
44
Rezultate
Validitatea factorial
i fidelitatea IPIP-50
n conformitate cu standardele de adaptare
(Albu, 1998), verificarea validitii factoriale se
impune n cazul msurrii mai multor dimensiuni prin acelai test. Pentru verificarea validitii
factoriale am optat pentru testarea mai multor
modele alternative de structurare a factorilor
IPIP-50, cu ajutorul modelrii prin ecuaii structurale (engl. structural equation modelling
SEM).
Analizele au fost realizate folosind rspunsurile oferite de 747 de studeni (70,83% gen
feminin, medie de varsta 21,03 ani), care au
completat chestionarul IPIP-50 ntre mai 2009 i
mai 2010.
Modelele testate au fost: (a) un model care
stipula existena unui singur factor latent; (b) un
model care stipula existena a 5 factori lateni
independeni i (c) un model care stipula existena a 5 factori lateni intercorelai. Primul
model, care stipula existena unui singur factor
latent, a fost testat pentru a evalua gradul n care
rspunsurile la itemii IPIP-50 sunt influenate de
distorsiunea datorat metodei comune (engl.
common method bias Podsakoff, MacKenzie,
Lee & Podsakoff, 2003). Dac acest model este
adecvat, putem afirma c rspunsurile nu sunt
influenate de existena celor cinci factori de
personalitate, ci de tendina participanilor de a
oferi rspunsuri similare la toi itemii acestui
instrument de evaluare. Cel de-al doilea model a
fost testat pentru a verifica dac cei cinci factori
de personalitate msurai de IPIP-50 se comport n conformitate cu perspectiva clasic
asupra modelului Big Five, care stipula existena a cinci factori necorelai. Cel de-al treilea
model testat ine cont de rezultatele meta-analizelor (van der Linden, te Nijenhuis & Bakker,
2010) care au indicat corelaii stabile ntre cei
cinci factori ai modelului.
4/17/2012
1:07 PM
Page 45
Tabel 3. Indici de potrivire a modelelor testate
GFI/AGFI
RMSEA
IFI
CFI
1 factor unic
(1175) = 8729.34, p <.001
.572/.536
.093
.350
.348
5 factori necorelai
(1175) = 5209.00, p <.001
.742/.721
.068
.653
.652
5 factori corelai
(1147) = 3435.54, p <.001
.829/.809
.052
.804
.802
Model
Tabel 4. Consistena intern i corelaiile inter-scale

St.emo.
Extrav.
Deschidere
Agreab.
Stab. emoional
.84
Extraversiune
.27
.84
Deschidere spre experien
.15
.35
.75
Agreabilitate
.12
.30
.34
.73
Contiinciozitate
.24
.17
.17
.27
Cont.
.78
N = 747. Valorile prezentate n italic reprezint coeficienii de fidelitate alfa-Cronbach. Toi coeficienii de corelaie sunt semnificativi la p<.01
Dup cum se poate observa din Tabelul 3,

modelul cu 1 factor unic are valori foarte sczute
ale indicilor de potrivire. Acest rezultat ne indic
faptul c eroarea metodei comune se manifest
n mic msur n setul de rspunsuri analizat.
De asemenea, rezultatele raportate n Tabelul 3
indic faptul c modelul a cinci factori corelai
este cel mai potrivit pentru a descrie structura
factorial a IPIP-50. Acest rezultat ne indic fap-
tul c IPIP-50 prezint un nivel acceptabil de

invarian configural, ceea ce permite utilizarea
instrumentului n cultura romn.
Din punct de vedere al consistenei interne,
scalele IPIP-50 nregistreaz valori acceptabile
(ntre .73 i .84), iar valorile corelaiilor ntre
scalele IPIP-50 sunt similare cu valorile medii ale
corelaiilor raportate n literatura de specialitate.
45
4/17/2012
1:07 PM
Page 46
Tabel 5. Indici path standardizai pentru modelul cu cinci factori corelai

Item
Stab.emo.
Extrav.
Deschidere
Agreab.
Contiinc.
ipip9
Aproape tot timpul sunt relaxat ().
.527
ipip19
Rareori m simt trist.
.198
ipip49
Adesea m simt melancolic()
-.518
ipip29
M supr cu uurin.
-.556
ipip24
Sunt perturbat() cu uurin.
-.596
ipip34
mi schimb dispoziia de multe ori.
-.603
ipip14
M ngrijorez.
-.640
ipip44
M irit uor.
-.653
ipip39
Am schimbri frecvente de dispoziie.
-.674
ipip4
M stresez cu uurin.
-.758
ipip31
Vorbesc cu o mulime de oameni diferii la petreceri.
.697
ipip21
Iniiez conversaii.
.650
ipip1
Sunt sufletul petrecerii.
.614
ipip11
M simt confortabil n jurul oamenilor.
.539
ipip41
Nu m deranjeaz s fiu n centrul ateniei.
.488
ipip36
Nu mi place s atrag atenia asupra mea.
-.402
ipip26
Am puine de spus.
-.553
ipip6
Nu vorbesc mult.
-.618
ipip46
n jurul oamenilor necunoscui sunt tcut.
-.627
ipip16
Rmn n umbr.
-.677
ipip50
Sunt plin() de idei.
.741
ipip25
Am idei excelente.
.633
ipip15
Am o imaginaie vie.
.520
ipip35
Sunt rapid() n nelegerea lucrurilor.
.495
ipip5
Am un vocabular bogat.
.481
ipip40
Folosesc cuvinte dificile.
.280
ipip45
mi petrec timp reflectnd asupra lucrurilor.
.178
ipip20
Nu sunt preocupat() de ideile abstracte.
-.358
ipip10
neleg greu ideile abstracte.
-.467
ipip30
Nu am o imaginaie bogat.
-.471
ipip42
Simt emoiile celorlali.
.561
ipip17
neleg sentimentele altora.
.540
ipip37
mi fac timp pentru alii.
.525
ipip27
Am o inim cald.
.511
ipip7
mi plac oamenii.
.449
ipip47
Fac oamenii s se simt linitii.
.387
ipip12
Jignesc oamenii.
-.263
ipip2
Sunt puin preocupat() de alii.
-.316
ipip22
Nu m intereseaz problemele altora.
-.470
ipip32
Nu sunt chiar interesat() de ceilali.
-.525
ipip33
mi place ordinea.
.617
ipip48
Sunt exact() n munca pe care o fac.
.539
ipip3
Sunt ntotdeauna pregtit().
.470
ipip43
Urmez un orar.
.446
ipip13
Sunt atent () la detalii.
.369
ipip23
Termin treaba repede.
.345
ipip18
nvlmesc lucrurile.
-.516
ipip38
mi evit responsabilitile.
-.564
ipip8
mi las lucrurile peste tot.
-.579
ipip28
Uit des s pun lucrurile napoi la locul lor.
-.604
46
4/17/2012
1:07 PM
Page 47
Tabel 6. Corelaii ntre scalele IPIP-50 i scalele echivalente din NEO-FFI i DECAS
NEO-FFI
DECAS
Stabilitate emoional
-.536
.698
Extraversiune
.383
.687
Deschidere spre experien
.437
.610
Agreabilitate
.248
.320
Contiinciozitate
.486
.583
N = 351. Toi coeficienii de corelaie sunt semnificativi la p < .01
scalele IPIP-50 i scalele echivalente din NEOFFI, respectiv DECAS. Excepia de la aceast
afirmaie este reprezentat de scala Agreabilitate, care nregistreaz corelaii mai sczute cu
scalele echivalente din NEO-FFI (r = .248) i
DECAS (r = .320).
Validitatea convergent
Validitatea convergent a fost evaluat prin
corelarea scorurilor IPIP-50 cu scorurile
obinute de 351 de studeni (76,84% gen feminin, medie de vrst 20,54 ani) la dou probe
paralele: DECAS (Sava, 2008) i NEO-FFI
(Costa & McCrae, 1992/2008). Aceste dou
probe msoar aceleai cinci variabile de personalitate, sunt avizate de Colegiul Psihologilor
i sunt distribuite sub licen n Romnia.
Dup cum se poate observa din Tabelul 6,
exist corelaii ridicate (ntre .38 i .70) ntre
Validitatea predictiv
Capacitatea de a prezice comportamentul
pornind de la un set de rspunsuri reprezint o
caracteristic de dorit a oricrui instrument psihologic. ntr-o meta-analiz recent, Green-
Tabel 7. Indicatori comportamentali relevani pentru scala Stabilitate emoional

IPIP-50
NEO-FFI
DECAS
-.17*
.13*
-.21**
234
232
227
-.17*
.14*
-.21**
227
226
222
-.34*
-.19
46
48
.21**
-.20
.23**
147
120
150
-.17*
-.09
147
150
.21**
.22*
148
.18*
-.21*
.13
121
120
117
Numrul de igri fumate pe zi
Frecventarea de baruri / cluburi / discoteci
Exerciiile fizice n mod regulat

151
Consumul de alcool
*corelaie semnificativ la p<.05; **corelaie semnificativ la p<.01
47
4/17/2012
1:07 PM
Page 48
wald, Poehlman, Uhlmann i Banaji (2009) au

artat c instrumentele de tip creion-hrtie coreleaz destul de modest cu evalurile comportamentale, raportnd o corelaie medie r = .27.
Aceast valoare este mult sub valorile nregistrate atunci cnd discutm despre corelaiile
ntre dou chestionare de auto-raportare a comportamentului (corelaii ce iau valori peste .40),
dar poate fi considerat o valoare-reper n evaluarea validitii predictive a scalelor IPIP-50.
Validitatea predictiv a IPIP-50 a fost evaluat prin corelarea corelaiilor dintre scorurile
obinute de subieci i o serie de indicatori comportamentali relevani pentru fiecare dimensiune n parte. Pentru a demonstra echivalena
dintre rezultatele IPIP-50 i rezultatele obinute
cu ajutorul unor chestionare consacrate, am
raportat i indicii de validitate predictiv obinui de aceste dou chestionare.
n tabelele ulterioare sunt raportate corelaiile propriu-zise ntre fiecare dintre aceti indicatori i rezultatele obinute la cele trei instrumente de evaluare psihologic. Deoarece datele
au fost culese n studii diferite, numrul de
subieci variaz de la un indicator la altul. Din
acest considerent, am raportat att valoarea coe-
ficientului de corelaie, ct i numrul de participani pe care a fost calculat acest coeficient.

Este important s menionm c n timp ce
scorurile ridicate la IPIP-50 i DECAS reprezint stabilitate emoional ridicat, scorurile mari
la NEO-FFI reprezint neuroticism ridicat (stabilitate emoional sczut).
La nivelul Stabilitii emoionale, s-a constatat c cele trei instrumente utilizate coreleaz
cu indicatorii comportamentali n aceai
direcie. Astfel, stabilitatea emoional sczut
este asociat cu frecventarea bisericii i
spovedania n momente dificile, cu un numr
ridicat de igri consumate zilnic i cu apelul la
consiliere sau terapie. Pe de alt parte, stabilitatea emoional ridicat este asociat cu
frecventarea de baruri / cluburi ca modalitate de
petrecere a timpului liber, practicarea exerciiilor fizice n mod regulat i consumul mai ridicat de alcool (vezi Tabelul 7).
n ceea ce privete Extraversiunea, cele trei
instrumente utilizate au artat un pattern similar
al corelaiilor cu indicatorii comportamentali.
Astfel, toate cele trei instrumente au scos n eviden corelaii pozitive ntre extraversiune i
participarea la programe de voluntariat i
Tabel 8. Indicatori comportamentali relevani pentru scala Extraversiune

IPIP-50
NEO-FFI
DECAS
.20**
-.10
.21**
234
232
227
.17**
.13*
.18**
232
232
227
.27**
.12
.32**
121
120
117
.19*
.001
.09
121
120
117
.22*
.35**
.17
121
120
117
.25**
.22*
.25**
126
125
121
Consumul de alcool
Participarea la aciuni / programe de voluntariat
Fumat
Evaluarea unei sarcini de scriere n oglind
* corelaie semnificativ la p<.05;

** corelaie semnificativ la p<.01
48
4/17/2012
1:07 PM
Page 49
Tabel 9. Indicatori comportamentali relevani pentru scala Deschidere spre experien

IPIP-50
NEO-FFI
DECAS
.12*
.10
.01
234
232
227
.13*
.14*
.13*
232
230
225
.11
.03
.16*
233
231
226
.17**
.10
.16*
234
232
227
-.23*
-.17
-.05
121
117
120
Numrul de persoane cu care a avut ntlniri romantice n afara unei

relaii stabile
Noapte alb pentru coal n timpul sesiunii

frecventarea bisericii. Tendina persoanelor

extravertite de a oferi rspunsuri pozitive (de
Jonge & Slaets, 2005) a fost evideniat de cele
trei instrumente prin corelaiile pozitive ntre
extraversiune i evaluarea unei sarcini de scriere
n oglind ca fiind plcut sau neplcut. Doi
din indicatorii comportamentali utilizai

(cosumul de alcool i tutun) au nregistrat corelaii diferite cu cele trei instrumente. Consumul
de alcool coreleaz pozitiv cu scala de extraversiune din cadrul IPIP-50 i DECAS, i negativ
(la nivel nesemnificativ statistic) cu scala de
Tabel 10. Indicatori comportamentali relevani pentru scala Agreabilitate
Numrul de persoane cu care a avut ntlniri romantice n afara unei relaii

stabile
IPIP-50
NEO-FFI
DECAS
-.14*
-.09
-.17**
232
230
225
-.13*
-.16*
-.18**
234
232
227
-.15*
-.10
-.14*
234
232
227
.18*
.12
.27**
147
232
151
-.12*
-.11
-.12*
225
223
218
.26**
.22**
139
142
Fumatul
Consumul de alcool
Auto-evaluarea performanei academice

(sunt n primii ... % din anul meu de studiu)
Timpul de studiu din sptmna anterioar

49
4/17/2012
1:07 PM
Page 50
extraversiune din cadrul NEO-FFI. n ce

privete fumatul, dei corelaiile cu cele trei
scale sunt n aceai direcie, doar corelaia cu
scala din cadrul IPIP-50 atinge un prag semnificativ. Aceste diferene pot fi explicate pe de o
parte de conceptualizarea diferit a Extraversiunii la nivel de faete n cele trei instrumente, i
pe de alt parte prin specificul i complexitatea
comportamentelor n cauz (consumul de tutun
i alcool).
A fost identificat un pattern similar al corelaiilor ntre indicatorii comportamentali specifi-
ci Deschiderii spre experien i scalele aferente

din cadrul celor trei instrumente. Participarea la
programe de voluntariat i frecventarea de
baruri n timpul liber coreleaz semnificativ
doar cu scala Deschidere spre experien din
cadrul IPIP-50, n timp ce celelalte dou scale
nregistreaz corelaii n direcia ateptat, dar
fr a atinge pragul de semnificaie. Comportamentul noapte alb pentru coal n timpul
sesiunii coreleaz semnificativ cu scala
Deschidere din IPIP-50, dar nesemnificativ cu
scala aferent din NEO-FFI, n mod consistent
Tabel 11. Indicatori comportamentali relevani pentru scala Contiinciozitate

IPIP-50
NEO-FFI
DECAS
.17**
.17**
.27**
231
228
223
.13*
.26**
.12
234
232
227
.20*
.22*
147
152
.19*
.22**
149
154
.19*
.22**
149
151
-.28**
-.11
-.27**
234
232
227
-.24**
-.04
-.20**
234
232
227
-.17**
-.08
-.13*
232
230
225
.24**
.08
.28**
233
231
226
.23**
.11
.21**
230
229
224
Performana academic
Punctualitate n participare la experiment
Sarcin experimental de persisten volum
Sarcin experimental de persisten adunri corecte
Sarcin experimental de persisten exactitate
Fumatul
Consumul de alcool
ntlniri romantice cu alte persoane dect iubitul/iubita
Spovedania de srbtori
r
.20**
.10
.19**
227
226
222
.19**
.10
.09
233
231
226
.10
.18**
.17**
234
232
227
.24**
.28**
146
151
Exerciiile fizice n mod regulat
Prezenele la orele de curs

50
4/17/2012
1:07 PM
Page 51
ateptat cu cele trei instrumente de evaluare a

personalitii.
cu conceptualizarea diferit a Deschiderii de

ctre Goldberg (1992) ca fiind bazat pe intelect, fa de conceptualizarea mai larg oferit
de Costa i McCrae (1992) care include aspecte
legate de valori i comportamente flexibile.
Tabelul 10 prezint un pattern asemntor
al corelaiilor ntre scalele de Agreabilitate ale
celor trei instrumente i indicatorii comportamentali specifici acestei dimensiuni. Conform
literaturii de specialitate, persoanele cu Agreabilitate ridicat au o varietate sczut a ntlnirilor romantice, precum i un consum redus de
tutun i alcool (Paunonen, 1998; Paunonen,
2003). Agreabilitatea ridicat este de asemenea
un predictor al participrii la aciuni de voluntariat (Carlo i colab., 2005). Aceste aspecte
sunt confirmate de patternul corelaiilor obinut
n acest studiu ntre scala Agreabilitate a
IPIP-50 i indicatorii comportamentali. Autoraportrile asupra performanei academice i a
timpului de studiu observate n Tabelul 10 sugereaz promovarea unei imagini pozitive a propriei persoane, specific Agreabilitii ridicate.
Tabelul 11 scoate n eviden faptul c cele
trei instrumente utilizate coreleaz cu indicatorii
comportamentali n aceai direcie (att cu cei
auto-declarai ct i cu cei observai). Toate cele
trei instrumente coreleaz pozitiv cu performana academic precum i cu performana participanilor ntr-o sarcin de persisten, n mod
consistent cu datele din literatura de specialitate
(Conrad, 2006; Poropat, 2009). Punctualitatea, o
trstur specific persoanelor cu contiinciozitate ridicat (Back i colab., 2006), coreleaz de
asemenea cu scalele de Contiinciozitate ale
celor trei instrumente. Un set de comportamente
auto-declarate care s-au dovedit a fi specifice
Contiiciozitii (Paunonen, 1998; Paunonen,
2003, Saroglou & Jaspard, 2000; Taylor & MacDonald, 1999) coreleaz de asemenea n direcia
Identificarea rspunsurilor dezirabile

n protocolul IPIP-50
Tendina persoanelor evaluate de a
rspunde ntr-o manier care s le pun ntr-o
lumin favorabil este o realitate cunoscut a
procesului de evaluare psihologic. Indiferent
dac aceast tendin este rezultatul unui proces
contient sau mai puin contientizat (precum
auto-amgirea), evaluatorul trebuie s aib la
dispoziie instrumentele necesare pentru a evalua probabilitatea de apariie a rspunsurilor
dezirabile din punct de vedere social.
Pornind de la aceste considerente, n
aceast seciune propunem un indicator de
dezirabilitate social pentru protocoalele IPIP50. Realizarea acestui indicator a presupus completarea Scalei de dezirabilitate social Marlowe-Crowne (Crowne & Marlowe, 1960),
alturi de IPIP-50.
Folosind regresia liniar multipl, am fost
interesai s identificm itemii IPIP-50 care
prezic cel mai bine scorurile ridicate la Scala de
dezirabilitate social Marlowe-Crowne. Am
ncercat s echilibrm itemii care intr n componena indicatorului, din punctul de vedere al
scalei din care fac parte. Varianta final a acestui indicator este format din 10 itemi, cte 2 din
fiecare factor de personalitate.
Analizele statistice au fost realizate pe un
lot de 520 de studeni (73,67% gen feminin,
vrsta medie 21,15 ani), care au completat
ambele probe. n varianta final, indicatorul de
dezirabilitate social din cadrul IPIP-50 (IPIP50-DS) coreleaz semnificativ cu Scala de
dezirabilitate Marlowe-Crowne (r(518) = .482,
p<.001) i poate fi calculat folosind urmtoarea
ecuaie de regresie:
Tabelul 12. Statistici descriptive ale indicatorului IPIP-50-DS

Indice de asimetrie
Min.
IPIP50-DS
11.140
Max.
24.280
Medie
17.728
Indice de boltire
Ab.standard
2.271
Valoare
Std. Error
Valoare
Std. Error
-.225
.107
-.078
.214
51
4/17/2012
1:07 PM
Page 52
Pornind de la valorile tendinei centrale i

dispersiei, putem afirma c scoruri mai mari de
21,12 la IPIP-50-DS indic o tendin accentuat a respondentului spre oferirea unei imagini
pozitive despre sine.
Discuii
Studiul realizat pe un eantion romnesc a
artat c Modelul cu Cinci Factori corelai
descrie cel mai bine structura factorial a scalei
IPIP-50, fiind n conformitate cu perspectivele
recente asupra modelului Big Five (van der Linden i colab., 2010). Modul n care itemii ncarc fiecare factor este n similar cu distribuia
acestora pe scale propus de Goldberg
(http://ipip.ori.org/), chestionarul demonstrnd
astfel un nivel acceptabil de invarian configural. n grupul testat, scalele IPIP-50 au artat o
consisten intern satisfctoare (ntre .73 pentru agreabilitate i .84 pentru extraversiune i
stabilitate emoional).
Corelaiile ridicate ntre scalele IPIP-50 i
scalele corespondente ale altor dou chestionare
care msoar cei cinci mari factori de personalitate (NEO-FFI i DECAS) sprijin validitatea
convergent a IPIP-50. Scala Agreabilitate din
cadrul IPIP-50 a fost singura care a nregistrat
corelaii mai sczute cu scalele corespondente
ale NEO-FFI i DECAS. Consistena intern
mai sczut a scalei de Agreabilitate din cadrul
IPIP-50 comparativ cu celelalte scale ale instrumentului (.73) este o potenial cauz a acestor
corelaii mai sczute. De asemenea, conceptualizarea uor diferit a Agreabilitii n cele trei
instrumente (la nivelul faetei cldur) poate
explica patternul corelaiilor observat ntre cele
trei scale de Agreabilitate.
n ce privete validitatea predictiv a scalei,
rezultatele au artat corelaii semnificative ntre
IPIP-50 i o serie de comportamente (att autodeclarate, ct i observate/obiective) specifice
celor cinci factori de personalitate. Comportamentele evaluate n cadrul studiului au fost deja
asociate trsturilor de personalitate n diverse
studii anterioare. De asemenea, patternul corelaiilor ntre IPIP-50 i indicatorii comportamentali este similar cu cel al corelaiilor ntre
52
NEO-FFI, respectiv DECAS i aceiai indicatori comportamentali.

Referitor la limitele IPIP-50, subliniem faptul c studiul de adaptare a fost realizat asupra
unui eantion format din studeni, posibilitile
de generalizare fiind aadar limitate. Totodat,
nu exist studii sistematice de validare a instrumentului, asfel nct acesta nu ndeplinete standardele testelor publicate profesionist. Utilitatea
IPIP-50 n practica psihologic nu a fost nc
investigat prin studii aplicative asupra unor
eantioane romneti. Din toate aceste considerente, IPIP-50 este un instrument extrem de util
n cercetare, dar se recomand folosirea lui cu
pruden n practica psihologic.
Utilizarea unei scale de dezirabilitate
social (Crowne & Marlowe, 1960) mpreun
cu IPIP-50 a permis identificarea itemilor IPIP
care prezic cel mai bine scorurile ridicate ale
scalei de dezirabilitate social. n acest fel a fost
construit un indicator al dezirabilitii sociale
format din 10 itemi IPIP-50 (cte doi din fiecare
scal). Aceast scal este, din cunotinele noastre, unic pn n acest moment. IPIP-50-DS va
permite cercettorilor s evalueze n ce msur
protocoalele IPIP sunt afectate de rspunsuri
dezirabile din punct de vedere social.
Considerm c sunt necesare studii i analize suplimentare asupra IPIP-50-DS nainte de a
recomanda utilizarea acestei scale n practica
psihologic. n absena unui eantion diferit de
cel pe baza cruia a fost dezvoltat scala, precum i a unui standard clar care distinge ntre
participanii cu dezirabilitate ridicat i cei cu
dezirabilitate sczut la Marlowe-Crown, nu a
fost realizat analiza sensibilitii i a specificitii scalei IPIP-50-DS, i nici nu a fost stabilit un scor optim de cut-off pentru aceast
scal. Cercetri viitoare n care participanii sunt
instruii s prezinte sau s nu prezinte o imagine
dezirabil social ar putea aduce aceste
mbuntiri scalei IPIP-50-DS. n acest fel
practicienii care utilizeaz scale de personalitate
n evaluarea psihologic vor avea posibilitatea
de a identifica acele persoane al cror profil
IPIP-50 prezint un risc crescut de biasare
datorat dezirabilitii sociale. Selecia de personal i alte domeni n care indivizii sunt motivai s se prezinte ntr-o lumin favorabil pot
4/17/2012
1:07 PM
Page 53
avea de ctigat prin utilizarea indicatorului de

dezirabilitate social IPIP-50-DS.
n concluzie, rezultatele raportate n acest
studiu arat c din punct de vedere al structurii
factoriale, fidelitii, validitii convergente i
predictive, IPIP-50 este un instrument satisfctor, care poate fi folosit cu succes n studii care
utilizeaz eantioane romneti.
Bibliografie
Albu, M. (1998). Construirea i utilizarea testelor psihologice. Cluj Napoca: Clusium.
Back, M., Schmukle, S., & Egloff, B. (2006). Who is
late and who is early? Big Five personality factors
and punctuality in attending psychological experiments. Journal of Research in Personality, 40(5),
841-848. doi: 10.1016/j.jrp.2005.11.003.
Carlo, G., Okun, M. A., Knight, G. P., Rosario, M., &
Guzman, T. D. (2005). The interplay of traits and
motives on volunteering : agreeableness, extraversion and prosocial value motivation. Personality
and Individual Differences, 38, 1293-1305. doi:
10.1016/j.paid.2004.08.012.
Conard, M. A. (2006). Aptitude is not enough : How
personality and behavior predict academic performance. Journal of Research in Personality, 40,
339-346. doi: 10.1016/j.jrp.2004.10.003.
Costa, P.T., & McCrae, R.R. (1992/2008). Manual
tehnic. Inventarul de personalitate NEO FFI.
Cluj-Napoca, Romania: Sinapsis.
Crowne, D.P., & Marlowe, D. (1960). A new scale of
social desirability independent of psychopathology. Journal of Consulting Psychology, 24,
349-354. doi:10.1037/h0047358.
De Jonge, P., & Slaets, J.P.J. (2005). Response sets in
self-report data and their associations with personality traits. The European Journal of Psychiatry,
19(4),
209-214.
doi:
10.4321/S021361632005000400002.
Goldberg, L. R. (1990). An Alternative Description of
Personality: The Big-Five Factor Structure. Journal of Personality and Social Psychology, 59(6),
1216-1229. doi: 10.1037/0022-3514.59.6.1216.
Goldberg, L. R. (1992). The Development of Markers
for the Big-Five Factor Structure. Psychological
Assessment, 4(1), 26-42. doi: 10.1037/10403590.4.1.26.
Goldberg, L. R. (1999). A broad-bandwidth, public
domain, personality inventory measuring the
lower-level facets of several five-factor models. In
I. Mervielde, I. Deary, F. De Fruyt, & F. Ostendorf
(Eds.), Personality Psychology in Europe, Vol. 7
(pp. 7-28). Tilburg, The Netherlands: Tilburg University Press.

Goldberg, L. R., Johnson, J. A., Eber, H. W., Hogan, R.,
Ashton, M. C., Cloninger, C. R., & Gough, H.G.
(2006). The international personality item pool
and the future of public-domain personality measures. Journal of Research in Personality, 40, 8496. doi: 10.1016/j.jrp.2005.08.007.
Gormly, J. (1984). Correspondence between personality trait ratings and behavioral events. Journal of
Personality, 52(3), 220-232. doi: 10.1111/j.14676494.1984.tb00878.x.
Greenwald, A.G., Poehlman, T.A., Uhlmann, E.L. &
Banaji, M.R. (2009). Understanding and using the
implicit association test: III. Meta-analysis of predictive validity. Journal of Personality and Social
Psychology, 97, 17-41. doi: 10.1037/a0015575.
Hendriks, A. A. J., Hofstee, W. K. B., & De Raad, B.
(2002). The Five-Factor Personality Inventory:
Assessing the Big Five by means of brief and concrete statements. In B. De Raad & M. Perugini
(Eds.), Big Five assessment (pp. 79-108). Gttingen, Germany: Hogrefe & Huber.
International Personality Item Pool: A Scientific Collaboratory for the Development of Advanced Measures of Personality Traits and Other Individual
Differences (http://ipip.ori.org/). Internet Web
Site.
John, O. P., Donahue, E. M., & Kentle, R. L. (1991).
The Big Five InventoryVersions 4a and 54.
Berkeley: University of California, Berkeley,
Institute of Personality and Social Research.
John, O. P., Naumann, L. P., & Soto, C. J. (2008).
Paradigm Shift to the Integrative Big-Five Trait
Taxonomy: History, Measurement, and Conceptual Issues. In O. P. John, R. W. Robins, & L. A. Pervin (Eds.), Handbook of personality: Theory and
research (pp. 114-158). New York, NY: Guilford
Press.
John, O. P., & Soto, C. J. (2007). The importance of
being valid: Reliability and the process of construct validation. In R. W. Robins, R. C. Fraley, &
R. F. Krueger (Eds.), Handbook of research methods in personality psychology (pp. 461494). New
York: Guilford Press.
McCrae, R., & Costa, P. (1997). Personality trait structures as a human universal. American Psychologist, 52, 509-516. doi: 10.1037/0003066X.52.5.509.
Paulhus, D.L. (1984). Two-component models of
socially desirable responding. Journal of Personality and Social Psychology, 46, 598-609.
doi:10.1037/0022-3514.46.3.598.
Paunonen, S. V. (1998). Hierarchical Organization of
Personality and Prediction of Behavior. Journal of
Personality and Social Psychology, 74(2), 538556. doi: 10.1037/0022-3514.74.2.538.
53
4/17/2012
1:07 PM
Page 54
Paunonen, S. V. (2003). Big Five Factors of Personality

and Replicated Predictions of Behavior. Journal of
Personality and Social Psychology, 84(2), 411424. doi: 10.1037/0022-3514.84.2.411.
Paunonen, S.V. & Ashton, M.C.(2001). Big Five factors
and Facets and the Prediction of Behavior. Journal
of personality and Social Psychology, 81(3), 524539. doi: 10.1037//0022-3514.81.3.524.
Podsakoff, P.M., MacKenzie, S.B., Lee, J.Y., & Podsakoff, N.P. (2003). Common method biases in
behavioral research: a critical review of the literature and recommended remedies. Journal of
Applied Psychology, 88, 879-903. doi:
10.1037/0021-9010.88.5.879
Poropat, A.E. (2009). A meta-analysis of the five-factor
model of personality and academic performance.
Psychological Bulletin, 135(2), 322-338. doi:
10.1037/a0014996
Saroglou, V. & Jaspard, J.-M. (2000). Personality and
religion: From Eysencks taxonomy to the FiveFactor Model. Archiv fr Religionpsychologie 23,
4170.
Saucier, G. (1994). Mini-markers: A brief version of
Goldbergs unipolar Big-Five markers. Journal of
Personality Assessment, 63, 506516. doi:
10.1207/s15327752jpa6303_8.
Saucier, G., & Goldberg, L. R. (1996). The language of
personality: Lexical perspectives on the Five-Fac-
tor Model. n J. S.Wiggins (Ed.), The Five-Factor

Model of personality: Theoretical perspectives
(pp. 21-50). New York: Guilford.
Sava, F.A. (2008). Inventarul de personalitate DECAS.
Manualul utilizatorului. Timisoara: ArtPress.
Soto, C., & John, O. (2009). Using the California Psychological Inventory to assess the Big Five personality domains: A hierarchical approach. Journal of Research in Personality, 43(1), 25-38. doi:
10.1016/j.jrp.2008.10.005.
Taylor A. & MacDonald D.A. (1999). Religion and the
five factor model of personality: An exploratory
investigation using a Canadian university sample.
Personality and Individual Differences, 27(6),
1243-1259. doi:10.1016/S0191-8869(99)00068-9.
Tupes, E. C., & Christal, R. C. (1961). Recur rent personality factors based on trait ratings. Technical
Report, U.S. Air Force, Lackland Air Force Base,
TX. Gsit la http://www.icodap.org/papers/SDI/
TUPES/ASD-TR-61-97-00.pdf
Van der Linden, D., De Nijenhuis, T., & Bakker, A.B.
(2010). The general factor of personality: A metaanalysis of Big Five intercorrelations and a criterion-related validity study. Journal of Research in
Personality,
44,
315327.
doi:
10.1016/j.jrp.2010.03.003
Acknowledgments
This research was supported by CNCSIS UEFISCSU, project number PNII IDEI 1076/2009.
This organization had no role in the design and implementation of the study.
54
4/17/2012
1:08 PM
Page 55
Anexa 1
IPIP-50
V prezentm mai jos propoziii care descriu comportamentul oamenilor. V rugm s folosii scala
de mai jos pentru a descrie gradul n care v caracterizeaz fiecare afirmaie. Descriei cum suntei n
general, acum, nu cum ai dori s devenii n viitor. Descriei ntr-un mod sincer: cum v vedei pe dumneavoastr, cum v vedei n relaiile cu persoanele cunoscute, de acelai sex, i de aproximativ aceai
vrst. Pentru a v putea descrie cu onestitate, rspunsurile voastre vor fi absolut confideniale.
V rog s citii fiecare afirmaie cu atenie i apoi marcai cu un x csua care corespunde opiunii
dumneavoastr.
Nr.
Itemi
Sunt sufletul petrecerii.
Sunt puin preocupat() de alii.
Sunt ntotdeauna pregtit ().
M stresez cu uurin.
Am un vocabular bogat.
Nu vorbesc mult.
mi plac oamenii.
mi las lucrurile peste tot.
Aproape tot timpul sunt relaxat ().
Dezacord
total
Dezacord
parial
Nici acord,
nici dezacord
Acord
parial
Acord
total
10 neleg greu ideile abstracte.

11 M simt confortabil n jurul oamenilor.
12 Jignesc oamenii.
13 Sunt atent () la detalii.
14 M ngrijorez.
15 Am o imaginaie vie.
16 Rmn n umbr.
17 neleg sentimentele altora.
18 nvlmesc lucrurile.
19 Rareori m simt trist.
20 Nu sunt preocupat() de ideile abstracte.
21 Iniiez conversaii.
22 Nu m intereseaz problemele altora.
23 Termin treaba repede.
24 Sunt perturbat() cu uurin.
25 Am idei excelente.
55
4/17/2012
1:08 PM
Page 56
Nr. Itemi
26 Am puine de spus.
27 Am o inim cald.
28 Uit des s pun lucrurile napoi la locul lor.
29 M supr cu uurin.
30 Nu am o imaginaie bogat.
31 Vorbesc cu o mulime de oameni diferii la petreceri.
32 Nu sunt chiar interesat() de ceilali.
33 mi place ordinea.
34 mi schimb dispoziia de multe ori.
35 Sunt rapid() n nelegerea lucrurilor.
36 Nu mi place s atrag atenia asupra mea.
37 mi fac timp pentru alii.
38 mi evit responsabilitile.
39 Am schimbri frecvente de dispoziie.
40 Folosesc cuvinte dificile.
41 Nu m deranjeaz s fiu n centrul ateniei.
42 Simt emoiile celorlali.
43 Urmez un orar.
44 M irit uor.
45 mi petrec timp reflectnd asupra lucrurilor.
46 n jurul oamenilor necunoscui sunt tcut.
47 Fac oamenii s se simt linitii.
48 Sunt exact() n munca pe care o fac.
49 Adesea m simt melancolic()
50 Sunt plin() de idei.
56
Dezacord
total
Dezacord
parial
Nici acord,
nici dezacord
Acord
parial
Acord
total
4/17/2012
1:08 PM
Page 57
To be or not to beidentified. Explorations of students (dis)identification

in a Romanian university
Smaranda Boro1, Petru Lucian Cureu2
Abstract
The paper explores the way in which contextual and dispositional factors impact on students development
of identification and disidentification. We investigate these relations in one cross-sectional and one longitudinal
study. The results indicate that need for identification moderates the impact of contextual variables upon disidentification and the transformation of ambivalent identification into disidentification. Based on these findings, the
proposed guidelines for building an effective strategy to foster students identification with their university follow two lines. The first one refers to the differential impact of policies on students, depending on their need for
identification and initial level of organizational identification. The second targets the manipulation in strategy
making of organizational level factors affecting identification, such as the incongruence of the organizations
identity and organizational prestige.
Keywords: ambivalent identification, organizational identity incongruence, need for identification
Rsum
Larticle explore la manire dont les facteurs contextuelle et personnel ont un impact sur le dveloppement
de lidentification et dsidentification des tudiantes. Nous explorons ces relations dans une tude longitudinale
et une enqute. Les rsultats montrent que le besoin didentification modrs limpact des variables contextuelles
sur dsidentification, tant bien que la transformation de lidentification ambivalente dans dsidentification. Sur la
base de ces constatations, nous proposons deux lignes directrices proposes pour la construction dune stratgie
efficace pour favoriser lidentification des tudiantes avec leur universit. La premire se rfre limpact diffrenci des politiques sur les lves, en fonction de leur besoin didentification et le niveau initial de lidentification organisationnelle. La seconde vise la manipulation dans la stratgie des facteurs au niveau organisationnel
affectant lidentification, tels que lincongruit de lidentit de lorganisation et le prestige de lorganisation.
Mots-cls: identification ambivalente, incongruence didentit organisationnelle, besoin didentification
Rezumat
Articolul exploreaz modul n care factorii contextuali i dispozionali afecteaz dezvoltarea identificrii i
dezidentificrii studenilor. Investigm aceste relaii ntr-un studiu longitudinal i unul transversal. Rezultatele
indic faptul c nevoia de identificare modereaz impactul variabilelor contextuale asupra dezidentificrii, precum i transformarea identificrii ambivalente n dezidentificare. Pe baza acestor rezultate propunem dou
direcii de luat n calcul n elaborarea unei strategii eficiente pentru stimularea identificrii studenilor cu universitatea lor. Prima se refer la impactul diferenial al politicilor asupra studenilor, n funcie de nevoia lor de iden-
Vlerick Leuven Gent Management School, Belgium.

Adres de corespondent: smaranda.boros@vlerick.com
2
Tilburg University, The Netherlands.
57
4/17/2012
1:08 PM
Page 58
To be or not to beidentified. Explorations of students (dis)identification in a Romanian university
tificare i de nivelul iniial de identificare organizaional. Al doilea vizeaz manipularea n cadrul strategiei a
factorilor care afecteaz nivelul de identificare, cum ar fi incongruena identitii organizaiei i prestigiul ei.
Cuvinte-cheie: identificare ambivalent, incongruena identitii organizaionale, nevoia de identificare.
Introduction
Budget cuts in higher education currently
affect universities world-wide. US proposes
budget cuts that would reduce support for higher education by $89 billion over 10 years (Johnson & de Vise, 2010), while the UK spending
review cuts 40% of the higher education budget,
from 7.1billion to 4.2billion by 2014
(Richardson, 2010). These cuts push all universities towards a business, self-sustaining model
of organizing. The endeavor to retain current
students and keep a steady inflow of prospective
students becomes more prominent (Lang, 2009;
Shulruf, Hattie & Trumen, 2008). In this race,
prestigious universities have a clear advantage.
They attract higher numbers of candidates, and
hence have a broader selection pool. Not only
can they retain more students (Jamelske, 2009),
but these students are also more engaged and
committed to their education, and put in a higher effort (Sung &Yang, 2009). This results in
better prepared professionals who will become
more successful, which reflects back on the
prestige of these universities and their competitiveness. The cornerstone process of this cycle is
students identification with their university.
Organizational identification is the perception of unity with or belonging to an organization (Ashforth & Mael, 1989), in this case, the
university. The process of identification consists
of a person defining themselves by the same
attributes that they believe define the organization (Dutton, Dukerich & Harquail, 1994).
Through organizational identification, organization members fulfil such psychological needs as
safety, affiliation, self-enhancement, and selfactualization, while the organization gets members who are more likely to act in ways congruent with organizational goals and needs (Mills,
Bettis, Miller & Nolan, 2005). In this case, identified students are more committed and find it
more difficult to drop their studies, they work
harder towards the collective goal (Cureu,
58
Janssen & Raab, 2012; Sung &Yang, 2009), and

get more involved in altruistic and helping
behaviors towards their peers (van Knippenberg
& Sleebos, 2006). Furthermore, identified students have a stronger desire to participate in
activities organized by the university and to
demonstrate their identifications by wearing distinctive insignia or logos (Mael & Ashforth,
1992). After graduation, they tend to be more
involved as alumni, both by financial (i.e., donations) and symbolic (i.e., positive advertising)
contributions (Mael & Ashforth, 1992).
However, the process of identification is
complex because neither the individual nor the
organization has a single identity or even consistency among identities (Mills et al, 2005). It
does not suffice for a university to be perceived
as a prestigious unit. For its members, the incongruent messages it sends with respect to who it
is and what it stands for (e.g., messages tailored
differently for various stakeholders, incongruence between formal policies and their implementation, etc) are also a capital element in their
identification. This incongruence in the universitys identity (Kreiner & Ashforth, 2004; Pratt,
1998) can trigger ambivalent identification or
even disidentification from its students.
Disidentification is a self-perception based
on (1) a cognitive separation between ones identity and ones perception of the identity of an
organization, and (2) a negative relational categorization of oneself and the organization (Elsbach
& Bhattacharya, 2001, pp. 393), while ambivalent identification is a dual state of both identification and disidentification with an organization.
Students who are disidentified with their
university will act in opposition to the values
and norms held by that unit. They will do anything to distance themselves behaviorally from
that organization. Clearly, students disidentification can be highly detrimental to universities
in the current situation.
The current study is an investigation in the
emergence of students disidentification with their
4/17/2012
1:08 PM
Page 59
universities. In our exploration, we focus on two

elements that universities can influence through
their policies and strategies: their perceived prestige and their organizational identity incongruence, and on the moderating role of students
intrinsic need for identification in this relation.
Context for the study

Higher education institutions in post-communist countries represent a fast-changing yet
underexplored terrain. The development of the
higher education system the past 18 years in
Romania is a typical example for Eastern European ex-communist countries (Teodorescu &
Andrei, 2009). The communist regime had prohibited the existence of private higher education
institutions. As a reaction to that, a large number
of private universities appeared after the fall of
the communist regime in 1989. Still, it is only in
recent years (namely 2006) that Romania has
started to develop a system that assesses the
quality of teaching in these institutions (the
ARACIS commission). Up until very recently
(i.e., 2011), no national classification for them
existed. In this context, research on higher education settings and dynamics has been more
than scant.
Furthermore, there was no real need until
recent years for research about students dynamics. Several contextual factors contributed to
this. First, students used to choose their university according to a desired profession and
depending on the region in which they lived.
These combined elements reduced the number
of universities to choose between. More importantly, the baby boom of the 1980s secured constantly high number of applications. Therefore,
universities did not need, until recently, to promote themselves: the number of applications
was steady anyhow. Second, since transfers to
other universities were rather rare, the need of
students retention was not a priority either.
Third, alumni didnt play a role in their university; in fact, the relation of students with their
university was broken after graduation.
However, all these conditions have undergone major shifts in the past years. Consequent-
ly, universities started to be proactive in their

policies of student attraction and retention.
Their current struggles in this respect resemble
now the ones of their West-European and American counterparts (Wraas & Solbakk,
2009).Therefore, the findings of this study are
relevant both for countries in which the business-model of universities is relatively recent,
as well as in countries where it has a long tradition, given the current economic constraints all
higher education units go through.
Literature
We started this paper by emphasizing the
positive impact of university prestige on student
identification. The first accounts on identification, coming from social identity theory, postulated that high status groups (i.e., prestigious
organizations), trigger higher identification
from their members (Tajfel, 1972). Being affiliated with a successful group provides individuals with an important way to establish and maintain a positive view of themselves. Therefore,
prestigious organizations trigger more identification and reduce the probability of disidentification in their members, since it is more difficult
not define yourself in a similar manner to a successful entity (Boro, Cureu & Miclea, 2011).
Nevertheless, this motivation alone fails to
explain why group members stick with their
group (i.e., remain identified) even through bad
times, when the groups negative image fails to
enhance their self-esteem (Ellemers, Spears
&Doosje, 1997).
Perceived organizational prestige is members perception of what outsiders think of the
organization. However, and more importantly,
members experience the organization first-hand.
They are aware of not only the organizations
intended image or how others perceive it, but
also of its identity. An organizations identity is
its shared answer to the question who we are
and what we stand for (Boro, 2009). It is the
structure of the central, enduring and distinctive
attributes that describe the organization (Albert
&Whetten, 1985). These attributes can form a
consistent, unitary image, and then the identity
59
4/17/2012
1:08 PM
Page 60
is defined as strong. Alternatively, an organization may send contradictory or mixed messages

to its stakeholders regarding what it stands for
and why it does so. In this case, the organization
has an incongruent identity. Organizational
identity incongruence can be caused either by
organizations having multiple and sometimes
conflicting identities (Pratt, 1998) or by a lack
of alignment between image (what we say we
stand for) and action (what our actions deem
important for us as an organization).
As a consequence of this incongruence,
members of the organization may have mixed
attachments towards the organization. In other
words, their identification becomes ambivalent.
Ambivalent identification is a dual state of both
identification and disidentification to an organization. It can take the form of identifying with
certain dimensions or traits of the organizations
identity or of simultaneous identification and
disidentification with the same traits (Kreiner &
Ashforth, 2004). The antecedents that can trigger ambivalent identification, are primarily the
incongruence of the organizations identity
(Kreiner & Ashforth, 2004) and the negative
image/prestige of the organization (Elsbach &
Bhattacharya, 2001). Based on the predictions
of cognitive consistency theories (Heider, 1958;
Festinger, 1957), ambivalent identification is
not a cognitive state one can endure for long
time, which means that one will try to find
strategies to come to terms with it. However,
this might be more difficult for students with a
propensity to identify with social groups or
organizations.
Although all individuals are at least somewhat receptive to identification (as a means of
fostering a sense of belonging and self), individuals might differ in their propensity to identify
with social objects (Glynn, 1998) a propensity termed need for organizational identification
(nOID). nOID is a desire to be imprinted upon
(Glynn, 1998:240) by an organization and
receptivity to socialization, and is negatively
associated with a desire for separateness from
the organization. nOID is an individual disposition that tends to generalize across various personal, situational or organizational contexts.
Individuals with a high need for organizational
identification present a marked tendency to pub60
licly identify themselves as organizational

members so as to extend the self to incorporate
the organization (Glynn, 1998). Unsurprisingly,
need for organizational identification is positively associated with identification and negatively associated with disidentification (Kreiner
& Ashforth, 2004). More importantly though,
people who differ in their need for identification
will react differently to similar situations
(Boro, Cureu & Miclea, 2011). Hence, need
for identification is a moderator of the external
antecedents of identification.
Therefore, our assumption is that ambivalent identification is a transient state for group
members only when it is associated with a low
need for identification. In this case, ambivalent
identification will in time transform into
disidentification. For people who preserve a
high need for identification, ambivalent identification will maintain over time, provided the
external conditions remain stable, but its consequences will alter. A number of recent studies
(Armstrong-Stassen, 2004; Doosje, Spears &
Ellemers, 2002; Worchel & Countant, 2001)
investigated how reactions to group status may
be moderated by initial levels of identification
and how the latter affects group perceptions and
cognitions. As Social Identity Theory predicts,
in the case of a high need for identification,
threat to the group will be met in terms of collective behavior responses and group affirmation strategies (Ellemers, Spears & Doosje,
2002). Hence, instead of disidentifying, these
group members might try to act in order to
improve the group status.
Building on these theoretical considerations, we propose three hypotheses. The first
one looks into the effect of organizational identity incongruence on ambivalent identification,
disidentification and identification, moderated
by need for organizational identification. In this
respect, we expect that:
Hypothesis 1a. Need for organizational
identification strengthens the positive impact of
organizational identity incongruence on
ambivalent identification.
Hypothesis 1b: Need for organizational
identification weakens the positive impact of
organizational identity incongruence on
disidentification.
4/17/2012
1:08 PM
Page 61
Hypothesis 1c: Need for organizational

identification strengthens the negative relation
between organizational identity incongruence
and identification.
Considering the same moderator, we then
look into the impact of perceived organizational
prestige on the three forms of identification and
predict that:
Hypothesis 2a: Need for organizational
identification strengthens the negative impact of
perceived organizational prestige on
ambivalent identification.
weakens the negative impact of perceived organizational prestige on disidentification.
Hypothesis 2c: Need for organizational
identification strengthens the positive impact of
perceived organizational prestige on organizational identification.
Finally, we explore the moderating role of
initial need for identification from a temporal
perspective, in order to explain contradictory
results arguing whether ambivalent identification
is a transient or permanent state (Kreiner & Ashforth, 2004). We argue that need for organizational identification moderates the development of
ambivalent identification in such a way that: people with ambivalent identification and high need
for identification will maintain their ambivalent
identification over time, whereas those with
ambivalent identification and low need for identification will become more disidentified in time.
Hence, we hypothesize that:
Hypothesis 3a: Need for organizational

identification strengthens the development of
ambivalent identification over time.
identification weakens the development of
ambivalent identification in disidentification
over time.
Method and findings

In order to test our hypotheses, we conducted a cross-sectional and a longitudinal investigation. The cross-sectional investigation was
carried out on a larger sample of students attending five different Schools of the university. Due
to the type of this design, in study 1 we could
only test hypotheses 1 and 2. Further on, to go
more in-depth with our investigation, and to be
able to study causality, we proceeded to a longitudinal study in one of the Schools. Study 2 tested the predictions of all three hypotheses. In the
next section, we report the procedure and the
results of each study.
Study 1. A cross-sectional
investigation
Respondents and procedure
344 students from four different Schools:
Psychology (117), Sociology (89), Geography
(69), and Economics (69) filled in a question-
Table 1. Overview of the measures used in the study

Name
Nr. of
items
Source
Response
scale
Sample item
Cronbach
1.Organization identity
incongruence
6 items
Kreiner&Ashforth, 5-point.Likert
2004
scale
My School sends mixed messages

concerning what it cares about
.85
2.Perceived
organizational prestige
8 items
Mael&Ashforth,
1992
5-point.Likert
scale
People in my community think highly

of this School
.71
3.Need for organizational

5 items
identification
2004
scale
Without an organization to belong to,

I would feel incomplete
.67
4.Organizational
identification
6 items
Mael&Ashforth,
1992
When I talk about the School I attend,

I usually say we rather than they
.81
5.Organizational
disidentification
6 items
2004
scale
I have tried to keep the School I

attend a secret from people I meet
.86
6.Ambivalent
identification
6 items
2004
scale
I find myself being both proud and

embarrassed to be a part of this
School
.75
5-point.Likert
scale
61
4/17/2012
1:08 PM
Page 62
Table 2. Descriptive statistics for the variables employed in study 1

Scale
SD
1.
2.
3.
4.
5.
6.
1.Organization identity incongruence
2.16
x.79x
xx1xx
-.317**
-.267**
-.191**
.440**
.500**
2.Perceived organizational prestige
3.88
.57
.267**
.245**
-.411**
-.400**
3.Need for organizational identification
3.47
.76
.508**
-.276**
-.155**
4.Organizational identification
3.25
.83
-.220**
-.059
5.Organizational disidentification
1.31
.56
.497**
6.Ambivalent identification
1.95
.86
* p<0.05, ** p<0.01
naire assessing organizational identification and

commitment. The respondents age ranged
between 19 and 49 (M=21.60, SD=2.62). 34 of
them were males, and 297 females. They were
chosen from all 4 study years: 33- 1styear, 1472nd, 62- 3rd and 102- 4th.
of the scales, samples of items, as well as the

alpha Cronbachs for each scale are reported in
Table 1. All measures used a 1 to 5 Likert scale,
in which 1 meant strong disagreement or very
low fit with the personal situation, and 5 strong
agreements or high fit with the personal situation.
Measures
Identification with the organization (in our
case, the School), as well as the antecedents of
identification were measured by adapting to university settings several corresponding scales from
the organizational studies literature. The sources
Results
Table 2 gives a concise overview of the
means, standard deviations and correlations
between all variables included in this study.
After inspecting these descriptive data and in
order to avoid the multicollinearity effect, we
Table 3. Regression analyses for hypotheses 1 and 2 (study 1).

Ambivalent identification
Organizational
disidentification
Organizational identification
Model 1
Model 2
Model 1
Model 2
Model 1
Model 2
Organizations identity
incongruence
.42***
.42***
.32***
.32***
-.03
-.02
Perceived organizational prestige
-.27***
-.27***
-.28***
-.28***
.10**
.11**
.03
.03
-.11**
-.10**
.47***
.47***
Need for identification

Interaction effect incongruenceneed for identification
-.00
-.14***
.04
Interaction effect prestige-need for

identification
-.02
.11**
.07
Overall F model
52.33***
31.27***
45.80***
33.25***
42.21***
25.78***
Adjusted R2
.31
.30
.28
.32
.27
.27
R2
.31
.00
.28
.04
.27
.00
* p<0.1, ** p<0.05, *** p<0.01

Note: all variables presented in these regressions were centered.
62
4/17/2012
1:08 PM
Page 63
centered all variables included in the study. The

centered variable is in fact the mean of the differences between each individual value taken by
this variable and its average value.
As reported in Table 3, only hypotheses 1 b
and 2 b received full support. Need for organizational identification (nOID) had a direct effect
on both identification and disidentification, but
not on ambivalent identification. However, it
only moderated the effect of organizational
identity incongruence on disidentification
(H1b). Organizational identity incongruence
had a direct positive relation with both disidentification and ambivalent identification.
Perceived organizational prestige had a
direct significant impact on all three dependent
variables. However, need for organizational
identification only moderated the impact of perceived organizational prestige on disidentification (H2b).
were females (11 males, 117 females). All 4

study years (the first year students are the only
Bologna generation in the sample) completed 2
waves of questionnaires, one academic year
apart. The final sample consisted of 52 subjects
(2 males, 50 females), aged 19 to 24 (m=20.46).
This was the sample considered for testing all
our hypotheses. Their distribution on study
years was: 32 freshmen (sophomores at wave 3),
13 sophomores (3rd year students at wave 3),
and 7 3rd year students (seniors by wave 3). For
each wave, the students were announced in
advance when they would have to complete the
questionnaire, and there were three occasions
for each wave when they could come and complete the questionnaire.
Results
The descriptive statistics for the data
obtained in study 2 are reported in Table 4. Once
again, and now considering the small final sample, we centered all variables included in the
study.
As reported in Tables 5 and 6, our data partially supported the proposed hypotheses. Need
for organizational identification at time 1
(nOID1) had a direct impact upon all three
dependent variables at time 2 (identification,
disidentification and ambivalent identification).
However, it only moderated the effect of organizational identity incongruence (time 1) on sub-
Study 2. A longitudinal investigation

Respondents and procedure
128 students in Psychology at BabeBolyai University were invited to participate in
a panel study, as a part of their Research Practice module. The students belonged to all 4
study years (24 freshmen, 48 sophomores, 14
3rd year, and 424th year students), they were
aged 18 to 28 (m=20.51), and most of them
Table 4. Descriptive statistics for the variables employed in study 2

Scale
SD
1.
2.
3.
4.
5.
6.
7.
1.Organization identity incongruence at time 1
1.76
x.60x
xx1xx
-.375**
-.460**
-.080
.397**
.479**
.360**
2.Perceived organizational prestige at time 1
4.18
.40
.194
.300*
-.092
-.251
-.144
3.Need for organizational identification at time 1
3.57
.57
.342*
-.354*
-.073
.100
4.Organizational identification at time 2
3.42
.84
-.249
-.068
.023
5.Organizational disidentification at time 2
1.14
.25
.269
.355**
6.Ambivalent identification at time 1
1.72
.74
.603**
7.Ambivalent identification at time 2
1.61
.80
* p<0.05, ** p<0.01
63
4/17/2012
1:08 PM
Page 64
Figure 1. Interaction plot for the impact of need for identification and organizations identity incongruence upon organizational disidentification in (a) study 1 (cross-sectional) and (b) study 2 (longitudinal)
Perceived organizational prestige (time 1)

had a direct significant impact on organizational identification (time 2). However, this effect
was not strengthened by nOID1 in any of the
cases (H2).
sequent disidentification (H1b). Organizational

identity incongruence at time 1 also had a significant direct positive impact on ambivalent
identification at time 2, but this relation was not
strengthened by need for organizational identification (nOID1), as predicted in hypothesis 1a.
Table 5. Regression analyses for hypotheses 1 and 2 (study 2).

Ambivalent identification
Organizational
disidentification
Organizational
identification
Model 1
Model 2
Model 1
Model 2
Model 1
Model 2
.50***
.44**
.32**
.09
.21
.27
Perceived organizational prestige
-.01
-.01
.07
.09
.30**
.30**
Need for identification
.33**
.29*
-.22
-.38***
.37**
.42***
Organizations identity incongruence
Interaction effect incongruence-need

for identification
-.11
-.40***
.09
Interaction effect prestige-need for

identification
.03
.10
-.07
Overall F model
4.48***
2.78**
3.79**
5.10***
4.09**
2.58**
Adjusted R2
.17
.15
.15
.28
.15
.13
R2
.21
.01
.20
.16
.20
.01
* p<0.1, ** p<0.05, *** p<0.01

64
4/17/2012
1:08 PM
Page 65
Figure 2. Interaction plot for the impact of need for identification and perceived organizational prestige upon organizational disidentification in (a) study 1 (cross-sectional) and (b) study 2 (longitudinal)
Furthermore, our data supported the

hypothesis that ambivalent identification is only a
transient state for people with low need for identification, and will in time transform into disidentification, whereas people with high need for
identification, unable to disidentify, will remain
in a state of ambivalent identification (H3).
As predicted, need for identification at time
1 had a positive impact on ambivalent identification at time 2 (=.21), and a negative impact
on disidentification at time 2 (=-.14). The

interaction effect of ambivalent identification 1
and need for identification 1 on ambivalent
identification 2 and disidentification 2 also had
opposite signs: =.27 for the former and =-.20
for the latter. The inspection of the slopes in Figures 3a and 3b confirm that need for organizational identification strengthens the development of ambivalent identification over time
(H3a), and at the same time it weakens the trans-
Table 6. Regression analyses for hypothesis 3 (study 2)

Ambivalent identification 2
Ambivalent identification 1
Need for identification 1
Model 1
Model 2
Model 1
Model 2
.66***
.62***
.08*
.05
.20
.21
-.14**
-.14**
Interaction effect
Overall F model
Organizational disidentification 2
.27
-.20***
15.27***
10.63***
5.54***
7.00***
Adjusted R2
.35
.36
.15
.26
R2
.38
.01
.18
* p<0.1, ** p<0.05, *** p<0.01

65
4/17/2012
1:08 PM
Page 66
Figure 3. Interaction plot for the impact of ambivalent identification and need for identification at time
1 upon (a) ambivalent organizational and (b) and disidentification at time 2
formation of ambivalent identification into

disidentification over time (H3b). In other
words, those who need to identify will carry on
an ambivalent identification, while those who
dont intrinsically need to, will simply start
disidentifying.
One final procedure investigated the impact
of initial levels of identification on subsequent
participation in the study. Our assumption here
was that students with lower levels of identification would slowly disappear from the study
(hence confirming the lack of engagement in
academic activities of non-identified students).
Therefore, we compared the levels of identification at time 1 between drop-outs and the sample
that continues (stayed in) the study. Betweengroup t-tests confirmed significant differences
in organizational identification (t=-1.71,
p=0.09), disidentification(t=2.28, p=0.02), and
ambivalent identification(t=2.33, p=0.02)
between drop-outs and remaining students:
drop-outs had slightly lower levels of identification (M=3.16, SD=.83), and significantly higher
levels of disidentification (M=1.26, SD=.64)
and ambivalent identification (M=1.97,
SD=.83) than remaining students (M=3.38,
SD=.72 for identification; M=1.09, SD=.26 for
disidentification, M=1.72, SD=.74 for ambivalent identification).
66
Hence, we received evidence in favor of

our assumption of identification forms (especially the negative ones) impacting on student
engagement (both in terms of participation to
activities and as mere class attendance).
Discussions and implications

Discussions
The present research set out to explore how
organizational and individual factors influence
the development of identification, ambivalent
identification and disidentification of students
with their Schools. Our results with respect to
positive identification are in line with previous
research on organizational identification. The
more prestigious the School, the more identified
its students are. Students propensity to identify
with various social entities enhances their positive identification, and acts in parallel with the
units prestige. The Schools identity incongruence does not seem to significantly impact on
students identification, regardless of their need
for identification. These results so far bring
nothing new to previous research (Kreiner &
Ashforth, 2004). The contribution of our
research is apparent when looking at other
4/17/2012
1:08 PM
Page 67
forms of identification (i.e., ambivalent identification and disidentification), which have been
largely ignored previously.
Organizational disidentification is a result
of the interaction effect between internal and
external variables. Although a lower prestige
and a higher incongruence have a direct immediate impact on disidentification (study 1), this
direct effect is no longer apparent in time (study
2). In time, disidentification is triggered by
lower prestige and higher incongruence only if
students have a less accentuated need for identification. In other words, need for organizational
identification is a really important moderating
factor when we look at how disidentification
develops in time. Considering the results for our
first two hypotheses, we can assert that external
(organizational) and internal (individual) variables have a different impact on the unfolding of
different types of identification. We argued that
external variables (such as organizations identity incongruence or organizational prestige),
being contextual in nature, have a more accentuated here and now impact, which allows organizational members to adapt flexibly to their environment. While these external variables have a
significant impact in the cross-sectional investigation, their effect is far less significant in the
longitudinal dataset. The individual-level variables (e.g., need for identification), on the other
hand, are those supporting consistency and stability in identification patterns, and their effect
is more evident in time. This pattern of results is
in accord with previous findings (Doosje, Ellemers & Spears, 1995; Ellemers & al., 1997) stating that people dont only identify when the
social groups (i.e., organizations) situation is
favorable; they remained identified even
through the bad times, because of a more powerful subjective factor (Ellemers & al, 1997). In
other words, the influence of external factors
upon the various dimensions of identification
proves indeed to be moderated by subjective
factors. Our data proves that this is even more
the case for ambivalent identification and
disidentification than for positive identification.
When comparing the data for hypothesis 1a
in study 1 and study 2, it becomes clear that
ambivalent identification is only momentarily
dependent upon organizational prestige (i.e., it
has a significant impact in the cross-sectional,

but not in the longitudinal analysis). What matters in time is a Schools organizational identity
incongruence. Identity incongruence directly
and immediately triggers ambivalent identification. Previous research has already pointed to
the fact that organizational identity incongruence brings about ambivalent identification
(Kreiner & Ashforth, 2004). Our results support
this assertion, but take it a step further. Ambivalent identification is a cognitively inconsistent
state, hence a disturbing one. Those students
who strongly need to identify with their organization can bear for longer the organizations
identity incongruence than those with a low
need for identification, and remain ambivalently
identified. For those with a lower propensity for
identification however, disidentification (hence
the rejection of the organization by defining
oneself through opposite attributes than the ones
used to define the organization) is the only
means powerful enough to solve this cognitive
inconsistency. It is important to notice here that
the moderating effect of need for organizational
identification is not so obvious in triggering
ambivalent identification (as it was for disidentification, for instance), but on its subsequent
development.
To summarize, our data showed that need
for identification is a moderator which influences the development of disidentification in
two ways: first, it filters the effect of external
variables on disidentification; second, it determines the development of ambivalent identification into disidentification. Our second important
finding regards external variables that impact
students identifications: while the Schools perceived prestige has a momentary impact on all
forms of identification, it is only relevant in time
for positive identification. However, it is important to remember that the impact of identity
incongruence remains visible in time and may
be the foundation of future disidentification.
These conclusions are important for strategy
makers in universities in their choice to focus on
foster positive identification and counteract
disidentification. Based on the findings of our
study, we draw in the following section some
outlines of applications for decision-makers in
universities.
67
4/17/2012
1:08 PM
Page 68
Applications and implications

Universities who want to have identified
students need to pay attention to several considerations: students traits (hence, individual level
factors) and external variables that can be influenced by strategy (at organizational level). The
individual level factors are definitely not the
focus of any policy or strategic plan. Yet, they
need to be taken into consideration whenever
implementing one, as important moderators of
the external factors, which are the ones manipulated through various interventions.
Students enter universities prone to be
imprinted upon: they are here not just to learn,
but also willing to develop a new identity. They
differ with respect to their need for identification, and are hence differently prompted to look
for cues that would increase their bond to the
university, or to react to these cues. One must
bear this in mind when assessing the efficacy of
a policy, in that its possible differential impact
might not be a flaw of the policy, but the natural
reaction of those targeted by it.
Bearing this in mind, the second and more
important aspect concerns the external factors
affecting identification: these are the ones that
any strategic policy-maker should focus on.
Although several researches have pointed to the
high impact of middle-level factors (i.e., grouplevel factors) on identification (Riketta & van
Dick, 2005; van Knippenberg & van Schie,
2000; van Dick, Wagner, Stellmacher & Christ,
2004), these factors are more difficult to control
in a university setting, given the large autonomy
(hence heterogeneity) of departments in general
and the organization of each department, and
further each course in particular. In these settings, the most relevant factors for policy makers are those pertaining to the organizational
level. The present research outlines some important aspects regarding two of these factors: organizations identity incongruence and organizational prestige. In terms of policy making, one
must know that the identity of the organization
strongly affects its members identification. This
identity needs first and foremost to be clear,
strong, coherent and unitary. In order to achieve
this, policies need to be articulated, so that they
wouldnt send contradictory messages. Even
more important, and often disregarded, their
68
implementation should be carefully monitored,

so that the actual form in which they reach their
target would keep the coherence and consistency of the universitys identity. One of the most
detrimental influences comes from a perceived
incongruence in the identity of the organization.
Not achieving either of the two demands of consistency above may lead to such a perceived
incongruence. This incongruence can be even
more harmful to member identification than the
prestige of the organization. The latter bears a
strong influence on identification. Yet, it is
incongruence that has the most negative effects.
This statement puts strategic policies into a new
light: not always can we obtain prestige, when
compared to other players in the field, but building and implementing a coherent and clear strategy, with the existent means, is a powerful tool,
which decision makers should not disregard.
References
Albert, S. &Whetten, D.A. (1985). Organizational identity. In L.L. Cummings and M.M. Staw (Eds).
Research in organizational behavior, vol. 7,
263295. Greenwich, CT: JAI Press.
Armstrong-Stassen, (2004). The influence of prior commitment on the reactions of layoff survivors to
organizational downsizing. Journal of Occupational Health Psychology, 9, 1, 46-60.
Ashforth, B. E., & Mael, F. (1989). Social identity theory and the organization. Academy of Management
Review,14, 2039.
Boro, S. (2009). Identity and image: The soul and face
of organizations. In S. Boro (Ed.) Exploring organizational dynamics, 644-654. London: Sage.
Boro, S., Cureu, P.L., &Miclea, M. (2011). Integrative
tests of a multidimensional model of organizational identification. Social Pychology, 42, 2.
Cureu, P.L., Janssen, S., &Raab, J. (2012). Connecting
the dots: Social network structure, conflict, and
Group Cognitive Complexity. Higher Education.
Downloaded from: http://www.springerlink.com/
content/l85015726726530u/fulltext.pdf
Doosje, B., Ellemers, N., & Spears, R. (1995). Perceived intragroup variability as a function of group
status and identification. Journal of Experimental
Social Psychology, 31, 410-436.
Doosje, B., Spears, R., & Ellemers, N. (2002). Social
identity as both cause and effect: The development
of group identification in response to anticipated
and actual changes in the intergroup status hierar-
4/17/2012
1:08 PM
Page 69
chy. British Journal of Social Psychology, 41,

57-76.
Dutton, J. E., Dukerich, J. M., & Harquail, C. V. (1994).
Organizational images and member identification.
Administrative Science Quarterly,39, 239263.
Ellemers, N. Spears, R. & Doosje, B. (2002). Self and
social identity. Annual Review of Psychology, 53,
161-186.
Ellemers, N., Spears, R., & Doosje, B. (1997). Sticking
together or falling apart: in-group identification as
a psychological determinant of group commitment
versus individual mobility. Journal of Personality
and Social Psychology, 72, 617-626.
Elsbach, K. D., & Bhattacharya, C. B. (2001). Defining
who you are by what youre not: organizational
disidentification and the National Rifle Association. Organization Science, 12, 393-413.
Festinger, L. (1957). A theory of cognitive dissonance.
Stanford, CA: Stanford University Press.
Glynn, M. A. (1998). Individuals need for organizational identification (nOID): speculations on individual differences in the propensity to identify. In D. A.
Whetten, and P. C. Godfrey (Eds.).Identity in
organizations: Building theory through conversations: 238-244. Thousand Oaks, CA: Sage.
Heider, F. (1958). The Psychology of Interpersonal
Relations. New York: Wiley.
Jamelske, E. (2009). Measuring the impact of a university first-year experience program on student GPA
and retention. Higher Education: The International Journal of Higher Education and Educational
Planning, 57, 3, 373-391.
Johnson, J. & de Vise, D (2010). Students protest cuts to
higher-education funds. Washington Post. Downloaded from http://www.washingtonpost.com/ wpdyn/content/article/2010/03/04/AR201003040130
7.html .
Kreiner, G. E., &Ashforth, B. E. (2004). Evidence
toward an expanded model of organizational identification. Journal of Organizational Behavior, 25,
1-27.
Lang, D.W. (2009). Articulation, transfer, and student
choice in a binary post-secondary system. Higher
Education: The International Journal of Higher
Education and Educational Planning, 59,
355-371.
Mael, F., &Ashforth, B. E. (1992). Alumni and their
alma mater: a partial test of the reformulated
model of organizational identification. Journal of
Organizational Behavior, 13, 103123.
Mills, M., Bettis, P., Miller, J. W., &Nolan, R. (2005).
Experiences of Academic Unit Reorganization:
Organizational Identity and Identification in Organizational Change. The Review of Higher Education, 28, 4, 597-619.
Pratt, M. G. (1998). To be or not to be? Central questions in organizational identification. In D. A.
Whetten, and P. C. Godfrey (Eds.), Identity in

organizations: Building theory through conversations: 171-207. Thousand Oaks, CA: Sage.
Richardson, H. (2010). Spending Review: Universities
to face 4.2bn cut. BBC News. Downloaded from
http://www.bbc.co.uk/news/education-11550619 .
Riketta, M.& Van Dick, R. (2005). Foci of attachment
in organizations: A meta-analytic comparison of
the strength and correlates of workgroup versus
organizational identification and commitment.
Journal of Vocational Behavior, 67, 490-510.
Shulruf, B., Hattie, J., &Tumen, S. (2008). Individual
and school factors affecting students participation
and success in higher education. Higher Education: The International Journal of Higher Education and Educational Planning, 56, 613-632.
Sung, M., & Yang, S-U. (2009). Studentuniversity
relationships and reputation: a study of the links
between key factors fostering students supportive
behavioral intentions towards their university.
Higher Education: The International Journal of
Higher Education and Educational Planning, 57,
6, 787-811.
Tajfel, H. (1972). La categorisation sociale. In S.
Moscovici (Ed.) Introductiona lapsychologie
sociale, 272-302. Paris: Larousse.
Teodorescu, D., & Andrei, T. (2009). Faculty and peer
influences on academic integrity: college cheating
in Romania. Higher Education: The International
Journal of Higher Education and Educational
Planning, 57, 3, 267-282.
van Dick, R., Wagner, U., Stellmacher, J., & Christ, O.
(2004). The utility of a broader conceptualization
of organizational identification: Which aspects
really matter? Journal of Occupational and Organizational Psychology, 77, 171-191.
van Knippenberg, D., & van Schie, E. C. M. (2000).
Foci and correlates of organizational identification. Journal of Occupational and Organizational
van Knippenberg, D., & Sleebos, E. (2006). Organizational identification versus organizational commitment: self definition, social exchange, and job attitudes. Journal of Organizational Behavior, 27,
571-584.
Wraas, A., & Solbakk, M.N. (2009). Defining the
essence of a university: lessons from higher education branding. Higher Education: The International Journal of Higher Education and Educational
Planning, 57, 449-462.
Worchel, S. and Countant, D. (2001). It takes two to
tango: Relating group identity to individual identity within the framework of group development. In
Hogg and Tindale (Eds.) Blackwell Handbook of
Social Psychology: Group Processes. Massachusetts: Blackwell Publishers.
69
4/17/2012
1:08 PM
Page 70
nvarea n echip i eficacitatea echipelor de munc:

Sinteza analizei studiilor publicate
Claudia L. Rus1, Adriana Bban2, Saul N. de Jesus3
Abstract
This study analyzes the research that have examined team learning in relation to work team effectiveness in
terms of sample, research design, methods of data gathering, and theoretical aspects investigated. This analysis
was conducted to reveal potential moderators of the relationship between these two concepts. The results reveal
that in the analyzed studies the two concepts were preponderantly examined with project, services and multiple
teams and self-report methods. All studies used a cross-sectional design. Most frequently, team learning was studied as a process in terms of learning behaviors and examined in relation to perceived performance as a criterion
of work team effectiveness. The studies that have analyzed team learning as a one-dimensional concept suggest
its positive relationship with work team effectiveness but those that have measured it as a multidimensional concept provide inconsistent results for this relation.
Keywords: team learning, work team effectiveness, synthesis
Rsum
Cette tude analyse les recherches qui ont examin lapprentissage en quipe par rapport leffectivit de
lquipe de travail en termes dchantillon, la conception de recherche, les mthodes pour collecter les donnes
et les aspects thoriques tudis. Les rsultats rvlent que la relation de ces deux concepts a t prpondrante
examine avec des quipes de projet, des services et multiples et des mthodes dauto-valuation pour collecter
les donnes. Toutes les tudes ont utilis une conception de recherche transversale. Le plus souvent, lapprentissage en quipe a t examin comme un processus en termes de comportements dapprentissage et examin par
rapport la performance perue comme un critre defficacit de lquipe de travail. Les tudes qui ont analys
lapprentissage de lquipe comme un concept unidimensionnel suggraient sa relation positive avec leffectivit
de lquipe du travail, mais ceux qui lont mesure multidimensionnel fournirent des rsultats incohrents pour
cette relation.
Mots cls: lapprentissage de lquipe, leffectivit de lquipe du travail, la synthse
Rezumat
Acest studiu analizeaz cercetrile care au examinat nvarea echipei n relaie cu efectivitatea echipei de
munc sub aspectul eantionului, designului de cercetare, metodei de colectare a datelor i aspectelor teoretice
investigate. Aceast analiz a fost realizat cu scopul de a identifica poteniali moderatori ai relaiei dintre cele
dou concepte. Rezultatele arat c n studiile analizate, aceste dou concepte au fost examinate preponderent cu
echipe de proiect, servicii i multiple metode auto-declarative de colectate a datelor. Toate studiile au utilizat un
design transversal. Cel mai frecvent, nvarea echipei a fost studiat ca proces, cu accent pe comportamentele de
Universitatea Babe-Bolyai.
Adres de coresponden: claudiarus@psychology.ro
2
Universitatea Babe-Bolyai.
3
University of Algarve.
70
4/17/2012
1:08 PM
Page 71
nvarea n echip i eficacitatea echipelor de munc: Sinteza analizei studiilor publicate
nvare, i examinat n relaie cu performana perceput ca i criteriu al efectivitii echipei de munc. Studiile
care au analizat nvarea echipei ca un concept unidimensional sugereaz existena unei relaii pozitive cu efectivitatea echipei iar cele care l-au msurat ca un concept multidimensional ofer rezultate inconsistente privind
aceast relaie.
Cuvinte cheie: nvarea n echip, efectivitatea echipei de munc, sintez
Introducere
Natura dinamic a mediului n care opereaz
echipele de munc genereaz nevoia ca o echip
s nvee continuu (Zaccaro, Ely, & Shuffler,
2008). Savelsbergh, van der Heijden i Poell
(2010), evalund atitudinile membrilor unor
echipe de munc i ale liderilor acestora fa de
factorii pe care i consider importani pentru performana echipei, au evideniat un acord puternic
ntre cele dou categorii de participani privind
importana comportamentelor de nvare ale
echipei ca indicator al nvrii n echip.
n literatura de specialitate, pot fi regsite
multiple definiii ale conceptului de nvare n
echip (London & Sessa, 2007). ns n ciuda
volumului de definiii, exist un nivel considerabil de ambiguitatea n definirea acestui concept
(Wilson, Goodman, & Cronin, 2007; Zaccaro et
al., 2008). Decuyper, Dochy i Van den Bossche
(2010) au identificat n literatura de specialitate
aproximativ 30 de definiii i descrieri. Unele
dintre aceste definiii pun accent pe procesul de
nvare n echip (Edmondson, 1999, Gibson &
Vermeulen, 2003), pe nvare ca rezultat (Ellis
et al., 2003), iar altele conceptualizeaz
nvarea n echip att n termeni de proces, ct
i de rezultat (Argote, Gruengeld, & Naquin,
2001). Definiiile nvrii ca proces adesea
cuprind aspecte precum reflecia i aciunea
(Edmondson, 1999; Gibson & Vermeulen,
2003), mprtirea i procesarea cunotinelor,
realizarea mbuntirii rezultatelor (Edmonson,
2002; Gibson, 2001; Kolb, 1984; apud. Jehn &
Rupert, 2008). n cadrul acestei perspective procesuale, nvarea a fost conceptualizat ca o
serie de activiti (e.g., achiziia informaiei; van
Woerkom, & Croon, 2009) sau de comportamente de nvare (e.g., discutarea, detectarea i
corectarea erorilor; Edmondson, 1999). Pe de
alt parte, n definiiile care vizeaz nvarea ca
rezultat, acest concept este descris n termeni de
schimbri la nivelul cunotinelor membrilor

echipei care rezult din interaciunile acestora
(Argote et al., 2001; Ellis et al., 2003).
Creterea ponderii utilizrii echipelor de
munc n cadrul organizaiilor moderne contribuie i la intensificarea preocuprilor privind
mbuntirea efectivitii acestora (Singh &
Muncherji, 2007). Conceptul de eficacitate a
echipei de munc este diferit de cel de eficiena
(msura n care echipa realizeaz sarcinile care
i revin cu o utilizare optim a resurselor
disponibile) sau auto-eficacitatea colectiv a
echipei de munc (convingerea membrilor
echipei c aceasta poate fi performant n viitor)
(Cureu, 2007). n ncercarea de a schia ceea ce
constituie eficacitatea echipei, Pia, Martnez i
Martnez (2008) disting pe baza literaturii de
specialitate dou categorii de modele. Prima
categorie definete eficacitatea echipei ca un
concept unidimensional i utilizeaz msurtori
obiective ale performanei echipei sau ale productivitii. Cea de-a doua categorie de modele
consider acest concept ca fiind unul complex,
multidimensional. Din aceast categorie, cel
mai cunoscut model este cel propus de Hackman (1987; apud. Singh & Muncherji, 2007)
care conceptualizeaz eficacitatea echipelor de
munc prin trei criterii:
1. performana sau rezultatele echipei (e.g.,
rezultate obiective ale echipei i evalurile
subiective ale acestora) reflect msura n care
un grup i realizeaz scopurile i gradul n care
rezultatul satisface misiunea pentru care a fost
alctuit.
2. viabilitatea perceput a echipei de a continua s lucreze mpreun n viitor. Aceast variabil a fost descris de unii autori ca fiind sentimentul de apartenen la grup (similar coeziunii
sociale) (Mathieu et al., 2008), n timp ce alii o
operaionalizeaz prin satisfacia membrilor, climatul sau atmosfera, angajamentul i coeziunea
echipei (Balkundi & Harrison, 2006). Din acest
71
4/17/2012
1:08 PM
Page 72
motiv, viabilitatea echipei de munc a devenit

un termen generic pentru o varietate de constructe diferite (Mathieu et al., 2008).
3. satisfacia membrilor echipei de munc
(dac membrii echipei au reacionat negativ la
aceasta experien, atunci costurile realizrii cu
succes a sarcinii sunt probabil mult mai mari
dect satisfacia trit) (Jordan, Feild, & Armenakis, 2002; Tekleab, Quigley, & Tesluk, 2009).
n prezent, pe lng aceste criterii, se
acord o atenie crescut inovaiei echipei ca
indicator al eficacitii echipei, fie c este abordat ca proces sau rezultat (Gil, Alcover, &
Peir, 2005; Mathieu et al., 2008).
Cu toate c studii anterioare au evideniat
relaia dintre nvarea n echip i competitivitate, legtura dintre nvarea n echip i performana echipei a fost examinat ntr-un numr
redus de studii (Chan, Pearson, & Entrekin,
2003) i avnd la baz perspective teoretice
diferite (Mo & Xie, 2009). Unele dintre aceste
studii au evideniat natura pozitiv a asocierii
dintre aceste dou concepte (Edmonson, 1999).
ns, alte studii au concluzionat c nvarea n
echip are un efect negativ asupra performanei
echipei (Lewis, 2003; apud. Mo & Xie, 2009).
Astfel, literatura de specialitate ofer rezultate
contradictorii privind relaia nvrii n echip i
eficacitatea echipei de munc. n ncercarea de a
explica aceste rezultate contradictorii, unii autori
au sugerat faptul c dei, nvarea n echip este
definit ca un proces multidimensional, de cele
mai multe ori acest concept este msurat ca un
proces unidimensional (Savelsbergh, van der
Heijden, & Poell, 2009). Cu toate c exist analize privind nvarea n echip, acestea s-au
focalizat mai mult pe organizarea diferitelor
aspecte ale literaturii de specialitate i ale rezultatelor obinute (Edmondson, Dillon, & Roloff,
2007) sau pe aspecte metodologice ale unor
studii selective privind nvarea n echip din
perspectiva unei singure definiii oferite acestui
concept (Goodman & Dabbish, 2011). n cadrul
acestor studii, nu s-a insistat ns asupra analizei
modului n care a fost studiat nvarea echipei
n relaie cu eficacitatea echipei de munc care ar
putea avea implicaii asupra relaiei dintre aceste
dou concepte, cu toate c a fost sugerat faptul
c nvarea n echipa de munc este un factor
72
important al eficacitii acesteia (Decuyper,

Dochy, & Van den Bossche, 2011).
n contextul informaiilor prezentate anterior, realizarea unei analize a studiilor empirice
care au examinat nvarea n echip n relaie
cu eficacitatea echipei de munc ar permite evidenierea unor factori de ordin teoretic i
metodologic care ar putea avea un rol explicativ
pentru rezultate inconsistente privind relaia
dintre aceste dou concepte. Astfel, n studiul de
fa ne propunem:
1. Identificarea metodologiei utilizate n
studiul nvrii n echip n raport cu eficacitatea
echipei de munc sub aspectul eantionului
(mrimea eantionului, tipul de echip, contextul
organizaional din care au provenit echipele de
munc), al tipului de design de cercetare, al
nivelului de raportare a rezultatelor (individual, de
grup), al metodei i al sursei de colectare a datelor.
2. Identificarea aspectelor nvrii n echip
i ale eficacitii echipei de munc care au fost
examinate n cadrul studiilor empirice sub aspectul perspectivei teoretice adoptate i a naturii
multidimensionale a constructului studiat.
Metod
Selectarea studiilor
Identificarea studiilor relevante pentru
aceast analiz a fost realizat printr-o cutare
computerizat n bazele de date internaionale
reprezentate de Web of Science, PsychArticles
(EbscoHost), PsychInfo (EbscoHost) i Psychology and Behavioral Sciences Collection
(EbscoHost). Cuvintele dup care s-a fcut
cutarea au fost: team learning and workteam
effectiveness, team learning and work team performance, grouplearning and workgroup effectiveness, group learning and workgroup performance. Perioada de cutare a studiilor a fost cea
cuprins ntre data de nceput permis de baza
de date (1899 ISI Web of KnowledgeSM, 1894
PsychArticles, 1800 PsychInfo, 1965 PsychologyandBehavioralSciencesCollection) i
31.07.2010. n total, au fost generate 3439 de
lucrri dintre care au fost selectate doar cele
publicate n limba englez.
4/17/2012
1:08 PM
Page 73
Pentru a fi incluse n analiz, studiile trebuiau: (a). s abordeze nvarea n echip n

relaie cu efectivitatea echipei utiliznd echipe
de angajai n mediul organizaional; (b). s
abordeze dintr-o perspectiv empiric cantitativ nvarea n echip n relaie cu efectivitatea
echipei de munc; (c). s abordeze echipa sau
grupul ca unitate de analiz; (d). s precizeze
nivelul de raportare al rezultatelor (individual
sau de grup); (e). s examineze cel puin dou
echipe de munc sau s ofere dou seturi de
msurtori.
Numrul final de studii incluse n analiz
este de 21. Aceste studii sunt marcate cu un * n
seciunea dedicat referinelor bibliografice.
Unul dintre studiile analizate a cuprins dou
eantioane diferite (Stalmeijer, Gijselaers,
Wolfhagen, Harendza & Scherpbier, 2007),
numrul eantioanelor independente incluse n
analiz fiind de 22.
Codarea studiilor selectate
Codarea studiilor care au ndeplinit criteriile anterior menionate a fost realizat independent de ctre doi cercettori. n cazul fiecrui
studiu au fost colectate informaii referitoare la:
(a) Eantionul utilizat: mrimea eantionului, tipul echipelor de munc (producie, servicii, management, proiect, aciune i performan, consultan, multiple schema de codare
utilizat de Sundstrom, McIntyre, Halfhill i
Richards n 2000), contextul organizaional din
care au provenit echipele de munc
(b) Tipul designului de cercetare: schema
de codare elaborat dup sistemul de clasificare
a metodelor de cercetare n psihologie propus de
Montero i Leon n 2007)
(c) Nivel de raportare al rezultatelor: grup
vs. individual
(d) Perspectiva teoretic din care a fost
definit nvarea echipei: proces (interaciune
ntre membrii grupului), rezultat (modificare
produs la nivel cognitiv, afectiv i comportamental) vs. alt categorie (proces i rezultat)
schema de codare elaborat dup Edmondson,
Dillon i Roloff (2008)
(e) Aspectul teoretic studiat al nvrii
echipei de munc n cadrul fiecrei perspective
teoretice: comportament de nvare (aciunile
prin care membrii grupului obin i proceseaz

date care le permit s se adapteze i s i
mbunteasc performana), activitate de
nvare (activiti prin care membrii grupului
obin i proceseaz date care le permit s se
adapteze i s i mbunteasc activitatea) vs.
alt categorie schema de codare elaborat dup
Edmondson i colaboratorii si (2008); rezultat
(schimbare) la nivel cognitiv, comportamental,
afectiv al echipei vs. alt categorie schem de
codare elaborat dup Kraiger, Ford i Salas
(1993);
(f) Aspectul teoretic studiat al eficacitii
echipei de munc: performan, satisfacia
membrilor fa de echip, viabilitatea echipei,
eficien, inovaie
(g) Natura multidimensional a constructului msurat: nvarea n echip considerarea
constructului msurat ca fiind unul global vs.
multidimensional (multi-faetat); eficacitatea
echipei de munc msurarea unui criteriu singular vs. criterii multiple de eficacitate
(h) Metoda de colectare a datelor:
observaional, psihofiziologic, psihofizic,
psihometric, scale i chestionare, intervievativ, jurnal i metoda narativ, focus-grup,
etnografic, cercetarea aciunii, de analiz a
arhivelor vs. mixt schem de codare elaborat
dup Breakwell, Hammond, Fife-Schaw i
Smith (2006), Brewerton i Millward (2001);
(i) Sursa de colectare a datelor: membrii
grupului, manageri, clieni, ali evaluatori
externi, mixt, nregistrri organizaionale,
schem de codare elaborat dup Sundstrom i
colaboratorii si (2000);
(j) Valoarea i forma coeficientului de asociere dintre nvarea i eficacitatea echipei de
munc.
Rezultate
Eantionul utilizat
Datele au evideniat faptul c mrimea
eantionului utilizat n studiile incluse n analiz
variaz de la 6 pn la 224 de echipe de munc.
n total au fost utilizate 1445, media fiind de
65.68 echipe per studiu (Tabel 1).
73
4/17/2012
1:08 PM
Page 74
Tabel 1. Caracteristici ale studiilor care au examinat nvarea echipei n relaie cu eficacitatea echipei
de munc
Nivel
Similaritate
raportare
Numr i tip echipe Context organizaional rezultate Metod
Surs
Eantion
Referina
Drach-Zahavy &
32 echipe de aciune Spitale
Pud (2010)
Van der Vegt,

de Jong,
Bunderson, &
Molleman
(2010)
46 echipe multiple
Bang, Fulesang,
56 de agende ale 8
Ovesen, &
echipe executive
Eilertsen (2010)
van Woerkom &

van Engen
84 echipe multiple
(2009)
Savelsbergh,
van der Heijden, 19 echipe de servicii
& Poell (2009)
van Woerkom &

88 echipe multiple
Croon (2009)
Huang, Chu, &

Jiang (2008)
60 echipe de proiect
Organizaii multiple
Organizaii multiple
Organizaii multiple
Banc
Organizaii multiple
Institut de cercetare
tehnologic
Somech&Drach224 echipe de servicii coli (educaie)

Zahavy (2007)
Tucker,
Nembhard, &
Edmondson
(2007)
74
Spitale
Grup
Grup
Grup
Grup
Individual
Grup
Nu
Da
Da
Da
Da
Da
Nu
Nu
Nu
Da
Da
Nu
Proces/
rezultat
Definiia nvrii echipei

de munc
Proces
nvarea echipei este ncorporat

ntr-un proces ciclic care implic
trei pai: colectarea datelor,
analiza datelor i formularea
concluziilor i implementarea
schimbrii (p. 795)
Proces
activitile prin care membrii

echipei caut s achiziioneze,
mprteasc, redefineasc sau
s combine cunoaterea relevant
sarcinii prin interaciunea unuia cu
cellalt. Comportamentele de
nvare ale echipei sunt o clas
specific de procese de
interaciune n echip (p. 348)
Proces
un proces emergent de reflecie i

aciune, caracterizat prin
adresarea ntrebrilor, cutarea
feedbackului, experimentare,
reflecia aspra rezultatelor i
discutarea erorilor sau rezultatelor
neateptate ale aciunilor (p. 353)
Proces
n definiia lui Huber (1991) i van

Offenbeek (2001) nvarea
echipei include procesele de
achiziie a informaiei, distribuirea
informaiei, interpretarea
informaiei i stocarea i
reactualizarea informaiei (p. 383)
Proces
proces de reflecie i aciune

colectiv caracterizat prin: (a)
explorare, (b) reflecie, (c)
discutarea erorilor i a rezultatelor
neateptate ale aciunilor, (d)
cutarea feedbackului, i (e)
experimentarea n i ca o echip
Proces
... Dup Argyris i Schn (1996) i

Edmondson (1999), ne focalizm
pe activitile de nvare realizate
de ctre membrii echipei prin care
echipa obine i proceseaz data
care i permit s se adapteze i s
se mbunteasc i prin care pot
fi obinute rezultate cum ar fi o
performan mai bun
Grup
Da
Da
Proces
Edmondson a definit nvarea

echipei ca un proces i ncearc
s articuleze comportamentul care
are ca rezultat adaptarea la
schimbare, o mai bun nelegere
sau performana mbuntit n
echipe (p. 1424)
Grup
Nu
Nu
Proces
un proces de interaciune, un
proces de echip cognitiv (p. 306)
Proces
Cnd echipele de proiect reflect

asupra aciunilor lor- incluznd
greelile pe care le-ar fi putut facei fac schimbri pentru a
mbuntii performana viitoare,
ele nva (Reagans i colab.,
2005) (p. 896)
Grup
Da
Da
4/17/2012
1:08 PM
Page 75
6 echipe de servicii
coal (educaie)
Grup
Da
Da
Proces
15 echipe de servicii
coal (educaie)
Grup
Da
Nu
Proces
Stalmeijer et al.
(2007)
Akgn, Lynn, &

165 echipe de proiect Organizaii multiple
Yilmaz (2006)
Yeh & Chou

(2005)
Organizaii multiple
Grup
Grup
Da
Da
Da
Da
Echipele pot s i ndeplineasc

pe deplin potenialul dac reuesc
valorificarea cunotinelor,
expertizei i experienei membrilor
lor pentru a mbogi performana
echipei. Acest proces a fost numit
ca fiind comportament de
nvare...
Proces
nvarea din perspectiva sociocognitiv ca un proces prin care

grupurile percep, interpreteaz,
stocheaz, reactualizeaz,
transmit i utilizeaz informaia
(Akgn i colab., 2003; Huber,
1991; Kiesler&Sproull, 1982)
(p.213). Sub-procesele procesrii
informaiei sunt: achiziia informaiei/cunoaterii, implementarea
informaiei/cunoaterii, diseminarea informaiei/ cunoaterii,
memoria, ne-nvrea, gndirea,
inteligena, improvizaia, crearea
semnificaiei (p. 212).
Proces
nvarea este abordat ca un

proces, prin care pot fi atinse
rezultate precum adaptarea la
schimbare, o mai bun nelegere
sau o performan mbuntit
(Edmonson, p. 353)
Van der Vegt &

Bunderson
(2005)
Organizaii din industria

petrolului i gazului
Grup
Da
Da
Proces
Din acest motiv considerm

comportamentul de nvare al
echipei ca un aspect al procesului
de interaciune al grupului
(Hackman& Morris, 1975) sau ca
un exemplu a procesului aciunii
grupului (Marks, Mathieu,
&Zacarro, 2001). (p. 534)
Edmondson
(2003)
Spitale
Grup
Nu
Nu
Proces
Proces
activitile din cadrul echipei care

implic adresarea ntrebrilor,
cutarea feedbackului unii de la
ceilali i discutarea problemelor
sau erorilor pot de-asemenea s
faciliteze performana echipei,
deoarece ele faciliteaz
dezvoltarea i rafinarea produsului
sau procesului echipei (p. 11)
Da
Proces
un proces emergent de reflecie i

aciune, caracterizat prin
adresarea ntrebrilor, cutarea
feedbackului, experimentare,
reflecia aspra rezultatelor i
discutarea erorilor sau rezultatelor
neateptate ale aciunilor (p. 353)
Da
Msura n care productorul/ prin

munca cu partenerul- a c?tigat
noi insigh-uri privind dezvoltarea
noilor produse, sarcinile cheie
implicate n procesul de producie,
Rezultat procesele noi de producie,
nelegerea nevoilor utilizatorului
final al produsului, precum ?i
practica personal, cuno?tine
procedurale ?i mecherii ale
pieei. (p. 486).
Edmondson
(1999a)
Edmondson
(1999b)
16 echipe aciune
51 echipe de
producie
51 echipe de
producie
Bstieler &
Hemmert (2010)
Organizaie de
producie de mobilier
Organizaie de
producie de mobilier

utilajelor
Grup
Grup
Grup
Da
Da
Da
Da
75
4/17/2012
1:08 PM
Page 76
Zellmer-Bruhn &
115 echipe multiple
Gibson (2006)
Akgn, Byrne,
Keskin, &Lynn
(2006)
Organizaii multiple din

industria farmaceutic i
produse medicale
Organizaii multiple
Grup
Da
Grup
Da
Nu
Achiziionarea, combinarea,
crearea colectiv i mprtirea
cunoaterii de ctre echipe
Rezultat (Argote, 1999) (p. 501), cu accent
pe nvarea actual realizat (de
exemplu, msura n care au fost
dezvoltate noi practici)
Da
Perspectiva comportamental a
nvrii echipelor de dezvoltare de
noi produse se refer la rezultatele
pe care le pot avea procesele.
Dintr-o perspectiv comportamental, nvarea echipei este
considerat ca ncorpornd leciile
Rezultat
nvate n timpul proiectului i ca
o corectare a problemelor ntlnite,
operaionalizat ca implementarea
sau utilizarea informaiei/cunoaterii de ctre membrii echipei
Moorman (79) i Lyn i colab. (72)
(p. 98)
Akgn, Byrne,
Keskin, Lynn, &
Imamoglu
(2005)
Organizaii multiple
Grup
Da
Da
Indicator al performanei echipei,

reflectnd nivelul de cunoatere
Rezultat pe care echipa l-a dobndit n
realizarea unei aciuni de elaborare a unui produs nou (p. 1109)
Sarin &
McDermott
(2003)

high-tech
Grup
Da
Da
Rezultat
Sub aspectul tipului de echipe din cadrul

acestor eantioane, nvarea n echip a fost
examinat n relaie cu eficacitatea echipei de
munc preponderent prin intermediul eantioanelor formate din echipe de proiect (40.90%),
o schimbare n comportament ca
rezultat al experienei (p.717)
servicii (18.18%) i echipe multiple (18.18%).

Aceste dou concepte au fost studiate mai puin
n cadrul eantioanelor formate din echipe de
aciune i performan (9.09%), management
(4.54%) i producie (9.09%) (Figura 1).
Figura 1. Distribuia studiilor incluse n analiz n funcie de tipul de echipe utilizate n cadrul acestora
Not:
76
1 = Echipe de proiect;
2 = Echipe de servicii;
3 = Echipe de performan i aciune;
4 = Echipe de management;
5 = Echipe de producie;
6 = Echipe de consultan;
7 = Echipe multiple (dou sau mai multe tipuri de echipe).
4/17/2012
1:08 PM
Page 77
Figura 2. Distribuia studiilor incluse n analiz n funcie de contextul organizaional din care au
provenit eantioanele utilizate n cadrul acestora.
Not:
1 = coli;
2 = Spitale;
3 = Bnci;
4 = Cercetarea produselor tehnologice;
5 = Industria utilajelor;
Contextul organizaional din care au provenit eantioanele incluse n aceast analiz este
variat. De cele mai multe ori, nvarea echipei
de munc a fost examinat n relaie cu eficacitatea acesteia n cadrul unor eantioane formate
din echipe de munc provenite din acelai
domeniu de activitate (63.64%). Restul eantioanelor a fost alctuit din echipe de munc
provenite din organizaii care activau n diverse
domenii de activitate (36.36%) (Figura 2).
Tipul designului de cercetare
Datele au evideniat c toate studiile analizate au avut la baz un design de cercetare de tip
corelaional, n care datele pentru ambele variabile de interes au fost colectate simultan.
Nivel de raportare al rezultatelor
Nivelul de raportare al rezultatelor obinute
n studiile analizate este cu preponderen la
nivel de grup (95.45%). Doar n cazul unuia dintre cele 22 de eantioane independente incluse
n analiz, rezultatele au fost raportate la nivel
individual (4.54%).
Metoda i sursa de colectare a datelor
n studiile analizate, toate aspectele nvrii
n echip au fost msurate prin intermediul metodelor auto-declarative (Figura 3). Astfel, 90.91%
6 = Producia de mobilier;
7 = High-tech;
8 = Industria gazului i petrolului;
9 = Industria farmaceutic i a produselor medicale;
10 = Diverse domenii de activitate.
dintre studii au utilizat metode de colectare a

datelor de tipul scalelor Likert. n cadrul acestei
categorii de studii, de cele mai multe ori, datele
au fost colectate de la membrii echipei (68.18%)
i mai puin frecvent de la managerii/supervizorii
echipei (18.18%) sau din evaluri realizate att de
membrii echipei, ct i de managerul acesteia
(4.54%). Restul studiilor (9.09%) au utilizat date
despre nvarea echipei provenite prin intermediul interviurilor aplicate supervizorilor sau
membrilor echipelor de munc.
De asemenea, rezultatele analizei arat c
datele despre eficacitatea echipei de munc au
fost colectate de cele mai multe ori prin intermediul scalelor de tip Likert. Aceste scale au fost
completate de ctre membrii echipei (40.90%),
membrii echipei i supervizorii acesteia
(13.63%), membrii echipei i un evaluator din
afara acesteia (4.54%), doar de supervizorii echipei (18.18%) sau alte persoane din afara grupului,
fie experii (4.54%) sau observatorii (4.54%).
Restul studiilor au utilizat date despre acest concept colectate prin intermediul metodei observaionale (4.54%), a analizei arhivelor sau a nregistrrilor organizaionale (4.54%) i a uneia mixte
n care datele au provenit prin intermediul scalelor i nregistrrilor organizaionale (4.54%).
77
4/17/2012
1:08 PM
Page 78
Figura 3. Distribuia studiilor incluse n analiz sub aspectul metodei de colectare a datelor
Not:
1 = Metoda observaional;
2 = Metoda psihofiziologic;
3 = Metoda psihofizic;
4 = Metode psihometrice;
5 = Scale i chestionare;
6 = Metoda intervievativ;
Perspectiva teoretic asupra nvrii

n ceea ce privete perspectiva teoretic din
care a fost abordat nvarea echipei, acest concept a fost examinat att ca proces, ct i ca
rezultat (Figura 4). n aproximativ trei sferturi
7 = Jurnale i metoda narativ;

8 = Metoda focus-grup;
9 = Metoda etnografic;
10 = Metoda cercetrii aciunii;
11 = Metoda de analiz a arhivelor;
12 = Metode mixte.
dintre eantioanele analizate (81.82%), nvarea n echip a fost examinat ca un proces de

interaciune a membrilor echipelor de munc.
Restul de 18.18% din totalul studiilor analizate,
s-au focalizat pe nvarea echipelor ca rezultat.
Figura 4. Distribuia studiilor incluse n studiu n funcie de perspectivele i aspectele teoretice ale
nvrii n echipa de munc
Not:
78
1 = nvarea ca proces;
2 = Comportamente de nvare;
3 = Activiti de nvare;
4 = Alte aspecte studiate ale nvrii echipei n
cadrul perspective procesuale;
5 = nvarea ca rezultat;
6 = nvarea ca rezultat cognitiv;
7 = nvarea ca rezultat comportamental;
8 = nvarea ca rezultat afectiv
4/17/2012
1:08 PM
Page 79
n cadrul abordrii nvrii echipei ca proces, cercettorii au studiat nvarea n echip n

termeni de comportamente (50%), activiti
(13.63%), mecanisme de nvare (4.54%, n=1),
proces de reflecie (4.54%), procesare multidimensional a informaiei de ctre echipa de
munc (4.54%) i creare de noi procese i practici (4.54%). Similar, nvarea n echip ca
rezultat a fost operaionalizat diferit. Unele
studii s-au focalizat pe aspecte cognitive ale
acestui concept (9.09%) iar altele pe cele de
ordin comportamental (9.09%).
Dimensionalitatea aspectelor
studiate ale nvrii
i efectivitii echipei de munc
Sub aspectul naturii multidimensionale a
nvrii n echip, cele mai multe studii au
abordat acest construct din perspectiv unidimensional (54.54%).

Mai mult de jumtate dintre studiile incluse
n analiz au investigat doar un aspect singular
al eficacitii echipei de munc (59.09%)
reprezentat de performana echipei. Studiile
care au analizat criterii multiple de evaluare a
eficacitii echipei de munc au inclus i criterii
reprezentate de eficien, calitatea relaiilor
interpersonale, satisfacia membrilor echipei i
nivelul de inovaie al echipei.
Rezultatele unei analizei mai detaliate a
studiilor care au abordat nvarea ca proces n
raport cu eficacitatea echipei de munc sub
aspectul naturii multidimensionale, metodei i a
sursei de colectare a datelor sunt incluse n
Tabel 2. Sumarizarea acestora este prezentat n
Tabelul 3.
Tabel 2. nvarea echipei ca proces i eficacitatea echipelor de munc

nvarea n echip
Eficacitatea echipei de munc
Referina
Relaie
Dimensiuni
Construct
Metod Surs
Dimensiuni
Construct
Metod
Surs
Scal
Observator
-.21
Performana n sarcin
Discutarea
despre
obiective
neclare
1. Bang et al. (2010)
Membri Multiplu
Comportament
Multidimensional
Discutarea
despre
comunicarea
nefocalizat
Construcia
semnificaiei
2. Stalmeijer et al.
(2007) Eantion 1
Scal
Scal
Membri
-.23
Calitate relaii
Scal
Membri
-.17
Satisfacia membrilor
Scal
Membri
-.42**
Performana n sarcin Scal
Observator
-.13
Scal
Observator
-.13
Scal
Membri
-.31*
Calitate relaii
Scal
Membri
-.23
Satisfacie membri
Scal
Membri
-.17
Legtura cu cunotine
Scal
anterioare
Membri
-.08
Coeren curs
Scal
Membri
-.28
Claritate obiective
Scal
Membri
.72
Evaluare
Scal
Membri
.55
Organizare
Scal
Membri
.63
Eficacitatea nvrii
Scal
Membri
.59
Calitatea sarcinilor
nvrii bazate pe
probleme
Scal
Membri
.32
Legtura cu cunotine
Scal
anterioare
Membri
.03
Coeren curs
Scal
Membri
.48
Claritate obiective
Scal
Membri
.94**
Evaluare
Scal
Membri
.46
Organizare
Scal
Membri
.88*
Eficacitatea nvrii
Scal
Membri
.50
nvrii bazate pe
probleme
Scal
Membri
.08
Performana n sarcin
Scal
Scal
Membri Multiplu
Membri Multiplu
Comportament
Multidimensional
Co-construcia
Scal
semnificaiei
Membri Multiplu
79
4/17/2012
1:09 PM
Page 80
(2007) Eantion 1
Comportament
Conflict
Multidimensional constructiv
Construcia
semnificaiei
(2007)Eantion 2
Factor latent
80
Scal
Comportament
Co-construcia
Scal
Multidimensional semnificaiei
Conflict
constructiv
4. Savelsberg et al.
(2009)
Scal
Scal
Membri Multiplu
Membri Multiplu
Membri Multiplu
Membri Multiplu
Legtura cu cunotine
Scal
anterioare
Membri
-.02
Coeren curs
Scal
Membri
.71
Claritate obiective
Scal
Membri
.81
Evaluare
Scal
Membri
.64
Organizare
Scal
Membri
.74
Eficacitatea nvrii
Scal
Membri
.66
nvrii bazate pe
probleme
Scal
Membri
.19
Legtura cu cunotine
Scal
anterioare
Experi
-.08
Coeren curs
Scal
Experi
.12
Claritate obiective
Scal
Experi
.27
Evaluare
Scal
Experi
.02
Organizare
Scal
Experi
.31
Eficacitatea nvrii
Scal
Experi
.09
nvrii bazate pe
probleme
Scal
Experi
.02
Legtura cu cunotine
Scal
anterioare
Experi
.06
Coeren curs
Scal
Experi
.26
Claritate obiective
Scal
Experi
.36
Evaluare
Scal
Experi
.24
Organizare
Scal
Experi
.46
Eficacitatea nvrii
Scal
Experi
-.18
nvrii bazate pe
probleme
Scal
Experi
.03
Legtura cu cunotine
Scal
anterioare
Experi
.12
Coeren curs
Scal
Experi
.71
Claritate obiective
Scal
Experi
.81
Evaluare
Scal
Experi
.64
Organizare
Scal
Experi
.74
Eficacitatea nvrii
Scal
Experi
.66
nvrii bazate pe
probleme
Scal
Experi
.19
Scal
Membri Singular
Performana echipei
Scal
Membri
.41**
Co-construcia
Scal
semnificaiei
Membri Singular
Performana echipei
Scal
Membri
.41**
Explorare
perspective
Scal
Membri Singular
Performana echipei
Scal
Membri
.46**
Analiz erori
Scal
Membri Singular
Performana echipei
Scal
Membri
.25*
Scal
Membri Singular
Performana echipei
Scal
Membri
.26**
Scal
Membri Singular
Performana echipei
Scal
Membri
.08
Reflecie
rezultate
Scal
Membri Singular
Performana echipei
Scal
Membri
.23*
Cutare
feedback
Scal
Membri Singular
Performana echipei
Scal
Membri
.21*
Experimentare Scal
Membri Singular
Performana echipei
Scal
Membri
.12
Comunicare
Comportament
Multidimensional erori
Reflecie
procese
4/17/2012
1:09 PM
Page 81
5. Edmondson
(2003)
Uurina
vorbirii
Interviu
Membri Singular
codare1
Succes implementare
Arhiv
nregistrri
organizaie
.63**
Uurina
vorbirii
Interviu
Membri Singular
codare2
Succes implementare
Arhiv
nregistrri
organizaie
.55*
Lrgirea
granielor
Interviu Membri Singular
Succes implementare
Arhiv
nregistrri
organizaie
.47*
Interviu
Membri Singular
codare1
Succes implementare
Arhiv
nregistrri
organizaie
.63**
Comportament
Uurina
Multidimensional vorbirii
Uurina
vorbirii
Lrgirea
granielor
Reflecie /
Practic
Interviu
Membri Singular
codare2
nregistrri
organizaie
nregistrri
organizaie
Succes implementare
Arhiv
Succes implementare
Arhiv
Succes implementare
Arhiv
nregistrri
organizaie
.66**
.40**
.55*
.47*
6. Van der Vegt et al. Comportament

(2010)
Unidimensional
Comportament
Scal
de nvare
Membri Singular
Performana echip
Scal
Supervizor
7. Huang et al.
(2008)
Comportament
Unidimensional
Comportament
Scal
de nvare
Membri Singular
Performana echip
Scal
Membri
8. Van der Vegt &

Bunderson (2005)
Comportament
Unidimensional
Comportament
Scal
de nvare
Membri Singular
Performana echip
Scal
Membri
9. Edmondson
(1999a)
Comportament
Unidimensional
Comportamente
Scal
de nvare
Membri Singular
Performana echip
Scal
Membri
.54**
Singular
Performana echip
Scal
Membri
.71*
Singular
Performana echip
Scal
Manageri
.52*
Singular
Performana echip
Scal
Membri
.34*
Singular
Performana echip
Scal
Manager
.81*
Eficacitatea
Scal
Membri
.57**
Satisfacie membri
Scal
Membri
.54**
Membri
10. Edmondson
(1999b)
Comportament
Unidimensional
Comportamente
Scal
de nvare
Manager
11. Yeh & Chou

(2005)
12. van Woerkom &

van Engen (2009)
13. Tucker et al.

(2007)
Comportament
Unidimensional
Comportamente
Scal
de nvare
Membri Multiplu
Factor latent
Scal
Membri Singular
Performana echip
Scal
Membri
.44**
Procesarea
informaiei
Scal
Membri Singular
Performana echip
Scal
Membri
.50**
Scal
Membri Singular
Performana echip
Scal
Membri
.33**
Scal
Membri Singular
Performana echip
Scal
Membri
.34**
Scal
Membri Singular
Succes implementare
Scal
Membri
.72**
Scal
Membri Singular
Succes implementare
Scal
Membri
.48*
Membri
.24
Calitatea obinut
Scal
Manageri
.18
Membri
.13
Manageri
.10
Membri
.52*
Manageri
.20
Activitate
Achiziia
Multidimensional informaiei
Stocare i
reactualizare
informaiilor
nva-ce
Activitate
Multidimensional nva-cum
Achiziia
informaiei
Scal
Membri Multiplu
Eficiena
Nivel inovaie
Calitatea obinut
14. van Woerkom &
Croon (2009)
Activitate
Procesarea
Multidimensional
informaiei
Scal
Membri Multiplu
Eficiena
Nivel inovaie
Calitatea obinut
Scal
Scal
Membri
.84***
Manageri
.60***
Membri
.83***
Manageri
.41**
Membri
.69***
Scal
Scal
Scal
Manageri
.18
Membri
.21
Scal
Manageri
Stocare i
reactualizare
informaie
Membri
Scal
Membri Multiplu
Eficiena
.36*
Scal
Manageri
Nivel inovaie
.46***
.40***
Membri
.26
Manageri
.19
Scal
81
4/17/2012
1:09 PM
Page 82
nvare
integrat
Interviu
SuperSingular
vizor
Erori medicaie
ObserMembri
vaie
-.45**
Interviu
SuperSingular
vizor
Erori medicaie
ObserMembri
vaie
.62**
Interviu
SuperSingular
vizor
Erori medicaie
ObserMembri
vaie
.41*
nvare
supervizat
Interviu
SuperSingular
vizor
Erori medicaie
ObserMembri
vaie
.40*
Performana echipei
Scal
Manageri
.20**
nvarea
echipei
Scal
Membri Multiplu
Nivel inovaie
Arhiv
nregistrri
organizaie
.20**
Achiziia
informaiei
Scal
Manager Singular
Succes produs
Scal
Manager
.44**
Implementare
informaie
Scal
Manager Singular
Succes produs
Scal
Manager
.54**
Scal
Manager Singular
Succes produs
Scal
Manager
.24*
Scal
Manager Singular
Succes produs
Scal
Manager
.08
Scal
Manager Singular
Succes produs
Scal
Manager
.35**
Improvizaie
Scal
Manager Singular
Succes produs
Scal
Manager
.16*
Crearea
semnificaiei
Scal
Manager Singular
Succes produs
Scal
Manager
.44**
Memorie
Scal
Manager Singular
Succes produs
Scal
Manager
.35**
Creare de
procese i
practici
Performan n sarcin Scal
Manager
.48**
Scal
Manager Multiplu
Calitatea relaiilor
interpersonale
Membri
.33**
nvare
Mecanisme de
non-integrat
15. Drach-Zahavy &
nvare
Pud (2010)
Multidimensional nvare
neuniform
Proces de
16. Somech &
nvare
Drach-Zahavy (2007)
Unidimensional
17. Akgn et al.

(2006)
Diseminarea
informaie
Procesele
Ne-nvare
nvrii
Multidimensional
Gndire
18. Zellmer-Bruhn & Proces

Gibson (2006)
Unidimensional
n Tabelul 3 n prima coloan sunt prezentate aspecte ale nvrii n echip iar n primul
rnd cele legate de eficacitatea echipei de
munc. La intersecia lor se gsete procentul de
studii n care se regsesc cele dou concepte studiate n funcie de instrumentele i sursa colectrii datelor. Studiile care au examinat comportamentele de nvare i eficacitatea echipei de
munc, au conceptualizat de cele mai multe ori
acest concept din perspectiva unidimensional
(27.27%). n cadrul acestor cercetri, nvarea
n echip a fost examinat mai frecvent n raport
cu un singur criteriu al eficacitii echipei de
munc (22.72%). Cele dou concepte au fost
evaluate prin acelai tip de instrumente de
msurare, i anume scale. Datele au fost colectate de la aceeai surs (membrii grupului,
18.18%) sau surse diferite (membrii echipelor
de munc i managerii acestora, 4.54%).
nvarea ca i comportament unidimensional a
fost examinat n raport cu criterii multiple ale
eficacitii echipei de munc doar n cadrul unui
studiu (performana perceput i satisfacia
membrilor echipei). Datele privind cele dou
concepte reprezint percepii ale membrilor
82
Scal
echipei de munc, msurate prin intermediul

scalelor. Toate studiile care au abordat unidimensional nvarea ca i comportament al
echipei n relaie cu eficacitatea echipei de
munc au evideniat existena unei relaii pozitive ntre cele dou concepte.
n schimb, cercetrile care au investigat
natura multidimensional a comportamentului
de nvare al echipei au examinat acest concept
mai mult n relaie cu criterii multiple ale eficacitii echipei de munc (13.63%). Dou dintre
aceste studii s-au focalizat pe conceptualizarea
multidimensional a performanei iar unul a
examinat att performana, ct i alte dou criterii reprezentate de viabilitatea i satisfacia
membrilor echipei de munc. Majoritatea studiilor care au examinat comportamente de
nvare n echip din perspectiv multipl au
utilizat scale pentru a msura cele dou concepte (18.18%). ns, n studiile care au relaionat nvarea cu un singur criteriu al eficacitii,
datele au fost colectate prin intermediul scalelor
de la membrii echipei (4.54%) i al unei metode
mixte alctuite din interviuri i nregistrri organizaionale (4.54%). Cercetrile care au
4/17/2012
1:09 PM
Page 83
Tabel 3. Sumarizarea rezultatelor analizei privind nvarea ca proces i eficacitatea echipei de munc
83
4/17/2012
1:09 PM
Page 84
tate (calitatea ndeplinirii sarcinilor, eficien i

nivel de inovaie). n cadrul acestuia, nvarea a
fost evaluat doar de ctre membrii echipei, ns
criteriile de eficacitate au fost evaluate att de
membrii echipei, ct i de managerul acesteia
prin intermediul scalelor. Criteriul de eficacitate
cel mai utilizat n aceast categorie de studii este
performana perceput a echipei. n ceea ce
privete forma relaiei de asociere dintre activitile de nvare i eficacitatea echipei de
munc, rezultatele sunt inconsistente.
Analiza nvrii n echip din perspectiv
procesual a evideniat un singur studiu care s-a
focalizat pe relaia dintre mecanismele de
nvare identificate din interviurile realizate cu
membrii echipei i performana obiectiv a
echipei, operaionalizat ca erori efectuate n
administrarea medicaiei pacienilor (DrachZahavy & Pud, 2010). Un alt studiu care a
examinat nvarea echipei ca proces de reflecie
unidimensional a utilizat metode diferite pentru
a msura cele dou concepte. Eficacitatea
echipei, conceptualizat prin criterii multiple, a
fost msurat prin intermediul scalelor i a
datelor obiective (Somech & Drach-Zahavy,
2007). ns aceste tipuri de msurtori au surprins fiecare criterii diferite ale acestui concept.
Studiul realizat de Zellmer-Bruhn i Gibson
(2006) a examinat nvarea n echip sub
aspectul crerii de procese i practici. Acest pro-
relaionat nvarea ca i comportament cu criterii multiple ale eficacitii echipei de munc

au utilizat doar date care au fost colectate prin
intermediul scalelor, att din aceeai surs
(membrii echipei de munc, 4.54%), ct i din
surse diferite (membrii echipei de munc i
managerul acesteia, 4.54%; membrii echipei de
munc i experi, 4.54%). Comportamentul de
nvare cel mai frecvent studiat este reprezentat
de construcia comun a semnificaiei de ctre o
echip de munc (13.63%). Acesta descrie
aciunile conversaionale ale membrilor echipei
prin care acetia rafineaz, construiesc sau modific semnificaia original atribuit astfel nct
ajung la noi semnificaii care nu le-au fost
disponibile anterior (Savelsbergh et al., 2009).
ns, analiza indicilor de asociere relev faptul
c acest comportament se relaioneaz diferit cu
performana echipei de munc atunci cnd
aceasta este evaluat de ctre membrii echipei
sau experii din afara echipei de munc.
nvarea n echip ca i activitate a fost
studiat doar din perspectiv multidimensional
(13.63%). Dou dintre studiile din aceast categorie s-au focalizat pe activitile de nvare n
raport cu un singur criteriu al eficacitii echipei
de munc. Datele pentru cele dou concepte
reprezint percepii ale membrilor echipei. Doar
un singur studiu a examinat activitile de
nvare n raport cu criterii multiple de eficaci-
Tabel 4. nvarea n echip n echip ca rezultat i eficacitatea echipelor de munc

nvarea n echipa de munc
Eficacitatea echipei de munc
Referina
Aspect studiat
Dimensiuni
Metod
Sursa
Dimensiuni
1. Bstieler &
Hemmert
(2010)
nvare
rezultat cognitiv
Unidimensional
Scal
Membri
Singular
2. Akgn et
al. (2005)
nvare
rezultat cognitiv
3. Akgn et
al. (2006)
nvare
rezultat
comportamental
4. Sarin &
McDermott
(2003)
nvare
rezultat
comportamental
** p < .01.
84
Unidimensional
Unidimensional
Unidimensional
Scal
Scal
Scal
Manager
Manager
Membri
Aspect
studiat
Relaia
Metod
Sursa
Performan
echip
Scal
Membri
.63**
Viteza spre
pia
Scal
Manager
.03
Succes
produs
Scal
Manager
.35**
Succes
produs
Scal
Manager
.55**
Viteza spre
pia
Scal
Membri
.17**
Nivel inovaie
Scal
Membri
.34**
Multiplu
Singular
Multiplu
4/17/2012
1:09 PM
Page 85
ces a fost relaionat cu performana echipei i

calitatea relaiilor interpersonale. Totodat,
Akgn, Lynn i Yilmaz (2006) au inclus
msurtori subiective ale proceselor de nvare
provenite doar de la conductorii sau managerii
echipelor de proiect.
Rezultatele unei analizei mai detaliate a studiilor care au abordat nvarea ca rezultat n raport cu eficacitatea echipei de munc sub aspectul naturii multidimensionale, metodei i a sursei
de colectare a datelor sunt incluse n Tabel 4.
Datele reflect faptul c nvarea n echip
ca rezultat a fost operaionalizat diferit n
cadrul celor patru studii incluse n analiz. Aceste operaionalizri au vizat rezultate cognitive
(9.09%) sau de ordin comportamental (9.09%).
ns, indiferent de aspectul examinat, n aceste
studii, nvarea a fost conceptualizat dintr-o
perspectiv unidimensional. Datele privind
acest concept i eficacitatea echipei reprezint
percepii ale managerilor (9.09%) i ale membrilor echipei (9.09%) msurate prin intermediul scalelor. Criteriul de eficacitate cel mai des
utilizat este performana echipei, evideniinduse preponderent o relaie pozitiv ntre aceste
dou concepte.
n cazul nvrii ca rezultat de ordin cognitiv, ntr-unul dintre studiile analizate acest concept a fost examinat n relaie cu un singur criteriu al eficacitii echipei. Datele pentru cele
dou concepte au fost furnizate de ctre membrii echipei (Bstieler & Hemmert, 2010). ntr-un
alt studiu, nvarea a fost examinat n relaie
cu criterii multiple ale eficacitii pe baza
datelor colectate de la managerii echipei de proiect (Akgn et al, 2005). n acest studiu,
nvarea ca rezultat a fost considerat drept un
indicator al performanei echipei alturi de alte
dou variabile reprezentate de rapiditatea cu
care echipa de munc a dezvoltat un produs i
succesul pe pia al acestui produs (p. 1109).
Studiile care au analizat nvarea echipei
ca rezultat exprimat n comportament au examinat acest concept n raport cu criterii multiple
ale eficacitii echipei percepute din perspectiva
membrilor echipei (Sarin & McDermott, 2003)
i n raport cu succesul noului produs ca indicator singular al eficacitii echipei de proiect
(Akgn et al., 2006).
Discuii
Rezultatele obinute n studiul de fa
reliefeaz faptul c o cunoatere a relaiei dintre
cele dou concepte este mai puin informat de
studii realizate cu eantioane cu un numr mare
de participani. Mai mult, aceast cunoatere
este bazat pe rezultate obinute doar cu
eantioane formate din echipe de aciune i performan, management i producie. Aceste
rezultate sunt similare cu cele existente n literatura de specialitate privind eficacitatea echipei
de munc. n cadrul unei analize a studiilor publicate n perioada 1999-2004, Nielsen, Sundstrom i Halfhill (2005) au artat c cel mai utilizat tip de echipe de munc n studiul
eficacitii echipelor de munc este cel
reprezentat de echipele de munc care ofer servicii (40%), echipele multiple (29%) i cele de
proiect (18%). n plus, n studiul de fa,
domeniile de activitate din care au provenit
echipele de munc sunt variate de la un studiu la
altul i nu permit evidenierea existenei unor
particulariti ale nvrii n echip n contextul
unui anumit tip de echip de munc.
La acestea se adaug accentul pus mai
degrab pe relaia dintre cele dou concepte prin
apelul la un design de cercetare corelaional n
care datele au fost colectate simultan. Acest
rezultat este n acord cu faptul c majoritatea
cercetrilor din domeniul organizaional au la
baz designuri de natur corelaional (StoneRomero, 2009, 2011). ns, utilizarea acestora
limiteaz posibilitatea de a explora relaii
dinamice i direcii de cauzalitate ntre conceptele studiate (Bang et al., 2010; Edmondson,
1999; Van der Vegt & Bunderson, 2005; Van der
Vegt et al., 2010; van Woerkom & Croon, 2009).
n plus, dei teoria sugereaz faptul c nvarea
echipei permite mbuntirea calitii relaiilor
interpersonale ca indicator al eficacitii echipei
de munc, este plauzibil ca aceast relaie s fie
reciproc (Zellmer-Bruhn & Gibson, 2006,
p.514). Analiza derulat n studiul de fa a relevat faptul c nici unul dintre studiile examinate
nu a avut drept scop examinarea existenei unei
astfel de relaii de reciprocitate. Plauzibilitatea
acestei relaii este informat de rezultatele unui
studiu longitudinal care a utilizat echipe de studeni (Mo & Xie, 2009). Avnd la baz perspec85
4/17/2012
1:09 PM
Page 86
tiva modelului IMOI (Ilgen et al., 2005), examinarea relaiei dintre cele dou concepte s-a realizat pe baza a dou seturi de date de la 55 de
echipe. Datele au evideniat c performana
echipei, msurat ca output la sfritul primului
set de msurtori, constituie o variabil input
important n cadrul celui de-al doilea set de
msurtori. Astfel, s-a artat c performana
echipei are influene pozitive asupra proceselor
subsecvente de nvare ale echipei. n plus, n
literatura de specialitate au fost identificate sugestii conform crora relaia dintre structura, procesele echipei, printre care i nvarea n
echip, i eficacitatea echipelor de munc ar
putea fi mai degrab una de tip spiral i nu una
liniar (Somech & Drach-Zahavy, 2007). Astfel,
se contureaz necesitatea realizrii unor studii
care s includ designuri de cercetare care s
permit surprinderea unor astfel de relaii
dinamice ntre cele dou concepte de interes.
Rezultate obinute n studiul de fa evideniaz c nvarea n echip a fost studiat
din perspective teoretice diferite, ns cele mai
multe studii au avut la baz definiia oferit de
Edmondson (1999b, p. 353): proces emergent
de reflecie i aciune colectiv. Aceast diversitate conceptual confirm i rezultatele existente n literatura de specialitate (Decuyper et
al., 2010), reliefnd lipsa de claritate n ceea ce
privete semnificaia conceptului de nvare n
echip (Edmonson et al., 2008; Wilson et al.,
2007; Goodman & Dabbish, 2011). n urma
analizei derulate, am identificat c unele dintre
studii au examinat nvarea echipei din perspectiv procesual, iar altele s-au focalizat pe
nvarea echipei ca rezultat. ns cele mai multe
s-au focalizat pe nvarea echipei ca proces, n
special pe comportamentele de nvare ale
echipei msurate unidimensional. Cu toate c n
literatura de specialitate exist conceptualizri
ale nvrii echipei att ca proces, ct i ca
rezultat (Argote et al., 2001; Decuyper et al.,
2010; Goodman & Dabbish, 2011; Wilson et al.,
2007), analiza realizat de noi nu a identificat
nici un studiu empiric care s fi avut la baz o
astfel de perspectiv teoretic. Aceste rezultate
sugereaz faptul c n cadrul studiilor analizate
focalizarea asupra examinrii fenomenelor de
interes a fost cu preponderen dintr-o singur
perspectiv, i anume cea procesual. Astfel,
86
acest lucru nu a permis furnizarea unu tip specific de cunoatere privind relaia ntre nvarea
n echip i eficacitatea echipelor din mediul
organizaional.
Diversitatea definiiilor i a aspectelor studiate ale nvrii n echip are implicaii asupra
generalizrii relaiei dintre acest concept i eficacitatea echipei de munc. Toate studiile care
au abordat unidimensional nvarea ca i comportament al echipei n relaie cu eficacitatea
echipei de munc, au evideniat o relaie pozitiv ntre aceste dou concepte. n schimb,
cercetrile privind relaia dintre multiple comportamente de nvare i eficacitatea echipei de
munc au evideniat existena unor relaii contradictorii. Acelai comportament de nvare
este asociat cu eficacitatea echipei de munc iar
alteori nu. Aceste rezultate sugereaz c relaia
dintre nvarea i eficacitatea echipei de munc
trebuie neleas din perspectiva teoretic adoptat n studiile analizate.
Datele din studiul de fa au evideniat faptul c, de cele mai multe ori, nvarea a fost
examinat n relaie cu un singur criteriu al eficacitii echipei de munc. Cel mai studiat criteriu de eficacitate este cel reprezentat de performana echipei de munc evaluat de ctre
membrii echipei. Aceste rezultate sunt similare
cu cele ale unor analize existente n literatura de
specialitate privind eficacitatea echipelor de
munc care au evideniat faptul c cele mai
multe dintre cercetrile dedicate acestui concept
au utilizat msurtori de ordin subiectiv (Nielson et al. 2005; Sundstrom et al., 2000). De asemenea, nu au fost identificate cercetri care s fi
inclus evaluri ale clienilor sau ale beneficiarilor proiectelor sau serviciilor oferite de ctre
echipa de munc, n ciuda faptului c aceste
tipuri de echipe au fost utilizate cel mai des n
cadrul studiilor analizate. n plus, similar unor
studii existente n literatura de specialitate (e.g.
Mathieu et al., 2008), datele obinute n studiul
de fa au evideniat c nvarea echipei este
examinat mai puin frecvent n relaie cu eficiena, viabilitatea, satisfacia membrilor i
nivelul de inovaie ca indicatori ai eficacitii
echipei de munc.
Majoritatea studiilor incluse n analiza de
fa au utilizat metode de colectare a datelor de
tip auto-declarativ, n general chestionare/scale.
4/17/2012
1:09 PM
Page 87
Acest lucru poate influena obinerea unui set de

rspunsuri consistente privind relaia dintre cele
dou concepte. Reducerea influenei unei astfel
de distorsiuni poate fi realizat prin utilizarea
unei abordri de tipul triangulaiei n cercetri
de tip follow-up (Savelsbergh et al., 2009) sau
utilizarea unor msurtori obiective ale acestor
concepte (van Woerkom & Croon, 2009; van
Woerkom & van Engen, 2009). n plus, studiul
de fa a evideniat c aproximativ 40.90% dintre cercetrile analizate au utilizat date colectate
prin instrumente de msurare similare oferite de
aceeai surs, n special reprezentat de membrii echipei. Pentru a reduce variana metodei
comune, Podsakoff i Organ (1986) au recomandat obinerea msurtorilor multiple ale
variabilelor de interes prin intermediul unor
metode multiple aplicate unor surse mutiple sau,
atunci cnd acest lucru nu este posibil, utilizarea
unor surse diferite pentru msurarea acestor
variabile. Restricionarea investigrii acestui
concept la un singur tip de metodologie permite
surprinderea unui aspect singular al acesteia,
care poate fi ns poate fi diferit de ceea ce au
investigat ali cercettori printr-un un alt tip de
metodologie (Edmondson et al., 2008).
Dat fiind natura corelaional a datelor de
ordin subiectiv despre relaia dintre nvarea n
echip i eficacitatea echipei de munc,
realizarea unor studii longitudinale, cu msurtori obiective, ar permite obinerea unor informaii relevante privind aceast relaie (van
Woerkom & Croon, 2009; van Woerkom & van
Engen, 2009). Utilizarea unui design de cercetare care s permit colectarea mai multor seturi
de date va putea oferi informaii suplimentare
despre stabilitatea i schimbarea celor dou concepte (Van der Vegt & Bunderson, 2005), precum i despre relaiile dintre ele (de Lange,
2005; Taris & Kompier, 2003; apud. Savelsbergh et al., 2009). Acest tip de cercetri este
necesar n condiiile n care nu ntotdeauna este
realizat distincia ntre nvare n echip i alte
concepte, cum este performana echipei de
munc (Wilson et al., 2007). n acest sens, n
analiza de fa a fost identificat un studiu n care
nvarea n echip ca rezultat era considerat ca
un indicator al performanei echipei (Akgn et
al., 2005, p. 1109).
Rezultatele obinute n studiul de fa pot fi

considerate drept complementare altor analize
existente n literatura de specialitate din acest
domeniu. Dar spre deosebire de acestea, studiul
de fa constituie un prim efort n ncercarea de
a analiza i integra ntr-o manier sistematic
ceea ce s-a studiat pn n prezent privind relaia
dintre nvarea i eficacitatea echipelor care i
desfoar activitatea n context organizaional.
n plus, pe lng identificarea perspectivelor
teoretice din care a fost examinat empiric
aceast relaie, a fost evideniat modul n care
aceast relaie a fost studiat. Acest lucru a contribuit la conturarea unei set de factori care ar
putea explica rezultatele inconsistente privind
relaia dintre nvarea n echip i eficacitatea
echipei de munc. ns, pentru a testa dac ntradevr aceti factori au o influen asupra
relaiei dintre cele dou concepte, studiile
viitoare ar putea construi pe contribuia analizei
derulate de noi prin ncercri de integrare metaanalitic a rezultatelor studiilor incluse n
aceast analiz.
n scopul crerii unei cunoateri ample
privind relaia dintre nvarea n echip i eficacitatea echipei, rezultatele obinute n studiul
de fa ar putea fi comparate cu cele ale unei
analize similare a cercetrilor care au examinat
aceast relaie n cadrul unor eantioane formate
din echipe de munc virtuale sau din alte tipuri
de echipe. n plus, studiile ulterioare ar putea
include n analiz i relaia dintre alte concepte
strns legate de nvarea n echip, cum ar fi
orientarea spre nvare sau reflexivitatea
echipei, i eficacitatea echipei de munc. Acestea ar putea contribui la aducerea unor informaii suplimentare necesare clarificrii rezultatelor inconsistente privind asocierea dintre
nvarea n echip i eficacitatea echipei.
Concluzii
Rezultatele obinute n studiul de fa evideniaz statutul de concept umbrel al nvrii echipei prin identificarea diversitii perspectivelor teoretice din care a fost studiat acest
concept n relaie cu eficacitatea echipei de
munc. Acest lucru apare n contrast cu faptul c
toate studiile incluse n analiz au utilizat acelai tip de design de cercetare iar majoritatea lor
au utilizat metode i surse de colectare a datelor
similare. Date fiind tipul echipelor de munc n
87
4/17/2012
1:09 PM
Page 88
care a fost studiat aceast relaie i convergena

studiilor analizate sub aspect metodologic, considerm c rezultatele analizei derulate de noi
pot constitui un reper important pentru realizarea unor studii empirice care s fac apel la
metode multiple de colectare a datelor despre
acelai fenomen din surse diferite.
Bibliografie
*Akgn, A. E., Byrne, J. C., Keskin, H., & Lynn, G. S.
(2006). Transactive memory system in new product development teams. IEEE Transactions on
Engineering Management, 53, 1, 95-111.
*Akgn, A. E., Byrne, J., Keskin, H., Lynn, G. S., &
Imamoglu, S. Z. (2005). Knowledge networks in
new product development projects: A transactive
memory perspective. Information & Management,
42, 1105-1120.
*Akgn, A. E., Lynn, G. S., & Yilmaz, C. (2006). Learning process in new product development teams
and effects on product succes: A socio-cognitive
perspective. Industrial Marketing Management,
35, 210-224.
Argote, L., Gruengeld, D., & Naquin, C. (2001). Group
learning in organizations. In M. Turner (Ed.),
Groups at work: Theory and research (pp. 369412). Mahwah, NJ: Erlbaum.
Balkundi, P., & Harrison, D.A. (2006). Ties, leaders,
and time in teams: Strong inference about the
effects of network structure on team viability and
performance. Academy of Management Journal,
49, 49-68.
*Bang, H., Fulesang, S. L., Ovesen, M. R., & Eilertsen,
D. E. (2010). Effectiveness in top management
group meetings: The role of goal clarity, focused
communication, and learning behavior. Scandinavian Journal of Psychology, 51, 253-261.
Breakwell, G.M., & Hammond, S., Fife-Schaw, C., &
Smith, J.A. (Eds.) (2006). Research methods in
psychology. London, Thousand Oaks, New Delhi:
Sage Publications.
Brewerton, P., & Millward, L. (2001). Organizational
research methods: A guide for students and
researchers. London, Thousand Oaks, New Delhi:
Sage Publications.
Chan, C.C., Pearson, C., & Entrekin, L. (2003). Examining the effects of internal and external team
learning on team performance. Team Performance
Management: An International Journal, 7/8,174181.
Cureu, P. L. (2007). Grupurile n organizaii. Iai: Editura Polirom.
88
Decuyper, S., Dochy, F., & Van den Bossche, P. (2010).

Grasping the dynamic complexity of team learning: An integrative model for effective team learning in organizations. Educational Research
Review, 5, 111-133.
*Drach-Zahavy, A., & Pud, D. (2010). Learning mechanism to limit medication administration errors.
Journal of Advanced Nursing, 66(4), 794-805.
*Edmondson, A. (1999a). A safe harbor from which to
venture out: Social psychological conditions
enabling boundary spanning in work teams. In R.
Wageman (Ed.), Research on managing groups
and teams Groups in context, vol. 2 (pp. 1792199). Stamford: JAI.
*Edmondson, A. (1999b). Psychological safety and
learning behavior in work teams. Administrative
Science Quarterly, 44, 350-383.
*Edmondson, A. C. (2003). Speaking up in the operating room: How team leaders promote learning in
interdisciplinary action teams. Journal of Management Studies, 40(6), 1419-1452.
Edmondson, A. C., Dillon, J. R., & Roloff, K. S. (2008).
Three perspectives on team learning: Outcome
improvement, task mastery, and group process. In
J. P. Walsh, & A. P. Brief (Eds.), The Academy of
Management Annals (vol. 1) (pp. 269-314). New
York: Taylor & Francis Group/Lawrence Erlbaum
Associates.
Ellis, A.P., Hollenbeck, J.R., Ilgen, D.R., Porter, C.O.,
West, B.J., & Moon, H. (2003). Team learning:
Collectively connecting the dots. Journal of
Applied Psychology, 88, 5, 821-835.
Gibson, C. B., & Vermeulen, F. (2003). A healthy
divide: Subgroups as a stimulus for team learning.
Administrative Science Quarterly, 48, 202-239.
Gil, F., Alcover, C. M., & Peir, J.-M. (2005). Work
team effectiveness in organizational contexts.
Recent research and applications in Spain and Portugal. Journal of Managerial Psychology, 20(3/4),
193-218.
Goodman, P. S., & Dabbish, L. A. (2011). Methodological issues in measuring group learning. Small
Group Research, 42, 379-404.
*Huang, C.C., Chu, C.Y., & Jiang, P.C. (2008). An
empirical study of psychological safety and performance in technology R&D teams. Proceedings
of the 2008 IEEE ICMIT, 1423-1427.
Ilgen, D. R., Hollenbeck, J. R., Johnson, M., & Jundt,
D. (2005). Teams in organizations: From I-P-O
models to IMOI models. Annual Review of Psychology, 56, 517-543.
Jehn, K. A., & Rupert, J. (2008). Group faultlines and
team learning: How to benefit from different perspectives. In V. I. Sessa, & M. London (Ed.), Work
group learning: Understanding, improving,
assessing how groups learn in organizations (pp.
119-147). New York: Taylor & Francis Group,
LLC.
4/17/2012
1:09 PM
Page 89
Jordan, M.H., Feild, H.S., & Armenakis, A.A. (2002).

The relationship of group process variables and
team performance: A team-level analysis in a field
setting. Small Group Research, 33, 121-150.
Kraiger, K., Ford, J. K., & Salas, E. (1993). Application
of cognitive, skill-based, and affective theories of
learning oucomes to new methods of training evaluation. Journal of Applied Psychology, 78, 311328.
London, M., & Sessa, V. I. (2007). How groups learn
continuously. Human Resources Management, 46,
4, 651-669.
Marks, M. A., Mathieu, J. E., & Zaccaro, S. J. (2001). A
temporally based framework and taxonomy of
team processes. Academy of Management Review,
26, 356-376.
Mathieu, J., Maynard, M.T., Rapp, T., & Gilson, L.
(2008). Team effectiveness 1997-2007: A review
of recent advancements and a glimpse into the
future. Journal of Management, 34, 410-486.
Mo, S., & Xie, X. (2009). Team learning, transactive
memory systems and team performance: A longitudinal study based on the IMOI approach. Front.
Bus. Res. China, 4(3), 409-422.
Montero, I., & Len, O.G. (2007). Aguide for naming
research studies in Psychology. International
Journal of Clinical and Health Psychology,7(3),
847-862.
Nielsen, T. M., Sundstrom, E. D., & Halfhill, T. R.
(2005). Group dynamics and effectiveness: Five
years of applied research. In S. A. Wheelan (Ed.),
The handbook of group research and practice (pp.
285-312). Thousand Oaks, California: Sage Publications.
Pia, M.I., Martnez, A.M., & Martnez, L.G. (2008).
Teams in organizations: A review on team effectiveness. Team Performance Management,
14(1/2), 7-21.
Podsakoff, P. M., & Organ, D. W. (1986). Self-reports in
organizational research: Problems and prospects.
Journal of Management, 12, 531-544.
*Sarin, S., & McDermott, C. (2003). The effect of team
leader characteristics on learning, knowledge
application, and performance of cross-functional
new development teams. Decision Science, 34(4),
707-739.
*Savelsbergh, C. M., van der Heijden, B. I., & Poell, R.
F. (2009). The development and empirical validation of a multidimensional measurement instrument for team learning behaviors. Small Group
Research
OnlineFirst,
doi:
10.1177/1046496409340055.
Savelsbergh, C. M., van der Heijden, B. I., & Poell, R.
F. (2010). Attitudes towards factors influencing
team performance: A multi-rater approach at estalishing the relative importance of team learning
behaviors in comparison with other predictors of
team performance. Team Performance Management, 16(7/8), 451-474.

Singh, A. K., & Muncherji, N. (2007). Team effectiveness and its measurement: A framework. Global
Business Review, 8, 119-133.
*Somech, A., & Drach-Zahavy, A. (2007). Schools as
team-based organizations: A structure-processoutcomes approach. Group Dynamics: Theory,
Research, and Practice, 11(4), 305-320.
*Stalmeijer, R. E., Gijselaers, W. H., Wolfhagen, I. H.,
Harendza, S., & Scherpbier, A. J. (2007). How
interdisciplinary teams can create multi-disciplnary education: The interplay between team processes and educational quality. Medical Education, 41, 1059-1066.
Stone-Romero, E. F. (2009). Implications of research
design options for the validity of inferences
derived from organizational research. In D. A.
Buchanan & A. Bryman (Eds.), The Sage handbook of organizational research methods (pp. 302327). Thousand Oaks, CA: Sage Publications.
Stone-Romero, E. F. (2011). Research strategies in
industrial and organizational psychology: Nonexperimental, quasi-experimental, and randomized experimental research in special purpose and
nonspecial purpose settings. In S. Zedeck (Ed.),
APA handbook of industrial and organizational
psychology, Volume 1: Building and developing
the organization (pp. 37-72). Washington, DC:
American Psychological Association.
Sundstrom, E., McIntyre, M., Halfhill, T., & Richards,
H. (2000). Work groups: From the Howthorne
studies to work teams of the 1990s and beyond.
Group Dynamics: Theory, Research, and Practice,
4(1), 44-67.
Tekleab, A.G., Quigley, N.R., & Tesluk, P.E. (2009). A
longitudinal study of team conflict, conflict management, cohesion, and team effectiveness. Group
Organization Management, 34, 170-205.
*Tucker, A. L., Nebrhard, I. M., & Edmondson, A. C.
(2007). Implementing new practices: An empirical
study of organizational learning in hospital intensive care units. Management Science, 53(6),
894-907.
*Van der Vegt, G. S., & Bunderson, J. S. (2005). Learning and performance in multidisciplinary teams:
The importance of collective team identification.
Academy of Management Journal, 48(3), 532-547.
*Van der Vegt, G. S., de Jong, S. B., Bunderson, J. S., &
Molleman, E. (2010). Power asymmetry and
learning in teams: The moderating role of performance feedback. Organization Science, 21(2),
347-361.
*van Woerkom, M., & Croon, M. (2009). The relationship between team learning activities and team
performance. Personnel Review, 38(5), 560-577.
*van Woerkom, M., & van Engen, M. L. (2009). Learning from conflicts? The relations between task and
89
4/17/2012
1:09 PM
Page 90
relationship conflicts, team learning and team performance. European Journal of Work and Organizational Psychology, 18(4), 381-404.
Wilson, J. M., Goodman, P. S., & Cronin, M. A. (2007).
Group learning. Academy of Management Review,
32(4), 1041-1059.
*Yeh, Y.-J., & Chou, H. W. (2005). Team composition
and learning behaviors in cross-functional teams.
Social Behavior and Personality, 33(4), 391-402.
Zaccaro, S.J., Ely, K., & Shuffler, M. (2008). The leaders role in group learning. In V. I. Sessa, & M.,
London (Ed.), Work group learning: Understanding, improving & assessing how groups learn in
organizations (pp. 193-214). New York: Taylor &
Francis Group, LLC.
*Zellmer-Bruhn, M., & Gibson, C. (2006). Multinational organizational context: Implications for team
learning and performance. Academy of Management Journal, 49(3), 501-518.
Nota autorului
*Autorii doresc s mulumeasc pentru suportul financiar din Programul co-finanat de Operaional Sectorial pentru Dezvoltarea Resurselor Umane 2007 2013, Contract POSDRU 6/1.5/S/4 STUDII DOCTORALE, FACTOR
MAJOR DE DEZVOLTARE AL CERCETRILOR SOCIO-ECONOMICE I UMANISTE.
90
4/17/2012
1:09 PM
Page 91
Restricia de amplitudine,
o ameninare ascuns la adresa validitii de criteriu
Marian Popa1
Abstract
Range restriction is manifested by reducing artificial variability in the measured values of sample variability relative to the reference population. Its direct effect is to reduce the Pearson correlation coefficient and underestimation of the predictive power of psychological tests. The article analyzes the sources and effects of range
restriction, and describes how to correct it.
Keywords: range restriction, variability, predictive power, psychological tests
Rsum
La restriction damplitude se manifeste par la rduction artificielle de la variabilit dans les valeurs mesures
sur chantillon, par rapport la population de rfrence. Son effet direct est de rduire le coefficient de corrlation de Pearson et de sous-estimation de la puissance prdictive des tests psychologiques. Larticle analyse les
sources et les effets de la restriction damplitude, et dcrit comment la corriger.
Mots-cls: la restriction damplitude, variabilit, puissance prdictive, tests psychologiques
Rezumat
Restricia de amplitudine se manifest prin diminuarea artificial a variabilitii valorilor msurate la nivelul eantionului n raport cu variabilitatea populaiei de referin. Efectul direct al acesteia este reducerea coeficientului de corelaie Pearson i subestimarea puterii predictive a testelor psihologice. Articolul analizeaz sursele i efectele restriciei de amplitudine, i descrie modalitile de corecie ale acesteia.
Cuvinte-cheie: restricia de amplitudine, variabilitate, putere predictiv, teste psihologice
Introducere
Fenomenul de restricie a amplitudinii
apare atunci cnd probabilitatea de selecie nu
este aceeai pentru fiecare dintre indivizii care
compun populaia, ceea ce nseamn c valorile
msurate pe eantion au o variabilitate limitat
(Aguinis, 1995). Este un lucru ndeobte cunoscut c mrimea coeficientului de corelaie Pearson depinde de dispersia variabilelor analizate.
La limit, dac presupunem, de exemplu, c una

sau ambele variabile, au o dispersie egal cu
zero, corelaia lor va fi i ea egal cu zero. Atunci cnd calculm corelaia ne ateptm ca variabilele msurate pe eantion s capteze ct mai
mult din variabilitatea populaiei. Dac aceast
presupunere nu se susine, iar dispersia
eantionului este mai mic dect dispersia populaiei de referin, atunci valoarea calculat a
lui r este mai mic dect valoarea real acestuia.
1
Universitatea din Bucureti.
Adresa de coresponden: popamarian@gmail.com
91
4/17/2012
1:09 PM
Page 92
Restricia de amplitudine, o ameninare ascuns la adresa validitii de criteriu
Acest fenomen, denumit restricie de amplitudine, a fost semnalat nc din anul 1903, de
ctre Karl Pearson, dar efectele lui sunt nc
destul de frecvent ignorate n analiza statistic a
datelor de cercetare (Raju & Brand, 2003).
Restricia de amplitudine i manifest
efectele negative nu doar n cazul coeficientului
de corelaie Pearson, ci i n contextul altor
statistici bazate pe modelul liniar: analiza de
regresie, analiza factorial, testele t pentru diferena dintre grupuri, analiza de varian. Aceste
proceduri pot fi utilizate, fie pentru analiza unor
date de cercetare, fie n contextul unor examene
de selecie psihologic. De asemenea, restricia
de amplitudine poate fi indus i de utilizarea
unor instrumente psihologice a cror adaptare
pentru cultura romneasc nu este adecvat. n
acest scop, n procesul de adaptare se impune
acordarea unei atenii speciale analizei variabilitii scorurilor obinute pe populaia romneasc, comparativ cu variabilitatea obinut pe
populaia pe care a fost creat instrumentul.
Ceea ce ne propunem n acest articol este
analizarea surselor restriciei de amplitudine,
evaluarea efectelor acesteia i descrierea unor
soluii pentru corecia valorilor coeficienilor de
corelaie afectai de restricia de amplitudine. n
final, vom aborda i problema dihotomizrii artificiale a criteriului, care reprezint o form specific de manifestare a restriciei de amplitudine.
Surse ale restriciei de amplitudine

Restricia de amplitudine este determinat
de orice factor care limiteaz n mod artificial
dispersia variabilelor msurate la nivelul
eantionului, n raport cu dispersia adevrat a
populaiei. Identificarea prezenei fenomenului
se face prin compararea celor dou dispersii.
Acest lucru presupune, desigur, ca ele s fie
disponibile, ceea ce n practic ridic unele
probleme, la care ne vom referi mai trziu. O
alt soluie de identificare const n utilizarea
unei msuri absolute a variabilitii, aa cum
este amplitudinea absolut (diferena dintre valoarea cea mai mare i valoarea cea mai mic a
variabilei) (Jex, 2002). De exemplu, dac avem
un chestionar cu 50 de ntrebri, cu doar dou
92
variante de rspuns (codificate cu 0 i 1), teoretic, scorurile ar trebui s se distribuie ntre 0 i

50. Dac scorurile obinute sunt cuprinse doar
ntre 0 i 30, atunci este posibil s avem o problem de restricie de amplitudine.
Analiznd fenomenul n contextul seleciei
psihologice, Thorndike (apud Aguinis, Henle, &
Ostroff, 2001) a definit trei cazuri tipice de
restricie de amplitudine:
Cazul I se refer la situaia n care restricia
afecteaz ambele variabile, dar avem informaii
cu privire la abaterea standard a lotului
restricionat i a celui nerestricionat numai pentru criteriu, nu i pentru predictor. Acest caz ar
corespunde unei proceduri de validare concurent n care, pentru valorile criteriului avem
date pentru un numr mai mare de angajai, n
timp ce pentru predictor avem valori doar pentru eantionul de validare.
Cazul II se refer la situaia specific procedurii de validare predictiv n care avem
abaterea standard a predictorului, att pentru
lotul nerestricionat (toi candidaii) ct i pentru
lotul restricionat (candidaii selecionai), dar
pentru criteriu avem doar abaterea standard a
lotului restricionat.
Cazul III corespunde situaiei n care avem
restricie de amplitudine att la nivelul predictorului ct i al criteriului, dar aceasta este
indus prin corelaia lor cu o alt(alte) variabil(e), de cele mai multe ori nespecificabile.
n literatura de specialitate, Cazurile I i II
sunt consacrate sub denumirea de restricie
direct, iar Cazul III, sub denumirea de restricie
indirect (Dunbar & Linn, 1991; Sackett &
Lievens, 2008). Dei, n general, restricia
direct este creditat cu cea mai mare influen
asupra validitii, restricia indirect nu trebuie
nici ea subestimat, contribuind ntr-o msur
mai mare la diminuarea validitii dect cea
direct (Schmidt & Shaffer, 2008). Figura 1 permite comparaia ntre graficul scatter-plot al
unei corelaii nerestricionate (a) dintre aptitudine (predictor) i performan (criteriu), cu
situaiile n care sursa restriciei este predictorul
(b) sau criteriul (c).
Borack (1994) scoate n eviden faptul c
restricia direct poate afecta n mod distinct,
4/17/2012
1:09 PM
Page 93
Figura 1. Diagrame scatter-plot, pentru corelaia nerestricionat i restricionate la nivelul variabilei

X (aptitudine) i Y (performan)
Lot nerestricionat
Restricie n X (predictor)
Restricie n Y (criteriu)
Figura 2. Restricia de amplitudine direct dubl, prin selecie i eliminare
att predictorul ct i criteriul. De exemplu, n

cazul validrii predictive dispersia valorilor testului este limitat prin decizia de angajare, iar n
cazul valorilor performanei profesionale dispersia poate fi limitat prin decizii de eliminare
a unor angajai nainte ca acetia s ajung s fie
evaluai sub aspectul criteriului (Figura 2).
n aceste condiii, coeficientul de validitate
este afectat de o dubl restricie direct, fiind
calculat doar pentru subiecii care au fost admii
i nu au fost exclui din procesul de munc
nainte de msurarea criteriului. Dac suprapunem mintal graficele din figurile 1b i 1c vom
nelege c efectul restriciei directe cumulate
este mai mare dect restricia simpl.
Restricia de amplitudine se poate manifesta n studii corelaionale i atunci cnd variabilele sunt msurate pe grupuri pariale, omogene, care compun un lot de subieci mai mare.
Am ilustrat aceast situaie n Figura 3, care
prezint graficele scatter-plot i corelaiile aferente (ipotetice) dintre salariu i satisfacia n
Figura 3. Restricie de amplitudine generat de nivelul de analiz
a) Corelaia salariu/satisfacie la nivelul compartimentelor

unei organizaii
b) Corelaia salariu/satisfacie
pe ansamblul organizaiei
93
4/17/2012
1:09 PM
Page 94
munc, la nivelul mai multor compartimente ale

unei organizaii (a) i la nivelul ntregii organizaii (b). Aa cum se poate observa, putem
obine corelaii mici pe grupuri izolate, afectate
de restricia de amplitudine, dar o corelaie substanial mai mare atunci cnd corelaia este calculat pe ntreaga organizaie.
Efecte ale restriciei de amplitudine

n principiu, restricia de amplitudine are ca
efect micorarea artificial a indicatorului de
asociere (Urbina, 2004). Acest efect a fost analizat mai ales n contextul validitii seleciei psihologice. Carreta i Ree (2003) evoc un studiu
efectuat de Thorndike, n anul 1944, n legtur
cu validitatea seleciei piloilor militari. Un
numr de 1036 candidai piloi din U.S. Army
Air Corps au fost admii n procesul de instruire, indiferent de scorul obinut la cinci teste de
aptitudini. Ulterior, au fost calculai coeficienii
de validitate pentru rezultatele la testele psihologice n raport cu un criteriu de performan n
instruire, pentru toi cei 1036 candidai i separat, pentru 136 de candidai care ar fi fost admii
dac s-ar fi aplicat standardul de selecie existent n uz. Scorul compozit rezultat din cele
cinci teste psihologice a obinut o corelaie de
0.64 pentru ntregul lot evaluat i de numai 0.18
pentru lotul care ar fi putut fi selectat conform
standardului existent. Diferena este imputabil
diminurii dispersiei predictorului n cazul lotului restricionat.
Dar efectele restriciei de amplitudine sunt
departe de a se limita la reducerea nivelului
validitii. Aceasta afecteaz rezultatele tuturor
procedurilor statistice care se bazeaz pe
asocierea variabilelor cantitative: analiza factorial, modelarea ecuaiei de structur (Kline,
2011; Muthn & Hsu, 1993), analiza relaiei de
mediere (Frazier, Tix, & Barron, 2004). Mai
mult dect att, restricia de amplitudine tinde s
subestimeze i diferenele dintre grupuri, ceea
ce afecteaz puterea testului t pentru diferena
dintre medii (Bobko, Roth, & Bobko, 2001), dar
i nivelul validitii examenelor de selecie,
atunci cnd candidaii aparin unor grupuri
diferite (gen, etnie, ras). Fenomenul este
94
cunoscut sub denumirea de dilema diversitatevaliditate: restricia de amplitudine face dificil

maximizarea validitii concomitent cu
realizarea unei diversiti ct mai mari la angajare, sub aspectul categoriilor n care se grupeaz candidaii (Ployhart & Holtz, 2008;
Pyburn, Ployhart, & Kravitz, 2008).
Corecii ale restriciei

de amplitudine
Soluii de corecie pentru restricia de
amplitudine au fost propuse nc din primele
decenii ale sec. XX de ctre Kelley, Otis, Gullikson, Thorndike .a. (apud, Alexander, Carson,
Alliger, & Barrett, 1984; Dunbar & Linn, 1991;
Mumford & Mendoza, 1983). Discuiile cu privire la formulele de corecie nu sunt nici n
prezent epuizate, avnd n vedere problemele pe
care le ridic necesitatea de a limita erorile de
corecie n diverse situaii: nclcarea condiiei
de normalitate, restricia direct cumulat,
prezena restriciei indirecte etc. (Alexander et
al., 1984; Mumford & Mendoza, 1983; Sackett,
Lievens, Berry, & Landers, 2007; Wiberg &
Sundstrm, 2009). O analiz extensiv a acestei
problematici nu face obiectul acestui articol.
n principiu, formulele propuse de diveri
autori se bazeaz pe coeficientul de corelaie
necorectat (observat) i abaterile standard sau
dispersiile, pentru lotul restricionat i nerestricionat. Aspectul cel mai delicat l constituie,
evident, faptul c n condiiile Cazului I nu
cunoatem abaterea standard a predictorului (X)
pentru lotul nerestricionat. n aceast situaie, o
soluie acceptabil este aceea de a utiliza
abaterea standard a lotului pe care a fost realizat etalonarea instrumentului de msur
respectiv, de preferin la nivel naional. Sackett
i Ostgaard (1994) au analizat abaterea standard
a scorului la testul de abiliti cognitive Wonderlic Personnel Test pentru 80 de profesii diferite,
comparativ cu un eantion mare, compus din
subieci de toate profesiile. Rezultatele au artat
c, n medie, abaterea standard pentru profesii
distincte (loturi restricionate) a fost cu 10% mai
mic dect pentru lotul general (nerestricionat).
Mai mult, 90% dintre abaterile standard
4/17/2012
1:09 PM
Page 95
Pentru ilustrare, am calculat corelaia ctorva scale ale chestionarului Armstrong Laboratory Personality Survey (ALAPS) (Popa, 2002;
Retzlaff, 2002; Retzlaff, King, McGlohn, &
Callister, 1996), cu o serie de indicatori ai performanei adaptative, evaluai prin intermediul
unui chestionar sociometric. Tabelul 1 prezint
rezultatele coreciei pentru restricia de amplitudine, calculat dup aplicarea prealabil a coreciei de atenuare pentru fidelitatea msurrii*.
Valorile din Tabelul 1 indic un impact
important al celor dou corecii succesive
asupra coeficientului de validitate. De exemplu,
n cazul corelaiei dintre depresie i preferinele exprimate, corelaia iniial, necorectat,
este r=-0.186, care crete la r=-0.212 dup
corecia de atenuare pentru fidelitate, i mai
departe, la r=-0.289, dup corecia pentru
restricia de amplitudine. Pe ansamblu, dup
aplicarea celor dou corecii succesive, coeficienii din Tabelul 1 prezint creteri care merg
de la 33% pn la 56% din valoarea lor iniial.
Pentru comparaie, evocm un studiu efectuat
de Campbell et al. (2010), cu privire la validitatea predictorilor de personalitate n raport cu
succesul n instruirea aviatorilor, care au constatat c nevrotismul, anxietatea i extraversia
apar cel mai frecvent ca fiind statistic semnificative. n urma coreciei pentru atenuarea de
fidelitate i restricia de amplitudine, cea mai
mare cretere a fost obinut pentru nevrotism
(r=-0.15/rc=-0.25). Gradul de corecie este cu
att mai mare cu ct fidelitatea este mai mic i
restricionate sunt mai mici cu 20% fa de lotul

general. La o concluzie similar au ajuns i
Lang et al. (2010), care au analizat rezultatele la
dou teste de abilitii cognitive, pentru 8276
candidai la nou profesii, n cadrul unor organizaii guvernamentale din Germania. Rezultatele
au artat c abaterile standard pentru fiecare
context profesional specific sunt mai mici, n
medie, cu 10-12% dect abaterea standard a
lotului naional de etalonare, i cu 4-6% mai
mici dect abaterea standard a lotului integral de
candidai. Ambele studii susin concluzia c
nivelul calculat al validitii pentru testele utilizate, este mai mic dect cel real, fapt care justific aplicarea unor corecii.
n continuare, vom prezenta cteva soluii
de corecie, pe baza unora dintre cele mai uzuale
formule propuse n literatura de specialitate.
Pentru corecia restriciei directe este vizat
mai ales situaia n care avem varianele predictorului, att pentru lotul integral, ct i pentru
cel restricionat prin selecie. n acest caz, una
dintre formulele utilizate este urmtoarea
(Bobko, 2001):
(1)
unde:
rxy=corelaia necorectat
rc=corelaia corectat
sx=abaterea standard a lotului restricionat
sx=abaterea standard a grupului nerestricionat
Tabelul 1. Corecia pentru restricia de amplitudine (dup corecia de atenuare pentru fidelitate)
(N=195)
Depresie
Orientarea pentru risc
Negativism
Preferine exprimate
Preferine primite
Preferine reciproce
Necorectat
-0.186
-0.182
-0.166
Corecia de atenuare
-0.212
-0.208
-0.189
Corecie RA
-0.289
-0.284
-0.259
Necorectat
0.259
0.225
0.235
Corecia de atenuare
0.294
0.255
0.267
Corecie RA
0.347
0.302
0.316
Necorectat
-0.164
-0.188
-0.186
Corecia de atenuare
-0.202
-0.231
-0.229
Corecie RA
-0.229
-0.262
-0.259
* Corecia de atenuare pentru fidelitate nu face obiectul acestui articol, dar pentru detalii se poate consulta: Popa, M.
(2011) Infidelitile coeficientului de fidelitate Cronbach alfa. Psihologia resurselor umane, vol. 9, nr. 1, pp. 85-99.
95
4/17/2012
1:09 PM
Page 96
cu ct dispersia lotului nerestricionat este mai

mare fa de dispersia lotului restricionat*.
Atunci cnd dispersia lotului nerestricionat este
mai mic dect cea a lotului restricionat, coeficientul corectat este mai mic dect cel necorectat. Acest lucru este imposibil n cazul seleciei
pe baza unui predictor unic, dar posibil atunci
cnd sunt utilizai mai muli predictori.
n exemplul nostru, am aplicat mai nti
corecia pentru fidelitate i apoi pe cea pentru
restricia de amplitudine. n practic, ordinea de
aplicare a coreciilor trebuie s se supun unei
anumite fundamentri, deoarece cele dou formule de corecie au fost create independent una
de alta. Regula de principiu se bazeaz pe natura estimrii fidelitii (calculat pe eantion
restricionat sau nerestricionat). Ca urmare,
atunci cnd coeficientul de fidelitate este el
nsui afectat de restricie, se efectueaz mai
nti corecia pentru fidelitate, altfel, prima este
corecia pentru restricia de amplitudine. Stauffer i Mendoza (2001) adopt o poziie critic
fa de acest raionament, susinnd c fundamentarea secvenei de corecie este impus de
natura restriciei de amplitudine i nu de natura
coeficientului de fidelitate. Ei consider c regula de mai sus se susine numai atunci cnd
corecia nu include chiar variabila dup care s-a
fcut selecia eantionului. Ca urmare, cei doi
autori susin primatul coreciei pentru fidelitate
numai atunci cnd corecia pentru restricie are
nevoie de rezultatul acesteia. n caz contrar,
corecia va ncepe cu restricia de amplitudine,
urmat de cea pentru fidelitate.
Procedura aplicat pentru obinerea valorilor din Tabelul 1 s-a bazat pe coreciile predictorului, situaie specific pentru Cazul II de
restricie, cel mai frecvent ntlnit n psihologia
industrial-organizaional. n condiiile acestui
caz cunoatem fidelitatea predictorului (calculat pe lotul nerestricionat) precum i dispersiile acestuia, pentru lotul nerestricionat i
restricionat. Totui, nivelul coreciei este limitat
de faptul c se ignor efectul de restricie a
amplitudinii asupra criteriului, care vizeaz doar
persoanele angajate pe baza unui proces selec-
tiv. Acest fenomen afecteaz, n primul rnd,

fidelitatea criteriului, a crui subestimare este cu
att mai mare cu este calculat pe un eantion
mai mic de persoane (Ercan, Yazici, Sigirli,
Ediz, & Kan, 2007). n al doilea rnd, se manifest o restricie a valorilor criteriului, care
descrie numai performana caracteristic persoanelor selecionate. Ambele aspecte au, n
ultim instan, un efect suplimentar de reducere a validitii, pe lng cel determinat de
restricia la nivelul predictorului. Acesta este
motivul pentru care corecia validitii ar trebui
s vizeze, simultan, att atenuarea de fidelitate,
ct i restricia de amplitudine a valorilor criteriului (Oswald & Converse, 2005).
Pentru corecia restriciei indirecte, care
n practic se utilizeaz mai rar, una din formulele utilizabile este urmtoarea (Dunbar &
Linn, 1991):
(2)
unde
rxy = corelaia corectat dintre X i Y, n condiiile restriciei impuse de corelaia lor cu Z
r = corelaia necorectat dintre variabilele respective
s2 = dispersia lotului neselecionat
s2 = dispersia lotului selecionat
Corecia efectuat cu formula 2 rezolv
doar restricia de amplitudine simpl, pentru
variabila X. n practic, aa cum am precizat mai
sus, este posibil s existe o restricia dubl, la
nivelul ambelor variabile. n acest caz, abaterile
standard restricionate se abat de la valorile nerestricionate, pentru variabila X, de exemplu,
att n funcie de restricia lui X ct i n funcie
de restricia lui Y. Acelai raionament este valabil concomitent i pentru Y. Pentru a rezolva
aceast problem, Wells i Fruchter (1970) au
elaborat o procedur compus din 7 pai, prin
care, mai nti sunt corectate abaterile standard
n funcie de cele dou surse de restricionare,
iar apoi acestea sunt utilizate n formula de
corecie pentru r. Eficiena acestei proceduri n
* Corecia pentru atenuarea de fidelitate i cea pentru restricia de amplitudine pot fi aplicate i independent una de
cealalt, dar utilizarea lor succesiv ne apropie mai mult de valoarea adevrat a corelaiei.
96
4/17/2012
1:09 PM
Page 97
raport cu formula clasic a fost analizat de

Alexander et al. (1984), care au ajuns la concluzia c rezultatele coreciei pentru ambele
variabile sunt superioare coreciei simple. Dezavantajul acestei soluii const ns n complexitatea calculelor, ceea ce o face dificil de aplicat
n absena unui program adecvat.
Revenind la corecia simpl, Sackett et al.
(2007) semnaleaz faptul c aplicarea coreciei
pentru restricie direct n condiiile prezenei
suplimentare a unei restricii indirecte, determin o subevaluare a validitii, uneori ntr-o
proporie de pn la 25% din valoarea adevrat a acesteia. Efectul de subestimare poate
fi chiar mult mai mare n cazul validrii cu mai
muli predictori coliniari (care coreleaz ntre
ei) sau atunci cnd se opereaz corecia pentru
restricie direct, dei restricia indirect este
mai important.
Formulele de corecie pentru restricia de
amplitudine funcioneaz corect atunci cnd
sunt ntrunite dou condiii: (i) liniaritatea
relaiei dintre predictor i criteriu (aceeai pant
de regresie pentru lotul restricionat i lotul nerestricionat); (ii) homoscedasticitatea regresiei
(aceeai varian a erorii n jurul dreptei de
regresie), att pentru lotul restricionat ct i
pentru lotul restricionat. Totui, se consider c
formula de corecie este robust n raport cu
nclcarea condiiei de homoscedasticitate, n
timp ce violarea condiiei de liniaritate tinde s
supraestimeze sau s subestimeze corecia coeficientului de validitate (Bobko, 2001).
Corecia pentru restricia de amplitudine
reduce numai o parte din eroarea de estimare a
validitii, valoarea corectat fiind ntotdeauna
mai mic dect cea real. Acest fapt este
determinat, fie de dificultatea coreciei complete
pentru restricia indirect, fie de alte cauze ale
restriciei pe care nu le putem controla. De
exemplu, n cazul criteriului, ca urmare a distorsiunilor de evaluare prin efectul de indulgen/severitate sau efectul tendinei centrale
(Landy & Conte, 2009; Muchinsky, 2006;
Pitariu, 2003).
Nu trebuie s omitem nici faptul c, devenind mai mici prin restricionare, coeficienii de
validitate ating mai greu pragul de semnificaie

statistic, ceea ce nseamn o reducere a puterii,
situaie care reclam un volum mai mare al
eantionului. De exemplu, pentru un coeficient
de validitate r=0.50, obinut pe un lot nerestricionat, este suficient un eantion de 38 de
subieci, pentru o putere statistic de 0.90. Pentru a conserva aceast putere n cazul unui
eantion restricionat cu 30%, volumul
eantionului trebuie s creasc la minimum 125
de subieci (Retzlaff, 2002). Totui, se apreciaz
c n cazul restriciei indirecte volumul
eantionului necesar atingerii pragului semnificaiei statistice este sensibil mai mic dect n
cazul restriciei directe (Sackett & Wade, apud
Retzlaff et al., 1996).
n paragraful anterior am evocat problema
semnificaiei statistice a coeficientului de corelaie (validitate) restricionat. Din pcate, dac
valoarea necorectat este statistic semnificativ
aceasta nu nseamn c i valoarea corectat i
pstreaz aceast proprietate. Motivul const n
modificarea caracteristicilor distribuiei de nul
pentru coeficientul corectat, ceea ce, implicit,
impune modificarea modului de calcul al erorii
standard a acestuia. Raju i Brand (2003) au
propus formule specifice pentru estimarea erorii
standard n cazul coeficienilor corectai pentru
fidelitate, pentru restricie, sau pentru ambele
situaii, cu ajutorul crora se poate calcula un
test Z pentru coeficienii corectai*.
Dihotomizarea, o form particular

a restriciei de amplitudine
Exist situaii n care validarea se bazeaz
pe un criteriu valorizat dihotomic, de exemplu,
admis/respins la examenul de finalizare a unui
program de instruire; succes/eec n dobndirea
licenei profesionale etc. n aceste exemple
dihotomizarea a criteriului reflect starea real
sub care se prezint informaia. Uneori ns,
exist tentaia de a dihotomiza o variabil (criteriul) msurat pe scal cantitativ, cu sperana
c prin compararea unor grupuri extreme, fie se
* Raionamentele i calculele sunt relativ laborioase pentru a fi reproduse aici, dar pot fi consultate n sursa citat.
97
4/17/2012
1:09 PM
Page 98
amplific mrimea efectului i, implicit, puterea

testului de semnificaie, fie se obine o simplificare a procesului de analiz statistic i de
prezentare a rezultatelor. n realitate, aceste
ateptri nu au un suport statistic real, ci dimpotriv. Clasificarea forat a unor valori continui n categorii dihotomice are efecte negative
importante pe mai multe planuri: pierderea de
informaie individual, prin clasificarea n
aceeai categorie a unor subieci cu valori sensibil diferite; diminuarea mrimii efectului i a
puterii testului statistic; riscul de mascare a unor
posibile asocieri de tip neliniar ntre variabile;
diminuarea fidelitii msurrii (Borack, 1994;
Salgado, Viswesvaran, & Ones, 2005; Schmidt
& Hunter, 2005).
Este important s precizm faptul c atunci
cnd una dintre variabile este exprimat cantitativ, iar cealalt este o dihotomie, nu se calculeaz coeficientul de corelaie Pearson (r), ci
coeficientul de corelaie punct-biserial (rpb)*.
Valoarea acestuia nu depinde doar de intensitatea relaiei dintre cele dou variabile, ci i de
proporia cazurilor (subiecilor) inclui n cele
dou grupuri dihotomice. Astfel, coeficientul
punct-biserial poate lua valoarea maxim numai
atunci cnd grupurile n care au fost mprite
valorile variabilei dihotomice sunt egale (50%50%). Chiar i aa, valoarea sa maxim este
doar o fraciune (0.798) din valoarea maxim
teoretic (1) pe care ar putea s o aib coeficientul de corelaie Pearson, dac variabila nu
ar fi dihotomizat. Cu ct dihotomizarea se
bazeaz pe o valoare care se ndeprteaz mai
mult de medie (presupunnd c distribuia criteriului este normal), cu att abaterea (e) a coe-
ficientului punct-biserial (rpb) de la valoarea

coeficientului bivariat (r) este mai mare. Transpus n formula (3) aceast afirmaie ne sugereaz c valoarea observat a coeficientului de
corelaie cu o variabil dihotomizat se ndeprteaz de valoarea corelaiei bivariate, cu att
mai mult ct eroarea este mai mare:
(3)
Din aceast expresie este dedus formula

de principiu a coreciei pentru dihotomizare:
(4)
Cohen (apud Salgado et al., 2005) a propus

o formul pentru calcularea erorii de estimare,
care a deschis calea coreciei pentru dihotomizare:
(5)
unde:
h = ordonata (nlimea) corespunztoare valorii
standard care separ cele dou grupuri, pe distribuia normal
p = proporia valorilor din oricare dintre cele
dou categorii rezultate prin dihotomizare
De exemplu, n cazul dihotomizrii la
medie, p=0.50, h=0.3989, iar valoarea lui e calculat cu formula (5) este 0.798, care este exact
limita teoretic pe care o poate atinge rpb, evocat anterior. Aceasta nsemn, de exemplu, c
pentru un coeficient de corelaie rpb=0.30, obinut cu criteriu dihotomizat n grupuri egale, valoarea corectat a corelaiei bivariate ar fi
r=0.30/0.798=0.37. n Tabelul 2 exemplificm
Tabelul 2. Coeficienii de corelaie dintre un predictor cu valori continui i criteriul cu valori continui
i dihotomizat n proporii diferite (N=474)
Criteriu (Y)
Dihotomizat (%)
Nedihotomizat
0.880**
50-50
40-60
30-70
20-80
10-90
0.519**
0.467**
0.410**
0.339**
0.257**
** Statistic semnificativ la pragul p<0.001

* SPSS calculeaz automat coeficientul de corelaie punct-biserial n locul celui bivariat (Pearson), atunci cnd valorile
unei variabile sunt de tip continuu (cantitative), iar ale celeilalte sunt dihotomice, codificate cu 0 i 1.
98
4/17/2012
1:10 PM
Page 99
efectul dihotomizrii asupra corelaiei dintre

dou variabile cantitative, ambele cu distribuii
asimetrice, dintre care una a fost dihotomizat n
grupuri cu proporii diferite.
Rezultatele arat c, pornind de la o corelaie r=0.88 ntre predictor i criteriul nedihotomizat, obinem prin dihotomizarea criteriului
corelaii din ce n ce mai mici, pe msur ce proporiile grupurilor sunt tot mai dezechilibrate.
Efectele negative sunt aceleai, indiferent
dac se dihotomizeaz una sau alta dintre variabile (criteriul sau predictorul), i se amplific
atunci cnd sunt dihotomizate simultan ambele
variabile (Borack, 1994; Campbell et al., 2010).
Astfel, dac dihotimizarea are loc att pentru
ambele variabile n aceeai proporie, subestimarea valorii corelaiei bivariate este mai mare
(de exemplu: r=0.30/(0.798*0.798)=0.47). n
mod obinuit, coeficienii de corelaie cu variabile dihotomizate (rpb) sunt mai mici dect cei
pentru valori continui (r), dar este posibil uneori
ca, accidental, acetia s fie mai mari, ca efect al
erorii de eantionare sau al unor anomalii ale
distribuiei variabilei continue (Borack, 1994).
Aplicarea procedurii de corecie pentru
dihotomizare nu este lipsit de controverse. Este
de precizat mai ales faptul c aceasta este recomandabil numai n cazul dihotomizrii artificiale a criteriului, nu i atunci cnd dihotomia
este natural (de ex., oferi care au avut accident/oferi care nu au avut accident) (Carretta
& Ree, 2001). Aspectul cel mai controversat
ns, rmne tentaia de a dihotomiza variabilele
msurate cantitativ, tentaie care, n mod normal, ar trebui evitat n practica de cercetare,
fiind declarat de ctre Royston et al. (2006)
drept o idee rea.
Concluzii
Restricia de amplitudine este una dintre
cauzele cele mai frecvente ale diminurii coeficienilor de corelaie n general, i a coeficienilor de validitate, n particular. Efectele de
subestimare a intensitii asocierii dintre variabile conduc la erori n interpretarea rezultatelor, i pot avea un impact negativ asupra
unor decizii practice. n cazul studiilor de vali-
dare de criteriu, de exemplu, restricia de amplitudine mascheaz o parte destul de important a

capacitii de predicie a testelor psihologice.
Este suficient s amintim aici doar controversa
legat de validitatea testelor de personalitate,
pentru a nelege ct de important poate fi
luarea n considerare a problemei restriciei de
amplitudine (Barrick, Mount, & Judge, 2001;
Campbell et al., 2010; Hogan, 1998; Hurtz &
Donovan, 2000; Leaetta & Frederick, 2000;
Morgeson et al., 2007; Nowack, 1997; Salgado
et al., 2005; Whetzel, McDaniel, Yost, & Kim,
2010). Atunci cnd avem dovezi ale prezenei
acesteia, aplicarea unei corecii adecvate reprezint o soluie util i necesar.
n conformitate cu standardele i recomandrile tiinifice uzuale (Judge, Klinger, Simon,
& Yang, 2008; SIOP, 2003), studiile de validare
trebuie s analizeze i s identifice artefactele i
sursele de contaminare care pot afecta coeficienii de validitate. Efectuarea tuturor coreciilor este legitim, dar trebuie nsoit de
furnizarea informaiilor complete cu privire la
procedurile utilizate, iar valorile corectate vor fi
nsoite ntotdeauna de valorile necorectate. De
asemenea, se recomand calcularea i publicarea erorii standard a coeficienilor corectai,
iar pe baza lor, a limitelor intervalului de
ncredere. n fine, generalizarea validitii
reprezint o modalitate promitoare prin care
variaiile locale ale validitii pot fi atenuate
prin studii de meta-analiz efectuate pe un
numr mare de studii.
Bibliografie
Aguinis, H. (1995). Statistical power problems with
moderated multiple regression in management
research. Journal of Management Research, 21,
1141-1158.
Aguinis, H., Henle, C. A., & Ostroff, C. (2001). Measurement in Work and Organizational Psychology.
In N. Anderson, D. S. Ones, H. K. Sinangil & C.
Viswesvaran (Eds.), Handbook of Industrial, Work
and Organizational Psychology (Vol. 1 Personnel Psychology, pp. 27-50).
Alexander, R. A., Carson, K. P., Alliger, G. M., & Barrett, G. V. (1984). Correction for Restriction of
Range when Both X and Y are Truncated. Applied
Psychological Measurement, 8(2), 231-241.
99
4/17/2012
1:10 PM
Page 100
Barrick, M. R., Mount, M. K., & Judge, T. A. (2001).

Personality and Performance at the Beginning of
the New Millennium: What Do We Know and
Where Do We Go Next? Personality and Performance, 9(1/2), 9-30.
Bobko, P. (2001). Correlation and Regression. Applications for Industrial Organizational Psychology
and Management (Second ed.). Thousand Oaks:
Sage Publications.
Bobko, P., Roth, P. L., & Bobko, C. (2001). Correcting
the Effect Size of d for Range Restriction and
Unreliability. Organizational Research Methods,
4(1), 46-61.
Borack, J. I. (1994). Estimating Predictive Validity
When Restriction Due To Range Restriction and
Attrition Is Present. Military Psychology, 6(3),
193-204.
Campbell, J. S., Castaneda, M., & Pulos, S. (2010).
Meta-Analysis of Personality Assessments as Predictors of Military Aviation Training Success.
International Journal of Aviation Psychology,
20(1), 92-109. doi: 10.1080/10508410903415872
Carretta, T. R., & Ree, M. J. (2001). Pitfalls of Ability
Research. International Journal of Selection and
Assessment, 9(4), 325-335.
Carretta, T. R., & Ree, M. J. (2003). Pilot Selection
Methods In P. S. Tsang & M. A. Vidulich (Eds.),
Principles and Practice of Aviation Psychology
(pp. 357-396). London: Lawrence Erlbaum Associates, Publishers.
Dunbar, S. B., & Linn, R. L. (1991). Range Restriction
Adjustments in the Prediction of Military Job Performance. In A. K. Wigdor & B. F. Green Jr.
(Eds.), Performance Assessment for the Workplace
Technical issues (Vol. II, pp. 127-157). Washington D.C.
Ercan, I., Yazici, B., Sigirli, D., Ediz, B., & Kan, I.
(2007). Examining Cronbach Alpha, Theta,
Omega Reliability Coefficients According to the
Sample Size. Journal of Modern Applied Statistical Methods, 6(1), 291-303.
Frazier, P. A., Tix, A. P., & Barron, K. E. (2004). Testing Moderator and Mediator Effects in Counseling
Psychology Research. Journal of Counseling Psychology, 51(1), 115-134.
Hogan, J. (1998). Personality and Job Performance.
Human Performance, 11(2/3), 125.
Hurtz, G. M., & Donovan, J. J. (2000). Personality and
Job Performance: The Big Five Revisited. Journal
of Applied Psychology, 85(6), 869-879.
Jex, S. M. (2002). Organizational psychology : a scientist-practitioner approach. New York: John Wiley
& Sons.
Judge, T. A., Klinger, R., Simon, L. S., & Yang, I. W. F.
(2008). The Contributions of Personality to Organizational Behavior and Psychology: Findings,
Criticisms, and Future Research Directions. Social
100
and Personality Psychology Compass 2(5), 19822000.

Kline, R. B. (2011). Principles and Practice of Structural Equation Modeling (Third ed.). New York:
The Guilford Press.
Landy, F. J., & Conte, J. M. (2009). Work in the 21st
Century: An Introduction to Industrial and Organizational Psychology (Third ed.): Wiley-Blackwell.
Lang, J. W. B., Kersting, M., & Hlsheger, U. R.
(2010). Range Shrinkage of Cognitive Ability Test
Scores in Applicant Pools for German Governmental Jobs: Implications for range restriction corrections. International Journal of Selection and
Assessment, 18(3), 321-327.
Leaetta, M. H., & Frederick, L. O. (2000). Personnel
Selection: Looking Toward the Future Remembering the Past. Annual Review of Psychology, 51,
631-666.
Morgeson, F. P., Campion, M. A., Dipboye, R. L., Hollenbeck, J. R., Murphy, K., & Schmitt, N. (2007).
Reconsidering the Use of Personality Tests in Personnel Selection Contexts. Personnel Psychology,
60(3),
683-729.
doi:
10.1111/j.17446570.2007.00089.x
Muchinsky, P. (2006). Psychology Applied To Work
(Eight ed.): Thomson Learning Inc.
Mumford, M. D., & Mendoza, J. e. L. (1983, March,
23-26). Range Restriction and Attenuation Corrections. Paper presented at the Annual Meeting of
the Southeastern Psychological Association,
Atlanta, GA.
Muthn, B. O., & Hsu, J.-W. Y. (1993). Selection and
predictive validity with latent variable structures.
Brilish Journal of Mathematical and Statistical
Nowack, K. (1997). Personality Inventories: The Next
Generation. Performance in Practice. American
Society of Training and Development (Winter
1996/1997).
Oswald, F. L., & Converse, P. D. (2005). Correcting for
Reliability and Range-Restriction in Meta-Analysis. Paper presented at the 20th Annual Conference
for the Society for Industrial and Organizational
Psychology, Los Angeles, CA.
Pitariu, H. (2003). Proiectarea fielor de post, evaluarea muncii i aprecierea personalului. Bucureti:
Irecson.
Ployhart, R. E., & Holtz, B. C. (2008). The
diversityvalidity dilemma: strategies for reducing
racioethnic and sex subgroup differences and
adverse impact in selection. Personnel Psychology, 61, 153172.
Popa, M. (2002). Armstrong Laboratory Aviation Personality Survey Inventory (ALAPS); Romanian
Release. Paper presented at the 25-th EAAP Conference, Warsaw.
4/17/2012
1:10 PM
Page 101
Pyburn, K. M., Ployhart, R. E., & Kravitz, D. A. (2008).

The diversityvalidity dilemma: overview and
legal context. Personnel Psychology, 61(1), 143151. doi: 10.1111/j.1744-6570.2008.00108.x
Raju, N. S., & Brand, P. A. (2003). Determining the significance of correlations corrected for unreliability
and range restriction. Applied Psychological Measurement, 27(1), 52-71.
Retzlaff, P. D. (2002). The Armstrong Laboratory Aviation Personality Survey: Development, Norming,
and Validation. Military Medicine, 167(12).
Retzlaff, P. D., King, R. E., McGlohn, S. E., & Callister, J. D. (1996). The Development of the Armstrong Laboratory Aviation Personality Survey
(ALAPS), : Interim Technical Report, Aerospace
Medicine Directorate, Clinical Sciences Division,
Brooks Air Force Base, TX.
Royston, P., Altman, D. G., & Sauerbrei, W. (2006).
Dichotomizing continuous predictors in multiple
regression: a bad idea. Statistics in Medicine, 25,
127-141.
Sackett, P. R., & Lievens, F. (2008). Personnel Selection. Annual Review of Psychology, 59:16.116.32
(http://users.ugent.be/~flievens/ar.pdf).
Sackett, P. R., Lievens, F., Berry, C. M., & Landers, R.
N. (2007). A Cautionary Note on the Effects of
Range Restriction on Predictor Intercorrelations.
Journal of Applied Psychology 92(2), 538544.
Sackett, P. R., & Ostgaard, D. J. (1994). Job-specific
applicant pools and national norms for cognitive
ability tests: Implications for range restriction corrections in validation research. Journal of Applied
Psychology, 79(5), 680-684.
Salgado, J. F., Viswesvaran, C., & Ones, D. S. (2005).
Predictors Used for Personnel Selection: An
Overview of Constructs, Methods and Techniques. In N. Anderson, D. S. Ones, H. K.
Sinangil & C. Viswesvaran (Eds.), Handbook of
Industrial, Work & Organizational Psychology

(Vol. 1 Personnel Psychology, pp. 165-199).
London: Sage Publishing.
Schmidt, F. L., & Hunter, J. (2005). Meta-Analysis. In
N. Anderson, D. S. Ones, H. K. Sinangil & C.
Viswesvaran (Eds.), Handbook of Industrial, Work
& Organizational Psychology (Vol. 1 Personnel
Psychology, pp. 51-70). London: Sage Publishing.
Schmidt, F. L., & Shaffer, J. A. (2008). Increased Accuracy for Range Restriction Corrections: Implications for the Role of Personality and General Mental Ability In Job and Training Performance.
Personnel Psychology, 61(4), 827868.
SIOP. (2003). Principles for the Validation and Use of
Personnel Selection Procedures (Fourth ed.):
Society for Industrial Organizational Psychology
(http://www.siop.org/_Principles/principles.pdf).
Stauffer, J. M., & Mendoza, J. L. (2001). The Proper
Sequence for Correcting Correlation Coefficients
for Range Restriction and Unreliability. Psychometrika, 66(1), 63-68.
Urbina, S. (2004). Essentials of Psychological Testing.
Hoboken: John Wiley & Sons, Inc.
Wells, D. G., & Fruchter, B. (1970). Correcting the Correlation Coefficient for Explicit Restriction on
Both Variables. Educational and Psychological
Measurement,
30(4),
925-934.
doi:
10.1177/001316447003000412
Whetzel, D. L., McDaniel, M. A., Yost, A. P., & Kim, N.
(2010). Linearity of PersonalityPerformance
Relationships: A large-scale examination. International Journal of Selection and Assessment, 18(3),
310-320. doi: 10.1111/j.1468-2389.2010.00514.x
Wiberg, M., & Sundstrm, A. (2009). A comparison of
two approaches to correction of restriction of range
in correlation analysis. Practical Assessment,
Research & Evaluation (http://pareonline.net/pdf/
v14n5.pdf), 14(5).
101
4/17/2012
1:10 PM
Page 102
Folosirea analizei factoriale exploratorii n cazul itemilor:

Aspecte specifice i recomandri
Paul Srbescu1
Abstract
In Romania, the use of exploratory factor analysis (EFA) in the development or cultural adaptation of psychological tests is a phenomenon that gained a very large ampleness in recent years. Many of the recommendations that apply when using EFA for variables will lead to erroneous results if EFA is used for items. The purpose
of this article is to provide relevant information about issues to be taken into account when EFA is used for items.
Basically, this article presents a series of recommendations about decisions on determining the need to implement
EFA, the sample used, choosing a factor extraction method, selecting the optimal number of factors, factor rotation and identifying false factors. Finally, a synthesis of recommendations made throughout the article is presented, along with the main techniques for forming item parcels and a guide for conducting a parallel analysis using
SPSS syntax.
Keywords: Exploratory factor analysis, item parcels, parallel analysis
Rsum
En Roumanie, lutilisation de lanalyse factorielle exploratoire (EFA) dans le procs de dveloppement ou
dadaptation culturelle des preuves psychologiques reprsente un phnomne qui a pris un grand essor pendant
les dernires annes. Beaucoup de recommandations valables pour lutilisation de la EFA pour les variables vont
mener des rsultats errons dans le cas de lutilisation de la EFA pour les items. Le but de cet article cest de
fournir des informations signifiantes sur les aspects qui doivent tre pris en considration lors quil sagit dutiliser la EFA pour les items. Pratiquement, le prsent article vient illustrer une srie de recommandations lies aux
dcisions concernant la dtermination du besoin dapplication de la EFA, lchantillon utilis, le choix dune
mthode dextraction des facteurs et lidentification des faux facteurs. A la fin, il y a une synthse des recommandations mentionnes tout au long de larticle, en prsentant les principales techniques pour la formation des parcelles ditems aussi bien quun guide pour la ralisation dune analyse parallle dans la syntaxe du logiciel SPSS.
Mots-cls: lanalyse factorielle exploratoire, parcelles ditems, analyse parallle
Rezumat
Utilizarea analizei factoriale exploratorii (eng. exploratory factor analysis sau EFA) n procesul de dezvoltare sau de adaptare cultural a probelor psihologice reprezint un fenomen ce a cptat o foarte mare
amploare n ultimii ani, n Romnia. Multe dintre recomandrile care sunt valabile pentru utilizarea EFA n cazul
variabilelor vor duce la rezultate eronate n cazul utilizri EFA pentru itemi. Scopul acestui articol este de a oferi
informaii relevante despre aspectele care trebuie luate n calcul atunci cnd EFA este utilizat n cazul itemilor.
Practic, acest articol prezint o serie de recomandri legate de deciziile privind determinarea necesitii de a aplica EFA, eantionul utilizat, alegerea unei metode de extracie a factorilor, selectarea numrului optim de factori,
rotirea factorilor i identificarea factorilor fali. La final este realizat o sintez a recomandrilor fcute pe par-
Universitatea de Vest din Timioara.

Adres de coresponden: paul.sarbescu@gmail.com
102
4/17/2012
1:10 PM
Page 103
Folosirea analizei factoriale exploratorii n cazul itemilor: Aspecte specifice i recomandri
cursul articolului, sunt prezentate principalele tehnici pentru formarea pachetelor de itemi, ct i un ghid pentru
realizarea unei analize paralele din sintaxa programului SPSS.
Cuvinte-cheie: Analiza factorial exploratorie, pachete de itemi, analiza paralel
Introducere
Analiza factorial exploratorie (EFA) este
una dintre cele mai utilizate tehnici statistice din
tiinele sociale. Scopul principal al acesteia
este identificarea numrului i a naturii factorilor care stau la baza unui set de variabile manifeste (Little, Cunningham, Shahar i Widaman,
2002; Bandalos, 2008; Sava, 2011).
Cu toate c tehnica a fost dezvoltat acum
mai bine de un secol i a fost aplicat constant
n diverse scopuri (dezvoltarea unor teorii ale
personalitii, construirea unor instrumente psihometrice), muli autori au criticat modalitatea
de aplicare a EFA (Armstrong, 1967; Gould,
1981). De asemenea, unele revizuiri ale
cercetrilor publicate n literatura de specialitate
ce folosesc EFA au scos la iveal carene evidente la nivelul deciziilor luate de autori n
derularea acestei tehnici statistice (Ford, MacCallum i Tait, 1986; Fabrigar, Wegener, MacCallum, i Strahan, 1999).
n cazul utilizrii analizei factoriale exploratorii pentru dezvoltarea/adaptarea cultural a
unor instrumente psihometrice, exist mai multe
aspecte ce trebuie luate n considerare, pentru a
evita obinerea unor rezultate distorsionate. Utilizarea EFA n cazul itemilor n acelai fel ca n
cazul variabilelor poate conduce, destul de des,
la obinerea unor structuri factoriale inadecvate.
Acest lucru se datoreaz, n principal, particularitilor specifice ale itemilor (Gorsuch, 1997).
Articolul de fa cuprinde o trecere n
revist a aspectelor specifice ale folosirii EFA n
cazul itemilor, urmat de o analiz a deciziilor
ce trebuiesc luate n cazul utilizrii EFA cu
itemi. Acest decizii sunt legate de: determinarea
necesitii de a aplica EFA, eantionul utilizat,
alegerea unei metode de extracie a factorilor,
selectarea numrului optim de factori, rotirea
acestora i identificarea factorilor fali. La final
sunt prezentate cteva informaii legate de formarea pachetelor de itemi, iar ncheierea este
format dintr-un set de concluzii i recomandri.
Aspecte specifice ale folosirii analizei factoriale exploratorii n cazul itemilor

Utilizarea EFA n cazul itemilor trebuie
fcut innd cont de faptul c itemii au proprieti diferite fa de scale, i necesit astfel
tehnici specifice, pentru evitarea obinerii unor
rezultate eronate. Prezentm n continuare principalele diferene dintre itemi i scale, ce pot
influena rezultatele unei analize factoriale
exploratorii (Gorsuch, 1997):
a) Itemii au o fidelitate mai sczut dect
scalele; combinarea unui set de itemi cu fidelitate sczut reduce eroarea itemilor prin combinarea varianei comune. Astfel, este mult mai
dificil de lucrat cu corelaiile dintre itemi,
deoarece acestea sunt mai sczute dect corelaiile dintre scale. Din aceast cauz, factorii
sunt mai instabili n cazul utilizrii EFA cu itemi
dect n cazul utilizrii EFA cu scale.
b) Itemii conin deseori varian strin
constructului care este msurat. Aproape orice
item din orice scal conine o fraz sau un
cuvnt care poate fi interpretat de cineva ntr-o
manier idiosincratic. Scalele diminueaz
aceste erori minore prin nsumarea confuziilor
aferente itemilor, astfel normalizndu-le.
c) Distribuiile itemilor difer frecvent
unele de altele, de multe ori n mod intenionat.
Pentru a putea surprinde toate variaiile unui
construct este necesar prezena unor itemi cu
medii care variaz de-a lungul scalei de rspuns.
Itemii cu mediile situate la extreme au, n general, distribuii asimetrice; aceste distribuii influeneaz corelaiile dintre itemi deoarece o corelaie poate fi puternic i pozitiv doar dac
itemii care sunt corelai au aceeai distribuie.
Astfel, prezena diferitelor distribuii reduce
valoarea corelaiilor dintre itemi. Scalele sunt
formate prin nsumarea unor itemi, normaliznd
astfel distribuiile; deci, scalele prezint n general corelaii mai ridicate dect itemii.
d) Rspunsurile la itemi sunt aproape tot
timpul formate dintr-o serie de categorii (dihotomice sau scale de tip Likert cu 5-7 variante), i
103
4/17/2012
1:10 PM
Page 104
foarte rar dintr-o scal de valori continu. Din

cauza faptului c valorile nu au o amplitudine
mai mare (ex. de la 0 la 50), rspunsurile la un
item pot reflecta doar diferene majore ntre
indivizi, privind constructul msurat (ex. chiar
dac 40 de persoane obin valoarea 4 la un anumit item, vor exista anumite diferene ntre ei,
privind constructul msurat, iar aceste diferene
ar putea fi identificate de un alt item). Deoarece
scorurile scalelor reprezint suma itemilor componeni, scalele funcioneaz mai degrab ca o
variabil continu, avnd astfel o mai bun posibilitate de difereniere ntre indivizi.
Din cauza acestor diferene, realizarea unei
analize factoriale exploratorii pe un set de itemi
n acelai fel ca pe un set de variabile va duce,
foarte des, la obinerea unor rezultate distorsionate. Sugestiile din literatura de specialitate pentru folosirea EFA n cazul itemilor (mai ales
atunci cnd distribuiile itemilor deviaz substanial de la normalitate) sunt urmtoarele
(Bernstein i Teng, 1989; Panter, Swygert,
Dahlstrom i Tanaka, 1997; OConnor, 2000):
pentru datele dihotomice, este recomandat
efectuarea analizei factoriale folosind matricea
corelaiilor de tip tetrachoric dintre itemi, n
locul matricii de corelaii Pearson. Pentru itemii
ce folosesc scale de tip Likert, este recomandat
efectuarea analizei factoriale folosind matricea
corelaiilor de tip polychoric dintre itemi, n
locul matricii de corelaii Pearson.
Corelaiile de tip tetrachoric i polychoric se
bazeaz pe presupunerea c toate categoriile de
rspuns (dihotomice sau scal de tip Likert) sunt
de fapt indicatori pentru variabile latente, care au
o distribuie normal. Analizele factoriale a matricelor de corelaie de tip tetrachoric sau polychoric sunt, n esen, analize factoriale a relaiilor dintre variabilele care stau la baza datelor
brute, i care se presupune c sunt continue i au
o distribuie normal (Panter i colab., 1997). Din
pcate, SPSS-ul nu ofer posibilitatea realizrii
EFA asupra unor astfel de matrici de corelaie.
n schimb, alte programe precum FACTOR*
(Lorenzo-Seva i Ferrando, 2006) ofer posibili-
tatea realizrii EFA pe matrici de corelaie de tip

tetrachoric i polychoric.
O alt recomandare pentru situaia n care
avem de-a face cu date dihotomice (dar nu
numai) este formarea pachetelor de itemi. Prin
intermediul acestei metode se grupeaz i se
analizeaz mai muli itemi mpreun, renunndu-se astfel la utilizarea itemului n sine ca unitate de msur (Bandalos, 2008; Sava, 2011).
Mai multe informaii despre situaiile cnd este
recomandat formarea pachetelor de itemi, ct
i despre principalele tehnici de formare a acestora, vor fi oferite ctre finalul acestui articol.
Principalele decizii n cazul folosirii

analizei factoriale exploratorii
Determinarea necesitii
de a aplica EFA
nainte de a trece efectiv la aplicarea EFA,
orice cercettor ar trebui s fie sigur c aceast
tehnic statistic este cea necesar pentru atingerea obiectivelor studiului (Fabrigar i colab.,
1999). Astfel, n cazul dezvoltrii unui nou
instrument psihometric, utilizarea EFA este
necesar pentru a identifica unidimensionalitatea sau multidimensionalitatea setului de itemi
supus analizei, i implicit structura factorial a
probei. De asemenea, n cazul adaptrii culturale a unui test, EFA poate fi utilizat la primele
aplicri ale probei, pentru a verifica n ce
msur structura factorial originar a probei
este reprodus pe populaie romneasc; de asemenea, ulterior aplicrii EFA este necesar i
aplicarea analizei factoriale confirmatorii (eng.
confirmatory factor analysis sau CFA), pentru
confirmarea modelului factorial al probei
respective. n schimb, n cazul n care o prob a
fost adaptat cu succes i utilizat intensiv,
folosirea EFA nu mai are foarte mare relevan,
n aceste situaii fiind mult mai indicat utilizarea CFA, pentru a verifica invariana instrumentului respectiv (Sava, 2011).
* Programul i instruciunile de utilizare sunt disponibile gratuit pe internet la adresa: http://psico.fcep.urv.es/utilitats/

factor/index.html.
104
4/17/2012
1:10 PM
Page 105
Eantionul utilizat
Pentru selectarea unui eantion adecvat trebuie avute n vedere dou aspecte: caracteristicile i volumul su (Gorsuch, 1997). Legat de
caracteristicile eantionului, soluia optim ar
consta n selectarea unui lot de omogenitate
medie, deoarece eantioanele foarte omogene
sau foarte eterogene pot conduce la obinerea
unor soluii factoriale incorecte. De asemenea,
foarte important este i similaritatea dintre
lotul utilizat i populaia int pe care va fi aplicat proba (ex. dac testul va fi aplicat pe femei
de peste 50 de ani, atunci lotul utilizat pentru
dezvoltarea/adaptarea cultural a probei ar trebui s fie format din femei de peste 50 de ani;
dac testul urmeaz s fie aplicat elevilor de
liceu, atunci lotul utilizat ar trebui s fie format
din elevi de liceu). Din pcate, n majoritatea
studiilor de acest gen sunt utilizate eantioane
formate din studeni, n special cei de la tiine
sociale, probabil din cauza gradului mare de
accesibilitate (ITC, 2010). Eantioanele astfel
construite au n general anumite caracteristici
(vrste ntre 19 i 25 de ani, preponderena
genului feminin etc.), care nu sunt ntotdeauna
adecvate pentru dezvoltarea/adaptarea oricrui
tip de instrument psihometric.
Legat de volumul eantionului, trebuie
menionat faptul c orice regul fix nu este valabil i nici foarte folositoare (Gorsuch, 1997;
MacCallum, Widaman, Zhang i Hong, 1999).
Att numrul minim de participani, ct i raportul minim participani-itemi necesare pentru a
asigura o analiz optim variaz n funcie de
anumite caracteristici ale itemilor inclui n
studiu; mai exact, acestea variaz n principal n
funcie de nivelul comunalitii itemilor, care se
afl ntr-o relaie invers proporional cu volumul eantionului. n cazul utilizrii EFA cu variabile (care au de obicei un nivel al comunaliti
mai ridicat dect itemii), soluii factoriale
optime au fost identificate pe eantioane mai
mici de 100 de persoane, avnd un raport participani-variabile de numai 3:1. n cazul utilizrii
EFA cu itemi, sunt n general necesare
eantioane mai mari (minim 100 de participani,
preferabil peste 200), i un raport participaniitemi de cel puin 5:1 (MacCallum i colab.,
1999; de Winter, Dodou i Wieringa, 2009;
Sava, 2011). Evident, dup cum menionam
anterior, aceste indicaii nu trebuie interpretate

ca nite reguli rigide, peste care nu se poate
trece sub nici o form (ex. structura factorial a
unui test de 12 itemi ce msoar 2 faete ale
aceluiai factor, are mari anse s fie reprodus
cu succes chiar i pe un eantion de 70-80 de
participani; de asemenea, structura factorial a
unui test multidimensional de 110 itemi este
foarte probabil s fie identificat pe un lot de
400 de participani, adic la un raport participani-itemi mai mic de 4:1)
Alegerea unei metode de extracie
a factorilor
Una dintre cele mai importante diferene
care trebuie menionat atunci cnd discutm
despre metodele de extracie a factorilor este
aceea dintre analiza componentelor principale
(eng. principal component analysis sau PCA)
i analiza factorilor comuni (AFC; eng. common factor analysis). O prim diferen dintre
cele dou se observ la nivelul modului de analiz a datelor. AFC identific trei surse ale variaiei datelor: factorii comuni, factorii specifici i
erorile de msurare, gruparea ultimelor dou formnd variana unic. n opoziie, PCA nu diferenieaz ntre variana comun i variana unic,
dispersia datelor avnd la baz dou surse: dispersia explicat i cea datorat erorilor de
msurare. De asemenea, teoria de la baza PCA
nu specific existena unor factori comuni,
folosind ca substitut termenul de component
principal (Fabrigar i colab., 1999; Sava, 2011).
O alt diferen dintre cele dou metode este
identificat la nivelul scopului fiecreia. Dac
obiectivul AFC este de a explica corelaiile dintre o serie de variabile manifeste prin determinarea unui numr restrns de factori care s le
influeneze, scopul PCA este de explica ct mai
mult din totalul dispersiei variabilelor manifeste
prin identificarea unui numr restrns de componente principale (Hogarty, 2005; Sava, 2011).
Astfel, avnd n vedere faptul c obiectivul
PCA nu este acela de a identifica factori comuni care s influeneze o serie de variabile manifeste, iar componentele principale nu sunt variabile latente, putem afirm faptul c PCA nu
este o metod de analiz factorial i, n consecin, nu ar trebui utilizat atunci cnd dorim
105
4/17/2012
1:10 PM
Page 106
realizarea unei analize factoriale exploratorii*

(Fabrigar i colab., 1999).
Cele mai utilizate metode de extragere
specifice EFA sunt analiza factorilor principali
(eng. principal axis factoring), factorizarea
Alpha (eng Alpha factoring) i analiza factorial bazat pe estimarea verosimilitii maxime
(eng. maximum likelihood factor analysis sau
ML FA). Prima prezint avantajul de a nu fi
foarte pretenioas n ceea ce privete prezena
unei distribuii multivariate normale a datelor;
cea de-a doua este orientat pe maximizarea
fidelitii ( Cronbach), iar cea de-a treia, dei
sensibil la forma distribuiei datelor, prezint
marele avantaj de a testa gradul de semnificaie
statistic a saturaiei factorilor i corelaiei dintre factorii extrai (Fabrigar i colab., 1999,
Sava, 2011). Din pcate, din multitudinea de
indicatori descriptivi pe care metoda i poate
oferi, SPSS-ul ofer doar testul de estimare a
raportului verosimilitii (eng. likelihood ratio
test sau LRT), bazat pe o distribuie , foarte
sensibil la volumul eantionului, care tinde s
supraestimeze numrul de factori care trebuie
extrai (Hayashi, Bentler i Yuan, 2007). n
schimb, programul CEFA** (Browne, Cudeck,
Tateneni i Mels, 2008) ofer toi indicatorii
afereni acestei metode, fiind astfel indicat n
cazul n care dorim s folosim metoda ML FA.
Selectarea numrului optim de factori
Probabil cea mai cunoscut i utilizat
metod pentru selectarea numrului de factori
este criteriul lui Kaiser. Metoda presupune
selectarea unui numr de factori comuni egal cu
numrul de eigenvalue mai mari dect 1. Chiar
dac metoda funcioneaz acceptabil atunci
cnd EFA este folosit n cazul variabilelor, n
cazul utilizrii EFA pentru itemi metoda va produce n majoritatea situaiilor soluii eronate,
supraestimnd numrul de factori care trebuie
selectai (Fabrigar i colab., 1999; Henson,
2006). Astfel, folosirea acestei metode atunci
cnd utilizm EFA pentru dezvoltarea/adaptarea

cultural a unor probe este contraindicat.
O alt metod la fel de cunoscut este criteriul grafic al lui Cattel. n aceast procedur,
valorile eigenvalue calculate sunt reprezentate
grafic n ordine descresctoare, iar graficul este
inspectat pentru a se identifica ultima scdere
substanial a intensitii eigenvalue (aa-numitul punct de cotitur), dup care se selecteaz
un numr de factori comuni egal cu numrul de
eigenvalue situai deasupra punctului de cotitur
(Fabrigar i colab., 1999). Metoda funcioneaz
bine n situaiile n care factorii comuni prezeni
sunt foarte bine reprezentai. Dezavantajele ei
apar atunci cnd punctul de cotitur nu este
foarte evident, sau n momentul n care exist
mai multe scderi brute ale intensitii eigenvalue, deoarece nu exist o explicaie obiectiv
a ceea ce nseamn o scdere substanial n
intensitate (Sava, 2011).
Una dintre cele mai precise metode pentru
selectarea numrului optim de factori este metoda analizei paralele (Horn, 1965). Aceasta se
bazeaz pe compararea intensitii eigenvalue
dintre datele reale i un set de date aleatoare,
urmat de selectarea unui numr de factori comuni egal cu numrul valorilor eigenvalue reale care
sunt mai mari dect valorile eigenvalue aleatoare
(Hayton, Allen i Scarpello, 2004). Chiar dac
metoda ofer rezultate valide n majoritatea situailor, exist i anumite situaii particulare (ex.
numr foarte mare de itemi, multe comunaliti
foarte sczute: < .30) cnd analiza paralel tinde
s supraestimeze uor numrul de factori care trebuie selectai (Buja i Eyuboglu, 1992). Un dezavantaj major legat de accesibilitatea analizei paralele este reprezentat de faptul c metoda nu se
poate realiza din programul SPSS fr apelul la
sintax (Anexa 1 prezint un ghid simplificat
pentru realizarea unei analize paralele folosind
sintaxa SPSS-ului).
O metod care ctig din ce n ce mai mult
teren n faa abordrilor clasice este selectarea
* Programul SPSS nu permite realizarea PCA n varianta clasic; sub denumirea de principal component method programul realizeaz o analiz factorial a componentelor principale. Diferena dintre cele dou este minim, opiunea din SPSS
introducnd o surs suplimentar de eroare, lucru ce o face i mai puin dezirabil.
** Programul mpreun cu manualul de utilizare sunt disponibile gratuit pe internet la adresa: http://faculty.psy.ohiostate.edu/browne/software.php.
106
4/17/2012
1:10 PM
Page 107
numrului de factori n funcie de indicatorii

descriptivi oferii de analiza factorial bazat pe
estimarea verosimilitii maxime (ML FA).
Aceast metod ofer indicatori prezeni i n
analiza factorial confirmatorie. Cel mai potrivit indicator absolut este RMSEA (eng. root
mean squared error of aproximation), care
reprezint o estimare a discrepanei dintre modelul testat i datele folosite, pe gradele de libertate ale modelului (Sass, 2009). n plus, este
oferit i un interval de ncredere pentru valoarea
calculat. Interpretarea sa se face n felul urmtor: dac limita maxim a intervalului nu
depete 0.08 avem un model bun, dac nu
depete 0.10 avem un model acceptabil, iar
dac depete 0.10 avem un model deficitar
(Fabrigar i colab., 1999). Un alt indicator este
ECVI (eng. expected cross validation index),
care estimeaz gradul de generalizare al soluiei
obinute la alte eantioane. Dei nu exist valori
absolute pentru interpretarea ECVI, acesta este
util n compararea modelelor alternative, valorile mai mici artnd un grad de generalizare
mai ridicat. Un indicator considerat problematic
i, n consecin, contraindicat pentru selectarea
numrului de factori, este testul de estimare a
raportului verosimilitii (LRT), din cauza
supraestimrii numrului de factori care trebuie
selectai (Hayashi, Bentler i Yuan, 2007).
Prezentm n continuare o situaie practic
n care criteriul lui Kaiser supraestimeaz
numrul de factori ce trebuie extrai, n timp ce

criteriul grafic al lui Cattel i analiza paralel
sugereaz soluia optim. Exemplul este extras
din procesul de validare al Scalei de ndeplinire
a nevoilor la locul de munc (Need Satisfaction
at Work Scale NSWS; Van den Broeck,
Vansteenkiste, De Witte, Lens & Soenens, 2010)
n Romnia. Autorii scalei au identificat o structur factorial format din trei factori: nevoia de
autonomie, de competen i de relaionare.
Primele 5 valori eigenvalue extrase sunt:
5.31, 2.53, 1.93, 1.27 i 0.97. Scree plot-ul obinut este prezentat n figura 1, iar rezultatele analizei paralele sunt prezentate grafic n figura 2.
Dup cum se poate observa n cazul scree
plot-ului, punctul de cotitur este situat dup
cel de-al treilea factor. De asemenea, analiza
paralel arat c primele trei valori eigenvalue
reale sunt mai mari dect valorile eigenvalue
aleatoare. Astfel, ambele metode sugereaz o
soluie cu trei factori. n schimb, criteriul lui
Kaiser sugereaz o soluie cu patru factori
(supraestimnd numrul de factori care trebuie
extrai), deoarece exist patru valori eigenvalue
mai mari dect 1. Un ultim aspect ce trebuie discutat se refer la situaiile cnd este selectat un
numr greit de factori. n literatura de specialitate exist un consens cu privire la acest aspect:
subestimarea numrului de factori este o eroare
mult mai grav dect supraestimarea acestora
(Fava i Velicer, 1992; Wood, Tataryn i Gor-
Figura 1. Scree plot-ul pentru NSWS
107
4/17/2012
1:10 PM
Page 108
Figura 2. Rezultatele analizei paralele pentru NSWS
such, 1995). Cnd se produce subestimarea, factorii selectai vor conine un nivel de eroare considerabil, deoarece itemii care ar fi trebuit s
ncarce factorii care nu au fost extrai pot ncrca incorect factorii extrai. De asemenea, nivelul de saturaie al itemilor care ar ncrca n mod
normal factorii extrai poate fi afectat. Astfel,
nivelul de eroare regsit n model crete direct
proporional cu numrul de factori subestimai.
n schimb, cnd se produce supraestimarea,
nivelul de saturaie pentru factorii autentici
conine, n general, mai puin eroare dect n
cazul subestimrii. De asemenea, supraestimarea cu doi, trei sau chiar patru factori induce
aproximativ la fel de mult eroare n model ca i
supraestimarea cu un singur factor (Wood i
colab., 1995).
Rotirea factorilor
Pentru a ajunge la o structur factorial ct
mai adecvat trebuie utilizat criteriul structurii
simple definit de Thurstone. Acest lucru nseamn c din multitudinea de rotaii posibile, cea
care va avea cea mai bun structur simpl va
avea o semnificaie psihologic real, va fi cea
mai uor de interpretat i de replicat. Practic,
structura simpl se refer la o soluie n care
fiecare item obine saturaii ridicate ntr-un singur factor i saturaii reduse n ceilali factori
(Fabrigar i colab., 1999).
Pentru obinerea structurii simple au fost
dezvoltate o serie de rotiri ale factorilor. Cea
108
mai important distincie ntre acestea le

mparte n rotiri ortogonale i rotiri oblice
(Browne, 2001). Rotirile ortogonale constrng
factorii s rmn necorelai, pe cnd rotirile
oblice permit factorilor s coreleze ntre ei.
Din categoria rotirilor ortogonale, rotirea
Varimax este de departe cea mai popular i cea
mai des utilizat. Cu toate c n foarte multe
situaii aceast rotire a produs rezultate adecvate, folosirea ei (i a oricrei rotiri de tip ortogonal) n situaia utilizrii EFA n cazul itemilor
nu este indicat (Gorsuch, 1997). Principalul
motiv care conduce la aceast recomandare este
legat de faptul c, n general, factorii msurai
de un anumit instrument pihometric tind s
coreleze ntre ei. Chiar i n situaiile n care factorii au o anumit independen, tot exist corelaii de magnitudini sczute (pn la .30) ntre
acetia. De aceea, este mult mai indicat utilizarea rotirilor de tip oblic pentru a ajunge la
cea mai simpl structur factorial.
Spre deosebire de rotirile ortogonale, nu
exist nici o rotire oblic care s domine n
cercetrile psihologice. Cele mai utilizate rotiri
oblice sunt: Direct Quartimin, Promax i rotirea
orto-oblic Harris-Kaiser (Fabrigar i colab., 1999;
Browne, 2001). Rotirile Promax i Harris-Kaiser
se regsesc n SPSS n mod direct, iar rotirea
Direct Oblimin (din SPSS) cu valoarea delta
prestabilit la zero corespunde unei rotiri Direct
Quartimin. Att programul CEFA ct i programul
FACTOR ofer o gam mult mai larg de rotiri.
4/17/2012
1:10 PM
Page 109
Mai trebuie menionat faptul c utilizarea

unei rotiri oblice nu necesit ca factorii s fie
corelai. Dac soluia cu cea mai bun structur
simpl implic factori ortogonali, o rotire oblic
va oferi valori ale corelaiilor dintre factori
apropiate de zero i va produce o soluie factorial apropiat de cea a unei rotiri ortogonale (Fabrigar i colab., 1999). Deci, n cazul utilizrii
EFA pentru dezvoltarea/adaptarea cultural a
unei probe este mult mai indicat folosirea unei
rotiri de tip oblic, dect a uneia de tip ortogonal.
Identificarea factorilor fali
n situaia n care dup efectuarea rotirii
apar factori fali, acetia trebuie identificai i
respini. n prezent, nu exist reguli clare pentru
identificarea acestor factori fali; totui, prezentm patru indicii care v-ar putea ajuta n luarea
unei decizii legate de autenticitatea factorilor
extrai (Wood i colab., 1995):
a) gradul de saturaie al itemilor este mult
mai probabil ca factorii n care itemii sunt puternic saturai s fie autentici, dect factorii ce
prezint saturaii sczute ale itemilor;
b) valorile corelaiilor dintre itemii care
ncarc ntr-un factor este foarte probabil ca un
factor s fie autentic dac itemii care l ncarc
sunt corelai ntre ei;
c) semnificaia factorului dac itemii care
sunt puternic saturai ntr-un factor prezint
asemnri evidente ntre ei, atunci este mult mai
probabil ca factorul s fie autentic, dect dac
similaritile ntre itemii respectivi sunt aproape
inexistente;
d) replicabilitatea factorului dac un factor apare n mai multe analize factoriale realizate pe eantioane diferite, este destul de probabil ca acesta s fie autentic. Factorii fali ar
trebui s fie instabili n eantioane diferite, pe
cnd factorii autentici ar trebui s fie replicai.
Practic, dac nici unul dintre elementele
prezentate mai sus nu este prezent, atunci este
foarte probabil ca factorul respectiv s fie fals, i
trebuie eliminat. Dac cel puin dou elemente
sunt prezente, factorul are anse foarte mari de a
fi autentic, i ar trebui reinut. Dac doar un singur element este prezent, atunci factorul respectiv
ar trebui etichetat ca fiind discutabil, necesitnd cercetri ulterioare (Wood i colab., 1995).
Formarea pachetelor de itemi

O tehnic foarte des utilizat, att n cazul
analizei factoriale exploratorii ct i n cazul
celei confirmatorii, este formarea pachetelor de
itemi. Aceasta presupune gruparea mai multor
itemi mpreun dup anumite criterii, i
folosirea ulterioar a acestora ca variabile manifeste n EFA sau CFA (Little i colab., 2002;
Sava, 2011). Formarea pachetelor de itemi
depinde n exclusivitate de obiectivele studiului
i de instrumentele folosite. Dac scopul
cercetrii este de a identifica efectele unei variabile latente la un nivel superior de generalitate,
atunci formarea pachetelor de itemi poate
reduce sau neutraliza efectele perturbatoare ale
factorilor aflai la un nivel inferior de generalitate. n aceste situaii, formarea pachetelor de
itemi este indicat, poate chiar necesar. n
schimb, dac scopul studiului este de a verifica
particularitile constructului msurat la nivelul
itemilor individuali, atunci minimizarea
efectelor de nivel inferior ar tinde s ascund
exact efectele dorite a fi studiate. n aceste situaii, formarea pachetelor de itemi este contraindicat (Little i colab., 2002).
n cazul n care pentru atingerea obiectivelor
studiului este necesar formarea pachetelor de
itemi, exist un aspect foarte important care trebuie luat n considerare: dimensionarea setului
de itemi trebuie determinat naintea formrii
pachetelor (Little i colab., 2002). Att n cazul
adaptrii culturale a unui test, i cu att mai mult
n cazul dezvoltri unui nou instrument psihometric, este necesar aplicarea unei analize factoriale exploratorii pentru a determina dac setul de
itemi este unidimensional sau multidimensional.
Acest lucru este important deoarece tehnicile de
formare a pachetelor de itemi difer n funcie de
dimensionarea setului de itemi supui analizei.
Vom prezenta n continuare cinci tehnici de formare a pachetelor de itemi, primele trei fiind
pentru seturi unidimensionale de itemi, iar
ultimele dou pentru seturi multidimensionale de
itemi (Little i colab., 2002):
a) Repartizarea aleatoare const n distribuirea fiecrui item, ntr-un mod aleator, ctre
unul dintre pachete. Aceast tehnic ar trebui s
duc la formarea unor pachete care conin, n
109
4/17/2012
1:10 PM
Page 110
aproximativ aceeai msur, dispersie datorat

factorului comun.
b) Ponderarea item-construct are ca scop
construirea unor pachete echilibrate, din punct
de vedere al relaiei item-construct. Se realizeaz n felul urmtor: presupunem c avem un
set unidimensional de 12 itemi, i dorim s formm 3 pachete; folosind nivelul de saturaie al
itemilor ca i criteriu, se aleg primii trei itemi cu
cel mai ridicat nivel de saturaie (ex. itemii 5, 8
i 11) i se distribuie (n ordine) celor trei
pachete (itemul 5 pachetului 1, itemul 8
pachetului 2 i itemul 11 pachetului 3). Se continu prin selectarea urmtorilor trei itemi (ex.
itemii 1, 7 i 12) i distribuirea lor n cele trei
pachete, dar n ordine invers (itemul cu saturaia cea mai ridicat se distribuie pachetului ce
coninea itemul din seria anterioar cu saturaia
cea mai sczut: itemul 1 pachetului 3, itemul 7
pachetului 2 i itemul 12 pachetului 1). Algoritmul continu pn la epuizarea numrului de
itemi analizai.
c) Reducerea erorii de aprobare const n
gruparea n acelai pachet a unor itemi formulai
pozitiv cu itemi formulai negativ, dup ce
acetia din urm au fost codai invers (ex. dac
avem un set unidimensional de 16 itemi, dintre
care 8 sunt formulai pozitiv i 8 negativ, i
dorim s formm 4 pachete, vom include n
fiecare pachet 2 itemi formulai pozitiv i 2 formulai negativ). Scopul acestei tehnici este
reducerea bias-ului negativ vs. pozitiv, legat de
constructul care este msurat.
d) Abordarea consistenei interne const
n crearea pachetelor folosind faetele constructului msurat ca i criteriu de grupare, i se realizeaz astfel: presupunem c avem un set multidimensional de 16 itemi ce conine 4 faete (A,
B, C i D), fiecare msurat prin 4 itemi (A: 1 4, B: 5 - 8, C: 9 - 12, D: 13 - 16); primul pachet
ar reflecta faeta A, i ar fi format din itemii 1, 2,
3 i 4; cel de-al doilea pachet ar reflecta faeta B,
i ar fi format din itemii 5, 6, 7 i 8, .a.m.d..
e) Abordarea domeniului caracteristic
const n combinarea itemilor ce aparin unor
faete diferite pentru a forma pachete (folosind
exemplul anterior, fiecare pachet ar conine cte
un item din fiecare faet msurat: primul
pachet ar conine itemii 1, 5, 9 i 13, cel de-al
110
doilea itemii 2, 6, 10 i 14, etc.). Aceast tehnic

ncearc s explice multidimensionalitatea prin
crearea unor pachete care cuprind nu numai variana comun (ca n abordarea consistenei
interne), dar i variana aferent faetelor unice
ale multiplelor dimensiuni.
ncheiem aceast seciune prin menionarea
unor aspecte complementare legate de formarea
pachetelor de itemi: numrul de itemi afereni
unei pachet poate varia ntre 3 i 6 (Little i
colab., 2002; Bandalos, 2008). Alegerea
numrului de pachete ce urmeaz a fi formate
depinde att de numrul total de itemi utilizai,
ct i de raportul itemi factori. De exemplu, nu
ar avea nici un sens ca pentru un test unidimensional de 12 itemi s formm 6 pachete a cte 2
itemi fiecare, fiind mult mai indicat s formm
3 pachete a cte 4 itemi fiecare. De asemenea,
pentru un test multidimensional de 40 itemi
cte 10 pentru fiecare faet, am putea: a) s
folosim abordarea consistenei interne, i am
forma 8 pachete cte dou pentru fiecare
faet, fiecare fiind compus din 5 itemi; b) s
folosim abordarea domeniului caracteristic, i
atunci am putea forma cte pachete dorim, dar
ar trebui s ne asigurm c fiecare pachet
conine itemi din fiecare faet i, preferabil,
fiecare faet este prezent n fiecare pachet n
aceeai msur (Little i colab., 2002).
De asemenea, mai trebuie s menionm
faptul c tehnicile pentru seturi multidimensionale de itemi pot fi aplicate n condiiile n care
instrumentul msoar faete ale aceluiai construct (ex. agresivitatea msurat prin faetele
agresivitate fizic, verbal, furie i ostilitate). n
cazul n care instrumentul msoar factori care
au o anumit independen (ex. un test care
msoar factorii modelului Big Five de personalitate, fr a msura ns i faetele lor), se
recomand folosirea unor tehnici de formare a
pachetelor unidimensionale pentru itemii afereni fiecrui factor n parte. n schimb, n cazul
unui test ce msoar i faetele modelului Big
Five, se poate aplica una dintre tehnicile multidimensionale de formare a pachetelor pentru
fiecare factor (cu faetele aferente), dup modelul prezentat n exemplul anterior.
4/17/2012
1:10 PM
Page 111
Concluzii
Scopul acestui articol a fost oferirea de
informaii relevante despre aspectele care trebuie luate n considerare atunci cnd analiza
factorial exploratorie este utilizat n cazul
itemilor. Toate informaiile i sugestiile prezentate au ca scop mbuntirea calitii utilizrii
EFA, n procesul de adaptare cultural sau dezvoltare a testelor. Astfel, acest articol poate fi
folositor att cercettorilor care se ocup cu dezvoltarea sau adaptarea cultural a instrumentelor
psihometrice pe populaie romneasc, ct i
studenilor (n principal masteranzi sau doctoranzi) a cror teme de disertaie/doctorat includ
astfel de analize.
La finalul acestui articol prezentm un set
de recomandri sub form de ghid, bazat pe
informaiile anterioare, legat de modalitatea optim de aplicare a EFA n cazul dezvoltrii/ adaptrii culturale a instrumentelor psihometrice:
1. Verificai dac EFA este tehnica statistic
de care avei nevoie pentru atingerea obiectivelor studiului.
2. Selectai un eantion care s respecte att
criteriile legate de volum (un raport participaniitemi de minim 5:1) ct i legate de caracteristici (omogenitate medie i similaritate ntre
eantion i populaia int a probei).
3. Verificai (pentru precauie) valorile
Skweness i Kurosis ale itemilor nainte de a
ncepe analiza. Dac valorile indic pentru mai
muli itemi distribuii ce deviaz substanial de
la normalitate, putei continua analiza din programul SPSS (asumndu-v riscul de a obine o
structur factorial neadecvat), sau putei apela
la programul FACTOR.
4. Realizai o EFA din programul SPSS,
urmnd urmtoarele recomandri: metod de
extracie analiza factorilor principali; metod
de selecie criteriul lui Kaiser (dei nu este o
soluie adecvat, se poate folosi n cazul realizrii unei EFA incipient, doar pentru a verifica
ci factori selecteaz programul fr nici o constrngere); rotire Direct Oblimin (cu valoarea
zero pentru delta).
5. Verificai scree plot-ul obinut i apoi
realizai o analiz paralel din sintaxa programului SPSS, folosind ghidul din Anexa 1. Dac
soluia indicat de analiza paralel este susinut
i de scree plot, putei fi destul de sigur c e

soluia corect. Dac cele dou difer, vei alege
soluia care are cel mai ridicat grad de interpretabilitate. n cazul n care ambele soluii au
un nivel sczut de interpretabilitate, putei ncerca realizarea EFA utiliznd unul dintre programele indicate n seciunile anterioare.
6. Realizai din nou o EFA, selectnd de
aceast dat numrul de factori indicat de
soluia aleas anterior. Matricea structurii factoriale va conine structura pe care o vei analiza.
7. Nu uitai ca n momentul n care vei
raporta rezultatele obinute s menionai toate
alegerile fcute, incluznd o justificare succint
pentru fiecare.
Bibliografie
Armstrong, J. S. (1967). Derivation of theory by means
of factor analysis or Tom Swift and his electric
factor analysis machine. The American Statistician, 21(1), 17-21.
Bandalos, D. L. (2008). Is Parceling Really Necessary?
A Comparison of Results From Item Parceling and
Categorical Variable Methodology. Structural
Equation Modeling: A Multidisciplinary Journal,
15(2), 211-240.
Bernstein, I. H., Teng, G. (1989). Factoring items and
factoring scales are different: Spurious evidence
for multidimensionality due to item categorization. Psychological Bulletin, 105, 467-477.
Browne, M. W. (2001). An Overview of Analytic Rotation in Exploratory Factor Analysis. Multivariate
Behavioral Research, 36(1), 111-150.
Browne, M. W., Cudeck, R., Tateneni, K. & Mels G.
(2008). CEFA: Comprehensive Exploratory Factor
Analysis, Version 3.03 [Computer software and
manual]. Gsit la adresa: http://faculty.psy.ohiostate.edu/browne.
Buja, A., & Eyuboglu, N. (1992). Remarks on parallel
analysis. Multivariate Behavioral Research, 27,
509-540.
Dinno, A. (2009). Exploring the Sensitivity of Horns
Parallel Analysis to the Distributional Form of
Random Data. Multivariate Behavioral Research,
44(3), 362-388.
Fabrigar, L. R., Wegener, D. T., MacCallum, R. C., &
Strahan, E. J. (1999). Evaluating the use of
exploratory factor analysis in psychological
research. Psychological Methods, 4(3), 272-299.
Fava, J. L., & Velicer, W. F. (1992). The effects of
overextraction on factor and component analysis.
Multivariate Behavioral Research, 27, 387-415.
111
4/17/2012
1:10 PM
Page 112
Ford, J.K., MacCallum, R.C. & Tait, M. (1986). The

Application of Exploratory Factor Analysis in
Applied Psychology: A Critical Review and Analysis. Personnel Psychology, 39(2), 291-314.
Gorsuch, R. L. (1997). Exploratory Factor Analysis: Its
Role in Item Analysis. Journal of Personality
Assessment, 68(3), 532-560.
Gould, S. J. (1981). The mismeasure of man. New York:
Norton.
Hayashi, K., Bentler, P. M., & Yuan, K.-H. (2007). On
the Likelihood Ratio Test for the Number of Factors in Exploratory Factor Analysis. Structural
Equation Modeling: A Multidisciplinary Journal,
14(3), 505-526.
Hayton, J. C., Allen, D. G., & Scarpello, V. (2004). Factor Retention Decisions in Exploratory Factor
Analysis: a Tutorial on Parallel Analysis. Organizational Research Methods, 7(2), 191-205.
Henson, R. K. (2006). Use of Exploratory Factor Analysis in Published Research: Common Errors and
Some Comment on Improved Practice. Educational and Psychological Measurement, 66(3),
393-416.
Hogarty, K. Y. (2005). The Quality of Factor Solutions
in Exploratory Factor Analysis: The Influence of
Sample Size, Communality, and Overdetermination. Educational and Psychological Measurement, 65(2), 202-226.
Horn, J. L. (1965). A rationale and technique for estimating the number of factors in factor analysis.
Psychometrika, 30, 179-185.
International Test Commision (2010). International Test
Commision Guideliness for Translating and Adapting Tests. Gsit la adresa: http://www.intescom.org.
Little, T. D., Cunningham, W. A., & Shahar, G. (2002).
To Parcel or Not to Parcel : Exploring the Question, Weighing the Merits. Structural Equation
Modeling: A Multidisciplinary Journal, 9(2),
151-173.
112
Lorenzo-Seva, U., & Ferrando, P. J. (2006). FACTOR:

a computer program to fit the exploratory factor
analysis model. Behavior research methods, 38(1),
88-91.
MacCallum, R.C., Widaman, K.F., Zhang, S. & Hong.,
S. (1999). Sample size in factor analysis. Psychological Methods, 4(1), 84-99.
OConnor, B. P. (2000). SPSS and SAS programs for
determining the number of components using parallel analysis and velicers MAP test. Behavior
research methods, instruments, & computers,
32(3), 396-402.
Panter, A. T., Swygert, K. A., Dahlstrom, W. G., &
Tanaka, J. S. (1997). Factor analytic approaches to
personality item-level data. Journal of Personality
Assessment, 68, 561-589.
Sass, D. (2009). Factor Loading Estimation Error and
Stability Using Exploratory Factor Analysis. Educational and Psychological Measurement, 70(4),
557-577.
Sava, F.A. (2011). Analiza datelor n cercetarea psihologic. Cluj-Napoca: Editura ASCR.
De Winter, J. C. F., Dodou, D. & Wieringa, P.A. (2009).
Exploratory Factor Analysis With Small Sample
Sizes. Multivariate Behavioral Research, 44(2),
147-181.
Van den Broeck, A., Vansteenkiste, M., De Witte, H.,
Soenens, B., Lens, W. (2010). Capturing autonomy, competence, and relatedness at work: construction and initial validation of the Work-related
Basic Need Satisfaction scale. Journal of Occupational and Organizational Psychology, 83 (4),
981-1002.
Wood, J. M., Tataryn, D. J., & Gorsuch, R. L. (1996).
Effects of under- and overextraction on principal
axis factor analysis with varimax rotation. Psychological Methods, 1(4), 354-365.
4/17/2012
1:10 PM
Page 113
Anexa 1: Ghid pentru realizarea unei analize paralele folosind sintaxa

programului SPSS
Prezentm n continuare 10 pai simpli pentru a realiza o analiz paralel folosind sintaxa programului SPSS*:
Pasul 1: Dup deschiderea programului, accesai sintaxa fcnd click pe: File New Syntax.
Pasul 2: Copiai urmtoarele comenzi n fereastra aferent sintaxei:
set mxloops=9000 printback=off width=80 seed = 1953125.
matrix.
Pasul 3: Introducei numele i locaia fiierului ce conine datele pe care dorii s le analizai
(fiierul trebuie s fie realizat n SPSS), dup comanda FILE = (ex. FILE = C:\Date\studiu.sav).
Dac specificai FILE = * , programul va analiza fiierul SPSS care este deschis n acel moment,
acest lucru fiind mult mai la ndemn. De asemenea, trebuie s specificai variabilele din baza de date
ce vor fi folosite pentru analiz (ex. dac ai numerotat itemii pe care dorii s-i analizai cu it1, it2,
it3,...., it20, atunci vei scrie VAR = it1 to it20). Comanda aferent descrierii de mai sus este:
GET raw / FILE = * / missing=omit / VAR = it1 to it20.
Pasul 4: Introducei numrul seturilor de date paralele ce dorii s fie analizate (n general, 1000
de seturi sunt suficiente; n cazul n care soluia obinut nu este ndeajuns de clar, putei utiliza mai
multe seturi de date):
compute ndatsets = 1000.
Pasul 5: Introducei valoarea percentilului dorit pentru realizarea analizei (de obicei, percentilul
95):
compute percent = 95.
Pasul 6: Alegei metoda de extracie a factorilor pe care dorii s o utilizai (1 pentru analiza
componentelor principale, sau 2 pentru analiza factorilor principali). Indiferent de metoda de
extracie pe care ai folosit-o, vei selecta opiunea 1, deoarece selectarea celei de-a doua opiuni tinde
s supraestimeze numrul de factori care trebuie extrai (Buja i Eyuboglu, 1992; Dinno, 2009):
compute kind = 1.
Pasul 7: Alegei modalitatea de generare a seturilor de date paralele (1 pentru generarea unor
date aleatoare cu distribuie normal, sau 2 pentru permutarea datelor brute). Alegei prima variant
n cazul n care itemii supui analizei au distribuii ncadrate n limitele normalitii. Cea de-a doua
variant este foarte precis n situaiile n care datele brute nu au o distribuie normal, dar dureaz mai
mult timp. Dac itemii utilizai n analiz ncalc condiiile de normalitate, realizai mai nti o analiz
paralel folosind prima metod (pentru a avea un punct de reper), iar apoi folosii cea de-a doua
metod.
compute randtype = 1.
* Ghidul a fost inspirat din indicaiile Prof. Univ. Dr. Brian OConnor despre analiza paralel, disponibile pe internet la
adresa: https://people.ok.ubc.ca/brioconn/nfactors/nfactors.html.
113
4/17/2012
1:10 PM
Page 114
Pasul 8: Introducei urmtoarele comenzi n fereastra aferent sintaxei:

compute ncases = nrow(raw).
compute nvars = ncol(raw).
do if (kind = 1 and randtype = 1).
compute nm1 = 1 / (ncases-1).
compute vcv = nm1 * (sscp(raw) - ((t(csum(raw))*csum(raw))/ncases)).
compute d = inv(mdiag(sqrt(diag(vcv)))).
compute realeval = eval(d * vcv * d).
compute evals = make(nvars,ndatsets,-9999).
loop #nds = 1 to ndatsets.
compute x = sqrt(2 * (ln(uniform(ncases,nvars)) * -1) ) &*
cos(6.283185 * uniform(ncases,nvars) ).
compute vcv = nm1 * (sscp(x) - ((t(csum(x))*csum(x))/ncases)).
compute evals(:,#nds) = eval(d * vcv * d).
end loop.
end if.
compute nm1 = 1 / (ncases-1).
compute vcv = nm1 * (sscp(raw) - ((t(csum(raw))*csum(raw))/ncases)).
compute realeval = eval(d * vcv * d).
compute evals = make(nvars,ndatsets,-9999).
loop #nds = 1 to ndatsets.
compute x = raw.
loop #c = 1 to nvars.
loop #r = 1 to (ncases -1).
compute k = trunc( (ncases - #r + 1) * uniform(1,1) + 1 ) + #r - 1.
compute d = x(#r,#c).
compute x(#r,#c) = x(k,#c).
compute x(k,#c) = d.
end loop.
end loop.
compute vcv = nm1 * (sscp(x) - ((t(csum(x))*csum(x))/ncases)).
compute evals(:,#nds) = eval(d * vcv * d).
end loop.
end if.
114
4/17/2012
1:10 PM
Page 115
compute num = rnd((percent*ndatsets)/100).

compute results = { t(1:nvars), realeval, t(1:nvars), t(1:nvars) }.
loop #root = 1 to nvars.
compute ranks = rnkorder(evals(#root,:)).
loop #col = 1 to ndatsets.
do if (ranks(1,#col) = num).
compute results(#root,4) = evals(#root,#col).
break.
end if.
end loop.
end loop.
compute results(:,3) = rsum(evals) / ndatsets.
print /title=PARALLEL ANALYSIS:.
print /title=Principal Components & Random Normal Data Generation.
else if (kind = 1 and randtype = 2).
print /title=Principal Components & Raw Data Permutation.
end if.
compute specifs = {ncases; nvars; ndatsets; percent}.
print specifs /title=Specifications for this Run:
/rlabels=Ncases Nvars Ndatsets Percent.
print results
/title=Raw Data Eigenvalues, & Mean & Percentile Random Data Eigenvalues
/clabels=Root Raw Data Means Prcntyle /format f12.6.
compute root = results(:,1).
compute rawdata = results(:,2).
compute percntyl = results(:,4).
save results /outfile=* / var=root rawdata means percntyl .
end matrix.
TSPLOT VARIABLES= rawdata means percntyl /ID= root /NOLOG.
Pasul 9: Executai sintaxa fcnd click pe: Run All.

Pasul 10: Alegei numrul optim de factori ce trebuie reinui, comparnd valorile eigenvalue din
datele reale cu cele din datele aleatoare. Trebuie s menionm faptul c, n varianta original a procedurii (Horn, 1965), valorile eigenvalue medii din datele aleatoare erau utilizate pentru comparaie. O
recomandare curent este utilizarea valorilor eigenvalue din datele aleatoare ce corespund percentilului ales (OConnor, 2000).
115
4/17/2012
1:10 PM
Page 116
Ticu Constantin, Pregtirea i realizarea

evalurii psihologice individuale. Norme,
metodologie i proceduri, Editura Polirom,
Iai, 2012, 350 pag.
Reprezentnd un subiect nsemnat al psihologiei muncii i organizaionale, evaluarea

psihologic necesit o tratare pe msura importanei sale. Cu deziderate precum contribuirea la
succesul unei organizaii prin asigurea celei mai
bune resurse umane i mbuntirea preformanei i a strii de bine a angajailor, psihologii
specializai n domeniul muncii i organizaional resimt nevoia unor mijloace susinute empiric
i manualizate. Bazndu-se pe o structur clar
i uor de urmrit, volumul de fa i propune
s ndeplineasc aceste necesiti.
Dac volumul anterior (Evaluarea psihologic a personalului, 2004) a fost primit de ctre
practicienii domeniului cu mult entuziasm, ediia
actual reuete o ancorare i mai puternic n
realitatea practicianului. Cum ine s menioneze
nc din prefa, prof dr. Ticu Constantin a concentrat n paginile lucrrii sale nu doar rezultatele unei documentari temeinice, ci mai ales
experiena sa de practician i produsul interaciunii cu ali practicieni. Astfel, putem vorbi de o
colecie de bune practici i recomandri testate
prin proprie experien. Toate acestea sunt concentrate n paisprezece capitole mprite n
patru seciuni i o sum impresionant de anexe
prin care sunt oferite tehnici, instrumente i alte
materiale cu caracter exemplificator.
n prima parte, cititorul este poziionat n
cadrul legal de realizare a activitilor de evaluare psihologic. Considerentele critice i claritatea limbajului scutesc cititorul de demersul
anevoios de a urmri textele legislative. Astfel,
primul capitol, intitulat Forme de organizare a
activitii psihologice, ne familiarizeaz cu
reglementrile profesiei de psiholog aa cum
sunt ele specificate n Legea nr. 213 din 2004.
Reglementrile ce vizeaz obinerea dreptului
de liber practic, diferitele niveluri de specializare, formele de exercitare a profesiei i
condiiile necesare infiinrii unui cabinet de
profil sunt prezentate succint, pstrndu-se doar
informaiile strict relevante. Avnd deja informaiile despre practicarea autorizat a profesiei
116
de psiholog, continum prin a primi informaii

specifice reglementrilor activitii de evaluare
psihologic. Analiza clar a cadrului legislativ
evideniaz tipurile de evaluri psihologice periodice i categoriile profesionale pentru care
aceast activitate este considerat obligatorie. n
acest context, este meritorie utilizarea informaiilor de actualitate, textul cuprinznd i noile
reglementri aduse prin Hotrrea nr. 1169 din
25.11.2011. n continuare, sunt tratate puncte
precum relaia cu medicul de medicina muncii i
raportul cu acesta, i avizul psihologic. La final,
sub titlul Obligaii i responsabiliti ale psihologului evaluator ne sunt prezentate principiile
generale i standardele etice, i normele privind
confidenialitatea actului psihologic, sub forma
n care sunt reglementate de ctre Colegiul Psihologilor din Romnia.
A doua parte poart titlul: Cadrul conceptual i metodologic al evalurii psihologice. Propunndu-i s clarifice conceptele cu care se va
opera n continuare, cea de-a doua seciune este
structurat pe trei capitole. Domeniile de analiz psihologic n firme/instutiii este primul
dintre acestea n cadrul cruia concepte, precum
evaluare la nivel individual, la nivel organizaional i la nivelul unor colectiviti exinse sunt
tratate intit. Incursiunea este susinut printr-un
limbaj adaptat practicienilor i printr-o suit de
reprezentri grafice. Capitolul numrul cinci sau
cel de-al doilea capitol al acestei seciuni ne va
clarifica conceptele de aptitudine, trasatur de
personalitate i atitudini, ilustrnd cu exemple
de instrumente specifice acestora. Relaia dintre
trsturile de personalitate i performan este
un subiect central n interesele psihologilor
practicieni n domeniul muncii i organizaional.
Astfel, tratarea factorilor modelului Big Five n
raport cu performana profesional are parte i
de o scurt inventariere a indicatorilor corelaionali. Informaiile despre valoarea predictiv
a factorilor de personalitate n ceea ce privete
performana (corelaiile deseori sczute sau cel
mult medii) beneficiaz de explicaii i clarificari. Mai mult, avem parte de un subcapitol dedicat strategiilor de explorare a relaiei dintre personalitate i performan, cu evidenierea
competenelor necesare derulrii unui astfel de
proces, dar i indicaii cu privire la evaluarea
performanei. Aceast ntreag suit de infor-
4/17/2012
1:10 PM
Page 117
Recenzii
maii este ilustrat prin mprtirea unui studiu

de caz.
Urmtoarele patru capitole descriu cea de-a
treia parte a volumului. Ele sunt dedicate
restriciilor impuse examenului de evaluare psihologic prin norme legale i deontologice,
evalurii psihologice la angajare n contextul
seleciei candidailor, evalurii psihologice periodice i altor contexte n care se preteaz o astfel de intervenie. Astfel, facndu-se apel la pruden, practicienii beneficiaz de o suit de
indicaii desprinse din reglementrile legale.
Aceste indicaii clarific contextele i condiiile
n care evaluarea psihologic este potrivit. Un
astfel de context este cel al seleciei profesionale. Capitolul dedicat acestui subiect face o
prezentare extensiv a ntregului proces de
recrutare i selecie, punctnd recomandri i
tehnici pentru fiecare etap n parte. De aceeai
atenie beneficiaz i evaluarea psihologic
periodic (n medicina muncii). n acest capitol,
pornind de la obiectivele procesului n sine i
pn la modul de completare a avizului psihologic, avem parte de o prezentare manualizat a
fiecrei etape n parte. Aceast seciune este
ncheiat prin prezentarea altor contexte profesionale n care evaluarea psihologic i gsete
rostul. Fie c este vorba de situaiile de promovare, de formarea personalului i dezvoltarea
organizaional sau de managementul carierei,
toate beneficiaz de atenia cuvenit.
Ultima parte abordeaz identificarea, analiza, cuantificarea i evaluarea criteriilor psihologice relevante. Astfel, beneficiem de
prezentarea unor strategii pe care putem s le
folosim n efortul de identificare a criteriilor psihologice. Mai apoi, ne sunt prezentate n mod
consecutiv modaliti de operaionalizare i
management al criteriilor deja identificate, fiind
constant satifcut necesitatea practicianului
prin nsoirea cu exemple. Capitolele dedicate
construirii ghidului de interviu i grilelor de
evaluare a conduitelor candidailor n timpul
interviului se bucur de o atenie deosebit, att
sub aspectul decelrii unor principii i strategii
de formulare ct i sub cel al exemplelor concrete oferite practicienilor. Penultimul capitolul
reuete cu succes a clarifica o dilem nsemnat a psihologului practician, i anume
selectarea instrumentelor standardizate pe care
le va folosi n evaluare. Astfel, pe lng normele

de construire, validare i standardizare care ne
sunt aduse n vedere spre a le urmri n alegerile
pe care le facem, este prezentat i o inventariere a principalilor dezvoltatori de probe psihologice acreditai de ctre Colegiul Psihologilor din Romnia. Cel din urm capitol ne
ghideaz spre a realiza evaluarea psihologic
descriind etapa de administrare a instrumentelor
psihologice. Cu acest scop suntem minuios purtai prin fiecare detaliu practic: stabilirea coordonatelor de timp i spaiu, anunarea i
invitarea candidailor, pregtirea materialelor
specifice instrumentelor folosite, administrarea
consemnului, monitorizarea modului de completare i modul n care ncheiem evaluarea. Pe
lng aceste informaii este adugat i o seciune robust dedicat controlului tendinei de
faad i a distorsiunii datelor evalurii psihologice, i una introductiv pentru interpretarea
datelor obinute.
Pregatirea i realizarea evaluarii psihologice individuale. Norme, metodologie i proceduri este o lucrare recomandat nu doar
proaspeilor absolveni dornici s i nceap
practicarea profesiei ct i profesionitilor ce
doresc s obin noi soluii pentru a-i
mbunti practica sau resimt nevoia clarificrii
anumitor aspecte. n concluzie, dac ar fi s
descriem volumul de fa prin intermediul unor
concepte cheie, termenii claritate, aplicabilitate
i pragmatism pot fi alei fr echivoc.
Drd. Andrei Rusu
Universitatea Alexandru Ioan Cuza din Iai
Universitatea de Vest din Timioara

Organisational Destroyers, 2011, 195 p.
Psihopaii sunt puini cunoscui n afara

imaginii lor criminale. Totui, dup cum recenta
criz finaciar evideniaz, comportamentul
unui numr mic de manageri poate dobor ntreg
sistemul vestic de business. Psihopaii reprezint 1% din oameni i sunt caracterizai prin lipsa
contiinei sau empatiei. Unii psihopai sunt vio-
117
4/17/2012
1:10 PM
Page 118
Recenzii
leni i sfresc n nchisoare, ns alii i cldesc cariere n corporaii. Aceast carte investigheaz cine sunt ei, ce fac i care sunt consecinele prezenei lor n organizaii.
Dei reprezint doar un procent mic de
angajai sau manageri, psihopaii corporatiti au
o nfluen important i semnificativ negativ
asupra activitii unei organizaii. Prezena psihopailor n cadrul organizaiilor provoac situaii conflictuale i determin niveluri crescute
de intimidare n rndul celorlali angajai, cresc
procentele de abandon, de constrngeri organizaionale, determin insatisfacie profesional i
niveluri mai sczute de responsabilitate social.
Organizaiile i societile care doresc s
supravieuiasc i s prospere vor trebui s elimine psihopaii corporatiti sau s le monitorizeze
ndeaproape activitatea.
Clive Boddy a studiat efectele pe care le
produc psihopaii corporatiti, aa cum sunt
trite de sute de manageri, n ultimii cinci ani.
Cercetarea a inclus colectarea informaiilor
privind prezena acestora n diferite niveluri
organizaionale, colectarea de rapoarte de incidente critice care implic poteniali psihopai
corporatiti i, mai recent, colectarea de date
privind efectele acestora asupra rezultatelor
organizaionale.
Volumul este alctuit din 15 capitole care
prezint psihopaii corporatiti i efectele pe
care acetia le produc n cadrul organizaional.
Primul
capitol
descrie
psihopaii
corporatiti ca Distrugtori de Organizaii.
Atunci cnd mari organizaii sunt distruse de
ctre aciunile directorilor, cnd angajaii i
pierd locurile de munc i cteodat mijloacele
de existen, cnd societatea pierde pri importante din infrastructura sa, directorii implicai
pleac cu contiina mpcat i cu mari sume de
bani, fr s le pese de haosul provocat.
Dei acetia pot prea buni, fermectori i
de succes, n realitate sunt aproape n ntregime
toxici pentru organizaiile n care lucreaz. Pui
n funcii de conducere, acetia distrug moralul
i starea de bine a angajaiilor, umilindu-i,
manipulndu-i, abuznd de regulile organizaionale pentru a-i controla, ncriminndu-i de
erori comise chiar de ei, hruindu-i i chiar constngndu-i la activiti sexuale.
118
Cercetarea prezentat este realizat pe un

eantion de 346 de persoane cu un nivel nalt de
educaie, att manageri, ct i muncitori, membrii ai Camerei de Comer, organizaii caritabile,
studeni i absolveni de studii universitare de
afaceri, precum i membrii din alte organizaii
din Perth, Australia. Chestionarul cuprinde n
total 572 de ntrebri, 345 despre managerii
actuali i 227 despre manageri cu care au lucrat.
Scala Psychopathy Measure-Management
Research Version a fost inclus n chestionar
pentru a determina prezena sau absena psihopailor ntr-un mediu de lucru.
Concluziile cercetrii sugereaz c psihopaii corporatiti au o influen asupra organizaiilor extrem de congruent cu ateptrile negative ale teoreticienilor de vrf, cercettorilor i
psihologilor specializai n psihopatii. n mod
evident, prezena acestora n cadrul organizaional are un impact semnificativ, un efect
msurabil asupra veniturilor organizaiei n termeni de timp pierdut, nivel sczut de performan, asupra creterii volumului de munc, a
condiii de lucru dificile.
n al doilea capitol The Origins of Corporate Psychopaths este prezentat o analiz a literaturii din care reiese c psihopatia la nivel
individual poate evolua n moduri diferite, n
funcie de cum este alimentat nc din
copilrie. Psihopaii pot deveni criminali antisociali sau se pot ncadra cu succes n societate.
Unii dintre ei pot intra n organizaii i, astfel,
pot deveni Psihopai Corporatiti. Psihopaii
mai inteligeni ce beneficiaz de o bun educaie
i un bun cadru social sunt deseori capabili s
aleag un mod de via n care nu se manifest
n mod evident anti-social i i pot folosi
farmecul i manipularea pentru a-i ascunde
comportamentele deviante.
Capitolul 3, Bullying, Conflict and Unfair
Supervision, abordeaz problema relaiei semnificative dintre psihopatia corporatist i agresivitatea, manifestrile conflictuale i supravegherea
neloial. Este de ateptat ca efectele negative s
se exprime pn la diminuarea performanei sau
a creativitii colective. 26 % din incidentele relevate n aceast cercetare, n care indivizii au fost
martorii unor tratamente nefavorabile la locul de
munc sunt asociate cu prezena Psihopailor
Corporatiti. Rezultatele sunt n acord cu literatu-
4/17/2012
1:10 PM
Page 119
Recenzii
ra de specialitate, conform creia psihopaii corporatiti se manifest prin comportamente agresive, precum umilirea, abuzul verbal sau criticile
nejustificate la adresa celorlali.
Urmtorul capitol demonstreaz influena
puternic a Psihopailor Corporatiti asupra
felului n care angajaii percep responsabilitatea
la locul de munc i asupra angajamentului
acestora fa de organizaie. Psihopaii Corporatiti sunt interesai doar de satisfacerea propriilor interese, nu de succesul pe termenul lung al
organizaiei n care lucreaz. Sunt preocupai s
conduc corporaii pentru putere, bani sau prestigiu, pe care i-l doresc cu nerbdare, neinnd
cont de soarta organizaiei sau a semenilor.
Rezultatele ulterioare ale studiilor, prezentate n carte, demonstreaz c prezena Psihopailor Corporatiti sporete semnificativ constrngerile organizaionale i volumul de munc
al celorlali angajai care sunt obligai s rectifice
greelile fcute de aceti manageri incompeteni.
De asemenea, acetia i nsuesc rezultatele
muncii celorlali, le atribuie erori comise chiar
de ei i i umilesc n public, genernd astfel
dizarmonii, crize i confuzii la locul de munc.
n capitolul 8, Corporate Psychopaths and
Organisational Seniority, este evideniat faptul
c dei aparent psihopaii reprezint un fenomen
universal, manifestrile comportamentale ale
sindromului pot fi modificate de influene culturale sau chiar de diferite nivele n cadrul corporaiilor. Conform datelor evideniate, Psihopaii Corporatiti sunt mai des ntlnii la
nivele nalte ale organizaiilor. Aceste rezultate
corespund punctului de vedere ipotetic al teoreticienilor conform crora Psihopaii Corporatiti sunt mai bine dotai s urce n ierarhia
organizaiei i s obin poziii de conducere.
Influena direct a Psihopailor Corporatiti
asupra angajailor se poate exprima n realitate
i printr-o rata mare de abandon sau de dezangajare emoional. Prezena acestora facilitnd un
mediu de munc toxic, angajaii fie caut s
evite contactul cu manageri psihopai, fie abandoneaz complet organizaia.
O parte important a acestei cri o reprezint studiul incidentelor critice n comportamentul Psihopailor Corporatiti. Numrul limitat de
incidente colectate demonstreaz c acetia se
comport n moduri previzibile: sunt lipsii de
scrupule, gata s mint i folosesc resursele

organizaiei n propriul interes.
Autorul atrage atenia asupra modului n
care corporaiile i fac reclam pentru posturile
disponibile sugernd pruden, deoarece n dorina lor de a angaja oameni care sunt dispui s
fac orice pentru a avea succes, las impresia c
faciliteaz i atrag angajarea Psihopailor Corporatiti.
Pe lng Psihopatia Corporatist, n capitolul 13, Corporate Psychopaths Compared
with Other Dark Leadership Personalities,
autorul prezint i alte tulburri de personalitate
pe care le pot avea liderii. Acestea nu se exclud
reciproc i se pot suprapune. Sunt evideniate:
personalitile anti-sociale, machiavelice, narcisiste, personalitile disociate i sociopate.
Mesajul important pe care l transmite
autorul prin intermediul acestei cri este acela
c identificarea psihopailor n corporaii este
important pentru longevitatea sistemului capitalist, pentru corporaii, dreptate social i chiar
stabilitatea financiar, pe termen lung. Principalul scop al acestei lucrri este acela de a face
munca psihologilor din zona psihopatiei i, mai
ales, a psihopailor de succes mai larg cunoscut responsabililor de resurse umane i a tuturor celor implicai n decizii organizaionale.
M.A. Florin Zamfirache
Horeca Expert
Anna Koch & Karl Westhoff, Task-AnalysisTools (TAToo) Step-by-Step Support for
Successful Job and Work Analysis, Pabst
Science Publishers, 2012, 93 p.
Dup cum reiese nc din titlu, Task-Analysis-Tools (TAToo) Step-by-Step Support for
Successful Job and Work Analysis, este o carte
scris sub forma unui ghid practic de analiz a
muncii i a postului.
Pe parcursul a patru capitole, autorii expun
un ghid util pentru practicieni, cu instruciuni
articulate pas cu pas pentru utilizarea n probleme reale, dar i cu o fundamentare tiinific
teoretic solid.
119
4/17/2012
1:10 PM
Page 120
Recenzii
n partea introductiv, autorii argumenteaz

pertinent necesitatea unei analize psihologice
aprofundate, dar mai ales calitative a muncii, ca
pas primordial n orice intervenie organizaional. Realizat cu profesionalism, analiza psihologic a muncii ofer posibilitatea identificrii
criteriilor de performan ntr-o anumit profesie, criterii care, n final, la nivel organizaional,
se traduc n productivitate crescut i succes
financiar. Paradoxal, dei aceste aspecte sunt
bine cunoscute iar repere calitative privind analiza i proiectarea activitilor muncii exist sub
form de recomandri i standarde, diverse
studii internaionale arat c doar 5% din companiile de nivel mediu recurg la astfel de analize
i intervenii. Se pare c, adevrata provocare
const n transpunerea n practic a principiilor
i modelelor teoretice avansate de cercettori.
Tocmai acesta este obiectivul autorilor prin
lucrarea de fa: s aduc o contribuie la
umplerea golului dintre teorie i practic, prin
propunerea unei scheme de lucru accesibile i
unui practician mai puin experimentat.
n al doilea capitol, dedicat principiilor teoretice i metodologice, autorii fac o trecere n
revist a principalelor tipuri de clasificri
privind analiza muncii i a posturilor, lmuresc
aspecte care in de deciziile necesare n derularea unui altfel de program i propun o serie de
recomandri n proiectarea unui design complet
i riguros de analiz a muncii, care stau i la
baza instrumentului prezentat n al treilea capitol Task-Analysis-Tools (TAToo).
Task-Analysis-Tools (TAToo) este un instrument de analiz a muncii care se bazeaz pe
o serie de proceduri i metode testate empiric n
scopul obinerii unei descrieri ct mai complete
a sarciniilor postului analizat, care s permit
decelarea criteriilor de performan i identificarea caracteristicilor psihologice i comportamentale responsabile de reuit ntr-o anumit
profesie.
Task-Analysis-Tools (TAToo) a fost construit astfel nct s se desfoare n trei etape
succesive, interdependente: Tool 1- collect, Tool
2- group and Tool 3- assess, pe care le vom
expune pe scurt.
Step 1 Collect. Primul pas const n colectarea informaiilor de la profesioniti buni
cunosctori ai unui post de munc (de exemplu
120
deintorul postului respectiv, managerii, colegii

sau clienii). Datele obinute vizeaz informaii
privind: (a) obiectivele, sarcinile i responsabilitile postului, (b) cunotinele, abilitile i
calificrile necesare, (c) situaiile critice i
comportamentele adaptative deja ntlnite, dar
i (d) alte evenimente probabile n viitor.
ntrebrile specifice acestei etape sunt
derivate din Tehnica Incidentelor Critice
(Flanagan, 1954) i Metoda Knowledge, Skills
and Abilities (KSA). n funcie de condiiile
ntlnite de analist n practic, autorii propun
trei scenarii metodologice de culegere a informaiilor: interviul structurat, chestionarul sau
workshop-ul. Studii privind superioritatea uneia
sau alteia dintre cele trei variante, raportul costbeneficiu, calitatea vs. cantitatea datelor
obinute au fost realizate anterior de ctre Koch
(2010). Rezultatele arat c dei cu ajutorul
chestionarului s-a strns cel mai considerabil
volum de informaii, interviul ca metod singular de colectare este mai eficient dect chestionarul. De asemenea, s-a demonstrat c workshop-ul ar fi de preferat s fie utilizat ca metod
complementar, pe lng celelalte dou. Ideal ar
fi ns mbinarea celor trei metode pentru conturarea unui profil al cerinelor ct mai exact, ce
urmeaz s se realizeze n a doua etap.
Step 2 Group. Scopul acestei etape const n condensarea informaiilor obinute anterior ntr-un profil al cerinelor. Pattern-urlie identificate ca fiind responsabile de succes n
profesia analizat sunt dezbtute n cadrul unui
workshop. Experii, buni cunosctori ai profesiei, grupeaz descrierile comportamentelor, le
operaionalizeaz i le convertesc n cerine, n
funcie de similaritatea coninutului.
Step 3 Assess. Rezultatele sunt integrate
de ctre analist ntr-un chestionar care este
transmis participanilor pentru a se evalua
importana i frecven fiecrui criteriu operaionalizat, posibilitatea antrenrii sau compensrii cerinelor psiho-comportamentale i gradul
n care acestea acoper n mod corespunztor i
integral profilul postului analizat.
Toate materialele necesare implementrii
metodei TAToo (chestionar de culegere a datelor, ghid de interviu i workshop) se regsesc
n anexe.
4/17/2012
1:10 PM
Page 121
Recenzii
Partea final a crii, dedicat discuiilor

generale, ne prezint aspecte privind calitile
psihometrice ale TAToo precum i rezultate
comparative ale acestuia cu alte instrumente.
Din punct de vedere psihometric, TAToo
este o prob ce ntrunete exigenele unei
metode tiinifice. Obiectivitatea a fost determinat pentru a treia etap i exprimat ca msur
a acordului dintre evaluatori (peste 70-80%)
privind importana cerinelor psihologice i
comportamentale pentru un anumit post, deduse
din aplicarea TAToo.
Validitatea instrumentului a fost evaluat
pentru a doua etap a metodei, gruparea patternurilor comportamentale n cerine ale profilului,
prin dou studii. Astfel, peste 90% din comportamente au fost grupate n mod asemntor cu
cerinele n termeni de coninut i concept, iar
peste 70% dintre angajai consider c postul lor
a fost descris n mod corespunztor.
Validitatea i fidelitatea au fost demonstrate
i prin repetarea evalurii la un interval de 16
luni.
Tattoo a fost comparat cu Position Analysis
Questionnaire (PAQ) i cu Experience-based
intuitive approach (EBIA) pe un eantion de
manageri. Studiul comparativ a artat c exist
diferene n ceea ce privete structura i coninutul profilului de cerine. PAQ este mai degrab
preocupat de activiti i de mediul de munc,
rspunznd la ntrebarea Ce trebuie fcut?,
spre deosebire de TAToo care este orientat spre
deducerea comportamentelor, interesndu-se i
de Cum trebuie fcut acel ceva?. Un alt dezavantaj al PAQ este c necesit un nivel ridicat de
expertiz din partea uilizatorului.
Pe lng cerinele cognitive, EBIA i
TAToo se focalizeaz i pe o descriere comprehensiv a palierului motivaional, emoional i a
cerinelor sociale care sunt absolut necesare
pentru descrierea unui comportament managerial de succes.
Diferena esenial dintre TAToo i EBIA
const n gradul de specificitate. Pe lng lista
de cerine, TAToo ncorporeaz cteva criterii
observabile i specifice de operaionalizare a
fiecrei cerine. Acesta este dezavantajul major
n aplicarea EBIA din moment ce nu exist pai
suplimentari care s ghideze cerinele spre operaionalizare.
Diferenele privind utilitatea celor trei

metode de analiz a muncii au fost evideniate
din perspectiva managerilor n funcie de dou
scenarii, anume selecia de personal i evaluarea
de personal.
Amndou au fost evaluate ca fiind utile,
EBIA surprinznd sensibil mai bine cerinele
pentru selecia de personal, n timp ce TAToo
pentru evaluarea de personal. Principalul avataj
al metodei TAToo fa de EBIA este faptul c
cea din urm depinde mult de expertiza evaluatorului, fiind o metod subiectiv. inndu-se
cont de toate criteriile, TAToo a fost considerat
mai bun dect EBIA i n mod semnificativ mai
bun dect PAQ.
Evaluatorii au notat economicitatea ca fiind
rezonabil att timp ct TAToo se poate aplica
foarte bine i pe computer, materiale fiind necesare doar pentru workshop-urile din Tool 1 i 2.
Evaluarea utilitii per ansamblu a fost realizat n mod subiectiv de ctre managerii care
intenionau s utilizeze i n viitor acest instrument. Autorii sugereaz c studii urmtoare ar
trebui s se focalizeze i pe msuri obiective,
precum performana.
Putem concluziona c TAToo este un instrument uor de implementat chiar i de nespecialiti, fiecare etap fiind structurat n mod
riguros. Un avantaj major al TAToo este c a fost
proiectat innd cont att de exigenele tiinifice, ct i de realitile practicii. Cartea este
redactat sub forma unui manual explicit care
ofer toate informaiile i materialele necesare
implementrii unui astfel de program de analiz
a muncii.
M.A. Andreea Butucescu
Universitatea Ovidius, Constana

Beyonders. Transcending average
leadership, LannooCampus, 2011
In their new book Beyonders. Transcending average leadership, Vlerick professors

Herman Van den Broeck and David Venter
explain how Beyonders, as leaders, make a dif121
4/17/2012
1:10 PM
Page 122
Recenzii
ference. Professor Herman Van den Broeck

(Vlerick Leuven Gent Management School,
University of Ghent) is specialist in change
management, emotional intelligence and leadership. Professor David Venter (Vlerick Leuven
Gent Management School) specialises in negotiation and mediation. He is born in South Africa
and assisted Nelson Mandela to manage the
transition from apartheid to democracy.
This book came as an integration of perspectives and out of the box thinking around the
question What are the trademarks of visionary
leaders? To answer this question, the authors
set out to research the traits and interactions of
managers in the financial industry-an industry in
deep need of visionary leaders these days. As a
result of this research, the authors coined the
term Beyonders for these beyond ordinary
leaders.
A Beyonder is a person who possesses and
radiates a deep-seated vision anchored in a clear
set of values. His vision provides the beacons
that guide his decisions, not short-sighted financial criteria.
Beyonders think and act in an innovative
and holistic manner. What others see as obstacles, they welcome as opportunities. They
choose and create environments that inspire passion, and persevere until they have achieved
their goals. Beyonders ask for second opinions,
as they appreciate that yes-men do not contribute added value. The bottom line being that
average leaders kill the talent of their co-workers. Beyonders exude creativity, and are able to
122
generate a myriad of ideas and solutions for any

problem. They see opportunities that no one else
has even thought of. In stark contrast, average
leaders merely perpetuate and protect that which
worked in the past, shunning change.
Beyondership is the result of the interplay
between seven unique leadership characteristics. As such, the Beyonder model holds up
seven mirrors in which leaders can reflect upon
themselves and ponder how to become a Beyonder. The seven building blocks of beyondership
emerged from a rich qualitative research on
managerial staff in the finance industry. The
conclusions derived from these interviews were
both the basis for developing the theory, as well
as for creating an assessment tool for the beyonder style indicator.
However, Beyonders tells more than an
empirically supported theory. It is a book that
seeks to explain and illustrate with many examples what Beyondership means, and how to truly
become a Beyonder. It not only helps you assess
the leadership qualities of your boss, but also of
yourself.
Being a Beyonder is not about being the best
in the world, but being the best for the world.
The book and the underlying research were
made possible by Vlerick Prime Foundation
Partner, TriFinance, a company that strongly
aspires to implement Beyondership.
PhD. Smaranda Boro
Vlerick Leuven Gent Management School,
Belgium
4/17/2012
1:10 PM
Page 123
Profesorul Robert A. Roe n Romnia psihologia organizaional

la timpul viitor, septembrie 2011, Bucureti
Ioana Breazu1
Dr. Robert A. Roe (1944) este profesor

emerit de Teorie Organizaional i Comportament Organizaional la Universitatea din Maastricht i preedintele Federaiei Europene a Asociaiilor Psihologilor (EFPA). De asemenea, Dr.
Robert A. Roe susine prelegeri la invitaia unor
universiti precum: Universitatea din Trento
(Italia), Universitatea din Valencia (Spania),
Universitatea din Leipzig (Germania) i Universitatea British Columbia (Canada).
Nu n ultimul rnd, profesorul Robert A.
Roe este preedinte fondator al Asociaiei Europene de Psihologia Muncii i Organizaional
(EAWOP, 1991-1995) i director fondator al
Centrului de Cercetare Industrial i Organizaional din Tilburg (Olanda). De-a lungul carierei sale, Robert A. Roe a lucrat ntr-o arie larg
de domenii: selecie, dezvoltarea carierei,
tehnologia informaiei, performana n munc,
schimbare organizaional, cultur organizaional, impactul transformrilor economice i
metode de evaluare organizaional. n prezent,
interesul su de cercetare este orientat spre management i comportament organizaional dintr-o
perspectiv temporal, cu accent pe motivaie,
competen i performan.
n septembrie 2011, Robert A. Roe a vizitat
Romnia la invitaia Centrului de Psihosociologie al Ministerului Administraiei i Internelor
(MAI), a Asociaiei de Psihologie Industrial i
Organizaional (APIO) i a Colegiul Psihologilor din Romnia, susinnd o serie de conferine i workshop-uri n Bucureti.
Seminarul susinut la Centrul de Psihosociologie al Ministerului Administraiei i
Internelor din Bucureti a avut ca tem Selecia
personalului n contextul aderrii la spaiul
Schengen. n cadrul dezbaterilor s-a fcut

referire la un model ideal de a construi un sistem
de selecie de personal, care s ia n calcul ct
mai multe variabile. De asemenea, temele ce au
mai fost atinse au vizat experiene, provocri,
reglementri, perspective cu privire la selecia
personalului n instituiile din domeniul aprrii,
ordinii publice i siguranei naionale.
La ntlnirea cu studenii din cadrul colii
Naionale de Studii Politice i Administrative
Bucureti, au fost abordate mai multe subiecte,
dintre care menionm: istoricul psihologiei
muncii industriale i organizaionale n Europa,
poziia i rolul psihologiei n Europa din perspectiva EFPA i, nu n ultimul rnd, dezvoltri
recente cum ar fi: certificatul EuroPsy. Referitor
la poziia i rolul psihologiei n Europa, profesorul Robert A. Roe a evideniat modul n care
psihologia se difereniaz de alte discipline, precum i faptul c expertiza psihologic poate
contribui la elaborarea politicilor comunitare n
domenii ca: sport, sntate, educaie, transport,
munc, etc. A fost subliniat i faptul c, n actualul context european, psihologii pot face mai
mult dect s-i trateze n mod individual
pacienii, putnd contribui la schimbarea comportamentului, a condiiilor de via i munc a
unui numr ct mai mare de oameni.
n concluzie, putem aprecia c, ntlnirea cu
profesorul Robert A. Roe a reprezentat un eveniment important, att pentru studenii SNSPA,
ct i pentru psihologii din domeniul industrial/
organizaional, innd cont de nalta anvergur
tiinific a invitatului, precum i de importana
i actualitatea subiectele abordate n cadrul discuiilor.
SNSPAFCRP, Bucureti.
Adresa de coresponden: ioana_breazu@yahoo.com
123
4/17/2012
1:10 PM
Page 124
Psycho-Social Health in Organizations: Research and Interventions

(International OHP Workshop), 15-17 December 2011, Timisoara, Romania
Gabriel Fischmann1
At the end of 2011, the Psychology Department of the West University of Timisoara hosted an international workshop in Organizational
Health Psychology (OHP). The workshop was
organized by Dr. Coralia Sulea (West University of Timisoara) in collaboration with Dr. Eva
Cifre Gallego (University Jaume I, Castelln),
the event being supported by CNCS-UEFISCDI, project no. PN II-IDEI/WE-2011-054).
The workshop was remarkable in many
ways, perhaps the most important one being the
impressive list of invited lectures; well-known
Romanian and foreign psychology researchers
agreed to meet and share details of their current
and future work.
The following topics were discussed: state of
the art and future directions in the study of work
engagement (Dr. Wilmar Schaufeli, Utrecht University), an evidence-based approach regarding
the monetary value of interventions in employee
well-being (Dr. Dragos Iliescu, SNSPA,
Bucharest), an overview of current and possible
future research on job insecurity (Dr. Hans De
Witte, Leuven University), research strategies
and recommendations for a scientific approach in
OHP (Dr. Florin A. Sava, West University of
Timisoara), ways to optimize psychosocial health
at work (Dr. Eva Cifre Gallego), the current state
of OHP in Romanian organizations (Dr. Delia
Virga, Dr. Irina Macsinga and Dr. Coralia Sulea,
West University of Timisoara). The workshop
also included a round table and a working session, where Dr. Corina Ilin (West University of
Timisoara) presented national and international
funding opportunities for OHP research and Dr.
Eva Cifre Gallego moderated the discussion
about international partnerships and cross-cultural future research projects.
West University of Timisoara.

Adresa de coresponden: gabriel.fischmann@yahoo.com
124
The presentations received a warm welcome from the actively involved audience,
which was quite mixed, comprising psychology
researchers, practitioners and graduate students.
For the students of the Organizational and
Occupational Health Psychology master program, attending this workshop was a rare opportunity to actually meet in person worldrenowned scientist whose work is mandatory
class reading. The Romanian and foreign
researchers were offered the chance, through the
round tables and work sessions, to create new
research groups, based on their common and
complementary interests, and to identify funding possibilities for this research. All workshop
participants were made aware of the existing
dualism in organizational and occupational
health psychology, where research may be performed by studying either the dark side (e.g.,
stressors, ill-being) or the bright side (e.g., work
engagement, well-being).
Theory and practice were masterfully combined by the lecturers, all presentations having
both a solid theoretical background and either
case studies or practical recommendations,
regarding how the existing knowledge can be
used to help improve employees and organizations health and well-being, while maximizing
their effectiveness and performance. Special
care was given to the distinction between scientific and pseudo-scientific approaches to organizational problems, and to the way in which OHP
can benefit both the organizational and the scientific domain. Invited lecturers presentations
and other details can be downloaded from:
http://occupationalhealthpsychology.wordpress.com/international-ohp-workshop/
4/17/2012
1:10 PM
Page 125
Psycho-Social Health in Organizations: Research and Interventions (International OHP Workshop)
By the end of the three days, the participants were already discussing the option that
this should not remain a singular event, and that
future versions of this workshop should happen
on a regular basis. It was a very productive and
pleasant opportunity for all parties involved to
share and to receive information, to form new
teams and to decide on future research directions. It must definitely happen again.
We are now looking forward to see each
other again at APIOs forthcoming conference in
Timisoara: Occupational Health Psychology: a
challenge for organizations (for further details,
please visit http://www.apio.ro/conferinta).
125
4/17/2012
1:10 PM
Page 126
Memento prof. univ. dr. Nicolae Jurcu sau Despre profesionalism

i omenie prof. univ. dr. Nicolae Jurcu
Prof. univ dr. Adrian Opre1
A trecut puin timp de cnd a plecat dintre

noi, pstrnd pn la capt senintatea chipului
i cldura adresrii profesorul univ. dr. Nicolae
Jurcu. O personalitate remarcabil, un om cu o
inut moral i academic deosebit. Bonomia,
modestia i generozitatea i-au fost carte de vizit i sincer ntmpinare pentru toi cei ce i-au
deschis ua biroului sau clcat pragul casei. Pentru cei din generaia mai tnr o necesar i
bineprimit punte de legtur cu prinii colii
romneti de psihologie. Graiei profesorului
Jurcu am avut privilegiul de a ne mprti i
noi din valorile pe care le-au creat promotorii i
continuatorii colii clujene de psihologie. Vitalitatea i rvna domniei sale pentru cunoatere au
nvins nu doar ineria fixaiilor paradigmatice
dar i piedicile dogmatice ale unui secol mult
prea zbuciumat.
Nicolae Jurcu s-a nscut n satul Hotrel,
judeul Bihor la data de 15.03.1940. A urmat
cursurile colii generale n zona natal, dup
care pe cele ale Liceului teoretic din oraul tei.
Dup susinerea examenului de bacalaureat este
admis la Universitatea Babe-Bolyai, obinnd n 1965 licena n Filosofie, specializarea
Psihologie. n anul 1974 obine titlul de doctor
n psihologie cu o tez coordonat de acad.
Alexandru Roca intitulat Diagnosticarea i
formarea aptitudinilor pentru profesiuni din industria textil.
Debutul profesional i-l face la Liceul nr. 2
din Dej, iar dup un an ncepe s desfoare o
asidu activitate de cercetare la Filiala din Cluj
a Academiei Romne. Aici rmne pn n anul
1976, acumulnd o valoroas experien ce va fi
ulterior mprtit discipolilor si, cu mult tact
i miestrie pedagogic. ncepnd din anul 1976
pn n anul 1983 va lucra ca lector universitar
Universitatea Babe Bolyai, Cluj Napoca.

Adresa de coresponden: adrianopre@psychology.ro
126
la Institutul Central pentru Perfecionarea Personalului Didactic Filiala Cluj i la Institutul

Politehnic din acelai ora. Ascensiunea academic a domnului Nicolae Jurcu continu,
astfel c n 1983 obine titlul de confereniar, iar
n anul 1990 pe cel de profesor universitar la
Universitatea Tehnic din Cluj-Napoca. Paralel
cu activitatea didactic i cea de cercetare deine
i o serie de funcii: eful Catedrei de tiine
Socio-Umane 1989-1997, iar din 1997 a ocupat
poziia de Dir. al Departamentului pentru Pregtirea Personalului Didactic, din UTCCJ.
n ceea ce privete activitatea tiinific
profesorul Nicolae Jurcu a acoperit, cu competen i responsabilitate, trei subdomenii aplicative ale psihologiei: psihologia organizaional,
industrial, i educaional. Fiind un excelent
cadru didactic, preocuprile sale s-au ndreptat
in ultimii ani din via mai apsat nspre domeniul Psihologiei Educaionale i a Pedagogiei
mprtind att mai tinerilor si colaboratori,
ct i studenilor din experiena acumulat.
ntreg parcursul profesional al profesorului
Nicolae Jurcu este unul ce reflect fidelitatea
pentru specializarea psihologie i credina c
psihologia este una din tiinele cele mai importante care abordeaz omul i probleme sale.
Profesorul Nicolae Jurcu este autor a peste
25 de lucrri de specialitate, relevante pentru
domeniul psihologie, publicate n edituri de
prestigiu, precum i peste 100 de studii i articole, pe care le putem gsi n reviste i culegeri
din ar i strintate. Suntem, din nefericire,
sufocai astzi de redundana operelor acelora
care vorbesc cu mult emfaz despre bunele
practici psihologice sau pedagogice, fr a avea
ns ndreptirea acestor asumri. Ei ne ofer,
demonstrativ, proceduri de intervenie corectiv
4/17/2012
1:10 PM
Page 127
Memento prof. univ. dr. Nicolae Jurcu sau Despre profesionalism i omenie prof. univ. dr. Nicolae Jurcu
sau formativ pe care nu le-au folosit nici mcar

o singur dat n activitatea lor practic. Contrastnd acestora, profesorul Nicolae Jurcu ne
ntmpin cu lucrri care poart n ele, nu numai
cunoaterea larg mprtit a comunitii academice dar, mai ales, amprenta expertizei unui
om de carier. Domnia sa a reuit s depeasc,
n chip minunat, condiia modest a acelor
nvcei care se mulumesc s invoce ndelunga lor experien didactic (fr efecte autoformatoare) pentru a-i impune, evident nefondat,
autoritatea n domeniu. Nicolae Jurcu a fost
dasclul care i-a valorificat superior experiena
reuind s o transpun n veritabile secvene formative, altruist mprtite colegilor de breasl.
Analizndu-i activitatea publicistic, oricine
poate lesne sesiza c autorul nu este un simplu
intermediar n perpetuarea metodelor didactice
ci, mai degrab un actor foarte activ n preschimbarea lor. Mai mult, dovedind o sensibilitate aparte la legtura profund i deopotriv
tainic dintre psihologie, pedagogie i realitatea
social Nicolae Jurcu s-a strduit ca, prin studiile sale, s pstreze n minile generaiei actuale
imaginea vie a valorilor i modelor umane autentice. n aceste scrieri discursul su depete
relaia teoretic dintre psihologie i pedagogie,
relevndu-ne mai degrab impactul tandemului
acestora asupra societii i culturii n general.
O prezenta vie n ntrunirile comunitii
tiinifice profesorul Jurcu a participat la peste
48 de manifestri tiinifice naionale i internaionale. Este autorul a dou invenii brevetate
i a trei inovaii, director sau colaborator la peste
15 contracte de cercetare. Excelenta pregtire
profesional, dublat de caliti personale deosebite explic alegerea sa ca membru al numeroase asociaii naionale i internaionale dar
i recunoaterea i aprecierea explicit i implicit a valorii sale academice i umane.
Profesorul Nicolae Jurcu a demonstrat, de
asemenea, extrem de mult sensibilitate fa de
nevoile studenilor. El a neles mai bine dect
muli dintre noi c principala surs a satisfaciei
profesionale a unui cadru didactic universitar o

constituie calitatea relaiilor sale cu studenii.
Adesea nu sesizm sau mai degrab refuzm s
acceptm faptul c atitudinile, afirmaiile i ntreaga noastr purtare poate avea un impact
major asupra personalitii lor. Amprenta noastr asupra minii celor ce ne ascult entuziati
sau ne suport stoic la orele de curs i seminar
nu o putem sesiza ns imediat. Cuvintele sdite
n minile lor au i ele nevoie de timp, reclam
i ele o perioad de gestaie. Atunci ns cnd ies
la lumin, ele reflect o mbinare plcut i original ntre gndul nostru druit i nota de subiectivitate a improprierii lui. La ei nu vom mai
regsi ideea original aa cum am sdit-o, dar
nici primitorul nu mai e cel care a fost la
momentul receptrii ei. Aceasta este metamorfoza ideilor i a oamenilor la care profesorul
Nicolae Jurcu a fost pe deplin prta.
Tot ei, studenii sunt ntr-o permanent cutare de repere profesionale, ei sper s i gseasc printre noi adevraii mentori. Din pcate
ns, modelele contemporane sunt tot mai rare i
din ce n ce mai istovite; lor parc le lipsete
ceva, probabil tocmai acea dimensiune ce a conferit for i carism elitei intelectuale din prima
jumtate a secolului trecut. Romnia interbelic
ne-a druit eroi i martiri dar, mai cu seam, nea druit caractere. Studentul nostru de astzi,
aflat la vrsta mplinirii caracterului i dorete
i caut tocmai aceste modele. Optimismul m
ndeamn s cred ca nc e posibil ca aceti
tineri s se bucure de adevrate valori. Pentru ei
profesorul Jurcu a fost, fr dar i poate, un
veritabil model profesional i caracterial.
Calm, echilibrat, cu (bun)sim al umorului,
spirit critic i mai ales autocritic, perseverent,
tenace, cu dragoste de adevr i de armonie ntre
oameni, empatic i inteligent iat chipul unui
om hrzit i druit s fie dascl. O personalitate
remarcabil care nu i-a ngropat talantul ci l-a
folosit cu nelepciune pentru a ne drui nou un
rod nsutit.
127
4/17/2012
1:10 PM
Page 128
PSIHOLOGIA RESURSELOR UMANE

Ghid pentru autori
Editorii
Acest document reprezint Ghidul pentru

Autori. Cuprinde instruciuni privind formatul
i limbajul care trebuie utilizat pentru manuscrisele trimise la revista Psihologia Resurselor
Umane. De asemenea, acest document poate fi
gsit i pe site-ul Asociaiei de Psihologie Industrial i Organizaional (www.apio,ro).
Acest Ghid Pentru Autori este alctuit
dup Manual de Publicat APA, Ediia a-VI-a.
Trimiterea i formatul manuscrisului

Toate manuscrisele pentru revista Psihologia Resurselor Umane trebuie trimise la urmtoarea adres de e-mail: revista@apio.ro.
Pentru redactarea manuscrisului v rugm
s utilizai fontul Times New Roman 12, spaierea de 1,5 i setrile paginii A4. Fiecare pagin va fi numerotat n colul drept de sus.
Marginea de sus i de pe lateral trebuie s aib
cel puin 2,5.cm sau 1 inch. Un exemplu complet de manuscris poate fi gsit n Manual de
Publicat APA, Ediia a-VI-a.
Publicaii
Lucrrile acceptare sunt re-editate. Autorii
trebuie s revizuiasc i s fac corecturi n noua
lucrare editat. Editorul revistei Psihologia Resurselor Umane va contacta autorul corespondent, dup ce n prealabil lucrarea a fost acceptat pentru a fi inclus ntr-un numr al revistei.
Dac lucrarea dumneavoastr a fost acceptat, v rugm s anunai editorul n ceea ce
privete schimbrile datelor dumneavoastr de
contact sau absenele ndelungate.
128
Pagina copert
Prima pagin a manuscrisului trebuie s includ urmtoarele informaii:
1. Titlu
Un titlu trebuie s fie o precizare concis a
temei principale a lucrrii i s permit identificarea aspectelor teoretice sau a variabilelor examinate i a relaiei dintre ele. Titlul trebuie s fie
redactat cu litere mari i mici (sentence case), s
fie centrat n raport cu marginile paginii i poziionat n partea superioar a acesteia.
2. Numele autorului i afilierea
(afilierile) instituionale
Numele autorului va fi prezentat n urmtoarea form: Primul prenume, iniialele altor
prenume i numele de familie.
Afilierea instituional trebuie s reflecte
instituia/locaia cu care a fost asociat autorul n
momentul derulrii cercetrii. Dac un autor nu
are afiliere instituional, trebuie specificate
oraul i ara de reziden sub numele autorului.
Afilierea instituional ar trebui centrat sub
numele autorului, n rndul urmtor.
3. Nota despre autor
Aceast seciune include urmtoarele:
Primul paragraf trebuie s includ afilierea
departamental la momentul cercetrii pentru toi autorii, dup urmtorul model:
numele autorului, exact cum apare la
seciunea anterioar, virgul, numele departamentului, virgul, numele universitii,
punct i virgul, numele urmtorului autor
.a.m.d. iar la sfrit se adaug un punct.
Al doilea paragraf trebuie s includ schimbrile privind afilierea autorului, survenite
dup realizarea cercetrii, dup urmtorul
model: [numele autorului] este acum la
[afilierea].
4/17/2012
1:10 PM
Page 129
Ghid pentru autori
Al treilea paragraf trebuie s includ mulumiri (numai pentru sprijinul financiar din
granturi sau de alt tip, orice alte nelegeri
speciale privind drepturile de autor, mulumiri pentru sprijin personal) i circumstane
speciale (care trebuie prezentate naintea
mulumirilor).
Al patrulea paragraf trebuie s includ
informaii despre autorul de contact: adresa
potal i o adres de e-mail.
Nota despre autor va fi plasat n pagina de
titlu, dup titlu, autori i afiliere. Sintagma
Nota despre autor va fi poziionat centrat.
Fiecare paragraf va fi scris separat. Nota despre
autor nu este numerotat sau citat n text.
Explorarea importanei problemei. Aceast

seciune prezint motivul pentru care problema necesit o nou cercetare. Autorul va
prezenta aceast problem n funcie de
tipul de cercetare (studiu empiric, review
sistematic i meta-analiz, lucrare metodologic sau studiu de caz).
Descrierea literaturii relevante i evidenierea continuitii logice ntre cercetrile
anterioare i cercetarea propus.
Precizarea fiecrei ipoteze formulate i
oferirea unui argument teoretic privind
modul n care a fost desprins din teorie sau
conectat logic cu studiile anterioare.
Metod
Pagina de rezumat
Rezumatul (abstract) i titlul lucrrii sunt
plasate pe pagina 2. Rezumatul nu trebuie s
depeasc 150 de cuvinte. Eticheta Rezumat
trebuie s apar cu litere mari i mici, centrat, n
partea de sus a paginii. Rezumatul trebuie s
aib un singur paragraf, adic s fie redactat fr
alineate. Autorul va propune i un titlu scurt.
Rezumatul va fi scris n limba Englez,
Francez i Romn. Este necesar s fie incluse
3-5 cuvinte cheie dup fiecare rezumat, n cele
trei limbi.
Paginile textului principal

n pregtirea manuscrisului, autorul va ncepe cu o introducere care va fi plasat pe pagina 3. Titlul manuscrisului va fi scris cu litere
mari i litere mici, centrat n partea de sus a
paginii i urmat de coninutul textului lucrrii.
Seciunile urmtoare ale lucrrii vor fi
prezentate fr spaii libere. Cnd ncepe o nou
seciune, aceasta nu trebuie poziionat pe o
pagin nou.
Aceast parte a lucrrii trebuie s includ:
Prezentarea problemei. Aceast seciune
prezint problema specific care va fi
investigat i descrie strategia de cercetare.
Aceast seciune nu trebuie etichetat ca
Introducere.
Aceast seciune descrie detaliat cum a fost

realizat studiul, incluznd i definiiile conceptuale i operaionale ale variabilelor utilizate n
studiu. Autorul ar trebui s includ:
Descrierea eantionului, prin descrierea
caracteristicilor majore ale acestuia, n special a caracteristicilor care pot conta n
interpretarea rezultatelor.
Procedura de eantionare, prin descrierea
modalitii de selecie a participanilor:
metoda de eantionare, procentul celor care
au fost contactai i au participant la cercetare, numrul participanilor care s-au oferit
s participe la cercetare etc.
Mrimea eantionului, putere i precizie.
Msurtorile prin descrierea metodelor utilizate pentru a colecta datele i a mbogi
calitatea msurtorilor.
Designul de cercetare.
Manipulri experimentale sau proceduri.
Descrierea sarcinilor.
Rezultate
Aceast seciune sumarizeaz datele colectate i analiza datelor realizat pentru a testa
ipotezele propuse. Autorul trebuie s raporteze
analiza datelor ct mai detaliat, astfel nct s
permit justificarea concluziilor. Pentru mai
multe informaii, autorul va consulta Manualul
APA, Ediia a-VI-a.
129
4/17/2012
1:10 PM
Page 130
Ghid pentru autori
Discuii
Aceast seciune evalueaz i interpreteaz
implicaiile rezultatelor, autorii fcnd referire
la ipotezele propuse. Autorul va examina, interpreta, cataloga rezultatele i va face inferene pe
baza lor. Autorul va insista pe consecinele teoretice sau practice ale rezultatelor obinute. De
asemenea, trebuie prezentate limitele studiului
i ale direciilor viitoare de cercetare.
Bibliografie
Referinele sunt citrile n ordinea alfabetic de la sfritul lucrrii. Aceast list trebuie
s includ toate lucrrile citate n cadrul
manuscrisului. Referinele trebuie scrise dup
urmtorul model:
1. Reviste (exemple selective)
Autor, A.A, Autor, B. B., & Autor, C.C.
(an). Titlul articolului. Titlul Jurnalului, xx, pppp. doi: xx.xxxxxxxxxx
Autor, A.A., Autor, B.B., Autor, C.C.,
Autor, D.D., Autor, E.E., Autor, F.F., Autor,
Y.Y. (an). Titlul articolului. Titlul Jurnalului, xx,
pp-pp. doi: xx.xxxxxxxxxx
Autor, A.A, Autor, B. B., & Autor, C.C. (an).
Titlul articolului. Titlul Jurnalului, xx, pp-pp.
Autor, A.A., & Autor, B.B. (in press). Titlul
articolului. Titlul Jurnalului. Retrieved from
http://cogprints.org/5780/1/ECSRAP.F07.pdf
2. Cri
Autor, A.A. (an). Titlul lucrrii. Locaie:
Editur.
Autor, A.A. (an). Titlul lucrrii. Retrieved
from http://www.xxxxxxx
Autor, A.A. (an). Titlul lucrrii. doi: xxxxx
Editor, A.A. (Ed.) (an). Titlul lucrrii.
Locaie: Editur.
3. Capitole din cri (exemple selective)
Autor, A.A., & Autor, B.B. (an). Titlul capitolului. In A. Editorul, B. Editorul, & C. Editorul (Eds.), Titlul crii (pp. xxx-xxx). Locaie:
Editur.
Autor, A.A, & Autor, B.B. (an). Titlul capitolului. In A. Editorul, B. Editorul, & C. Edi-
130
torul (Eds.), Titlul crii (pp. xxx-xxx). Retrieved from http://www.xxxxxxx

Autor, A.A., & Autor, B.B. (an). Titlul capitolului. In A. Editorul, B. Editorul, & C. Editorul (Eds.), Titlul crii (pp. xxx-xxx). Locaie:
Editur. doi: xxxxxxxx.
4. Conferine i simpozioane
(exemple selective)
Contributor, A.A., Contributor, B.B., Contributor, C.C., & Contributor, D.D. (an, lun).
Titlul lucrrii. In E.E. Chairperson (Chair),
Titlul simpozionului. Symposium conducted at
the meeting of [Numele Organizaiei], Locaia.
Presenter, A.A. (an, lun). Titlul lucrrii sau
posterului. Paper or poster session presented at
the meeting of [Numele Organizaiei], Locaie.
4. Lucrri nepublicate
(exemple selective)
Author, A.A. (an). Titlul manuscrisului.
Unpublished manuscript [or "Manuscript submitted for publication," or "Manuscript in
preparation"].
Pentru o descriere detaliat a procedurii
privind citarea altor tipuri de lucrri dect cele
listate anterior, autorii vor consulta Manualul
APA, Ediia a-VI-a.
Note de subsol
Notele de subsol sunt utilizate pentru a oferi
informaii suplimentare sau pentru a confirma
statutul drepturilor de autor.
Anexe
Anexele manuscrisului (etichetate ANEXA 1,
ANEXA 2 etc.) conin materiale suplimentare fa
de coninutul lucrrii, cum ar fi informaii legate de
proceduri metodologice lungi, calcule etc.
Tabele i figuri
Autorul trebuie s numeroteze toate tabelele i figurile cu cifre arabe, n ordinea n care
au fost menionate pentru prima dat n textul
4/17/2012
1:10 PM
Page 131
Ghid pentru autori
manuscrisului, indiferent dac o discuie mai

detaliat a tabelului sau figurii este prezent
ulterior n text. Autorul ar trebui s le eticheteze
Tabelul 1, Tabelul 2 .a.m.d. sau Figura 1, Figura 2 .a.m.d.. Prezentai prima dat toate
tabelele, ulterior figurile. Plasai tabelele i figurile dup anexele de la sfritul manuscrisului i
indicai poziia fiecruia/fiecreia n text astfel:
-----------------------------------Inserai Tabelul 1 aici
-----------------------------------Fiecare tabel sau figur trebuie s aib o
propoziie de introducere n text. Formatul
acceptat este cel standard (canonic). Fiecare
tabel trebuie s raporteze un singur tip de analiz (care trebuie s fie identificat din denumirea tabelului) i fiecare coloan i rnd trebuie s conin sun singur tip de date.
Citri
Este important ca autorul s menioneze
fiecare lucrare citat n manuscris n seciunea
Bibliografie. Autorii pot cita n text astfel:
1. Un singur autor
Numele i anul: S-a artat c X este asociat
cu Y (Autor, an)
Numai anul: Autorul (an) a artat c
2. Doi autori sau mai muli autori
Cnd o lucrare are doi autori, autorul trebuie s citeze numele ambilor autori ori de cte
ori apare referina n text.
Cnd o lucrare are trei, patru sau cinci autori, autorul trebuie s citeze toi autorii prima
dat. n citrile ulterioare va fi inclus numele primului autor urmat de sintagma et al. (fr Italic
i cu un punct dup al.) i anul apariiei lucrrii.
3. Dou sau mai multe lucrri citate

Autorul trebuie s ordoneze citrile n ordine
alfabetic. Dou sau mai multe lucrri ale aceluiai autor (sau ale aceluiai grup de autori) publicate n acelai an vor fi marcate prin adugarea
unui a, b .a.m.d. dup anul publicrii.
4. Lucrri fr autor identificat
sau Autor anonim
Cnd o lucrare nu are un autor identificat,
autorul trebuie s citeze n text titlul lucrrii i
anul. Se va poziiona ntre ghilimele titlul articolului, capitolului sau a paginii web i se va
marca cu Italic numele revistei, crii, brourii
sau a raportului:
privind angajamentul organizaional
(Study Report, 2011)
cartea Corelate motivaionale (2011)
5. Numrul paginilor n citri
Pentru a cita o parte specific dintr-o anumit surs, autorul trebuie s indice pagina,
capitolul, figura, tabelul sau formula. ntotdeauna va fi menionat numrul paginii:
(Johnny, 2011, p. 13)
6. Sursele secundare
Cnd sursele originale nu sunt disponibile
n format fizic, autorul trebuie s menioneze a
doua surs in lista bibliografic iar n text s
menioneze lucrarea original i s citeze sursa
secundar:
raportul elaborat de Minnie (citat n Smith,
2011).
V mulumim pentru atenia acordat

instruciunilor specificate n acest ghid va
uura munca celor implicai n publicarea acestei reviste.
131
4/17/2012
1:10 PM
Page 132
HUMAN RESOURCES PSYCHOLOGY

Guide for authors
The Editors
This document represents the Guide for

Authors. It covers the format and language to
be used for manuscripts submitted to Human
Resources Psychology. Also, this document can
be found on the webpage of the Romanian Association of Industrial and Organizational Psychology (www.apio.ro).
This Guide for Authors follows the 6th
APA Publication Manual.
Manuscript Submission and Format

All manuscripts for the journal Human
Resources Psychology should be submitted to
the following e-mail address: revista@apio.ro.
To edit the manuscript please use Times
New Roman 12-point type, 1.5 line spacing and
the A4 page setting. Each page will be numbered in the upper right corner. The top and side
margins should be left of at least one inch or
2.54 cm. A full example of a manuscript can be
found in the 6th APA Publication Manual.
Publications
Accepted papers are copy-edited and
retyped. Authors have to review edits and proofread their work. The editor of Human Resources
Psychology will contact the corresponding author
after the editor assigns your work to an issue.
If your work is accepted, please keep the
editor informed of changes in your contact
information and of long absences.
Front Page
The first page of the manuscript should
include the following information:
132
1. Title
The title should be a concise statement of
the main topic and should identify the variables
or theoretical issues under investigation and the
relationship between them. It should be typed
in sentence case, centered between left and
right margins, and positioned in the upper half
of the page.
2. Author name(s)
and institutional affiliation(s)
Author name(s) will be presented in the following form: first name, middle initial(s), and
last name.
Institutional affiliation should reflect the
institution/location where the author(s) were
when the research was conducted. When an
author has no institutional affiliation, the city
and state of residence below the authors name
should be specified. The institutional affiliation
should be centered under the author's name, on
the next line.
3. Authors note
This section should include the following:
First paragraph should include the departmental affiliations at the time of the study
for all authors as follows: name of the
author as it appears in the byline, comma,
department name, comma, university name,
semicolon, next author name, and so on,
and end with a period.
Second paragraph should include any
changes in author affiliation subsequent to
the time of the study as follows: [authors
name] is now at [affiliation].
Third paragraph should include acknowledgments (only for grants or other financial
support, any special agreements concerning
authorship, thanks for personal assistance)
and special circumstances (disclose them
before the acknowledgements in this paragraph).
4/17/2012
1:10 PM
Page 133
Guide for authors
Fourth paragraph should include information about the person to contact in terms of
mailing address and e-mail.
Place the author note on the title page,
below the title, byline, and affiliation. Center the
label Author Note. Start each paragraph of the
note with an indent, and type separate paragraphs for the authors' names and current affiliations, changes in affiliations, acknowledgments, and special circumstances, if any, along
with the person to contact. The author note is
not numbered or cited in the text.
Abstract Page
The abstract as well as the title of the work
go on page 2. The abstract should be no longer
than 150 words. The label Abstract should
appear in sentence case, centered, at the top of
the page. Type the abstract itself as a single
paragraph without paragraph indentation. Place
a running head (short title).
The abstract will be written in English,
France and Romanian. It is necessary to include
3-5 key words after each abstract, in all these
three languages.
Main body text pages

In preparing your manuscript, begin the
introduction on page 3. Type the title of the
manuscript in sentence case centered at the top
of the page, and then type the text. The remaining sections of the article follow each other
without a break; do not start a new page when a
new heading occurs.
This section should include the following:
Introduction of the problem. This section
will present the specific problem under the
study and describe the research strategy.
There is no need to label this section as
Introduction.
Explore importance of the problem. This
section states why the problem deserves
new research. State explicitly this problem
according to the type of the study (empirical study, literature review and meta-analysis, methodological paper and case study).
Describe relevant scholarship by discussing

the relevant related literature and demonstrating the logical continuity between previous and present work.
State each tested hypothesis clearly and
provide a theoretical argument for how it
was derived from theory or is logically connected to previous data and argumentation.
Method
This section describes in detail how the
study was conducted, including conceptual and
operational definitions of the variables used in
the study. Authors should include the following:
Sample description, by describing the main
characteristics with particular emphasis on
characteristics that may have bearing on the
interpretation of results.
Sampling procedure by describing the procedures for selecting participants in terms
of sampling method, the percentage of the
sample approached that participated, the
number of participants who selected themselves into the sample.
Sample size, power and precision.
Measures and covariates by describing the
methods used to collect data and to enhance
the quality of the measurements.
Research design.
Experimental manipulations or procedures.
Task description.
Results
This section summarizes the collected data
and the analysis performed on the data to test
the proposed hypotheses. Report the data analysis in sufficient detail to justify your conclusions. For more information please consult the
6th APA Publication Manual.
Discussion
This section evaluates and interprets the
implications of the results, especially with
respect to original hypotheses. Examine, inter133
4/17/2012
1:10 PM
Page 134
Guide for authors
pret, and qualify the results and draw inferences

and conclusions from them. Emphasize any theoretical or practical consequences of the results.
Also, the limits of the study and possible
future studies can be considered in this section.
References
References are your entries in the alphabetical list at the end of your article or research
note. This list should include all the works you
have cited throughout the manuscript. The references should be formatted as follows:
1. Periodicals (selective examples)
Author, A.A, Author, B. B., & Author, C. C.
(year). Title of article. Title of Periodical, xx,
pp-pp. doi: xx.xxxxxxxxxx
Author, A. A., Author, B. B., Author, C. C.,
Author, D. D., Author, E. E., Author, F.F.,
Author, Y.Y. (year). Title of article. Title of Periodical, xx, pp-pp. doi: xx.xxxxxxxxxx
Author, A.A, Author, B. B., & Author, C. C.
(year). Title of article. Title of Periodical, xx,
pp-pp.
Author, A.A., & Author, B.B. (in press). Title
of article. Title of Periodical. Retrieved from
2. Books
Author, A. A. (year). Title of work. Location: Publisher.
Author, A. A. (year). Title of work.
Retrieved from http://www.xxxxxxx
Author, A. A. (year). Title of work. doi:
xxxxx
Editor, A. A. (Ed.) (year). Title of work.
Location: Publisher.
3. For chapters in a book or entry
in a reference book (selective example)
Author, A.A., & Author, B.B. (year). Title
of chapter or entry. In A. Editor, B. Editor, & C.
Editor (Eds.), Title of book (pp. xxx-xxx). Location: Publisher.
Author, A.A, &Author, B.B. (year). Title of
chapter or entry. In A. Editor & B. Editor (Eds.),
Title of book (pp. xxx-xxx). Retrieved from
http://www.xxxxxxx
134
Author, A.A., & Author, B.B. (year). Title

of chapter or entry. In A. Editor, B. Editor, & C.
Editor (Eds.), Title of book (pp. xxx-xxx). Location: Publisher. doi: xxxxxxxx
4. Meeting and symposia
(selective examples)
Contributor, A.A., Contributor, B.B., Contributor, C.C., & Contributor, D.D. (Year,
Month). Title of contribution. In E.E. Chairperson (Chair), Title of symposium. Symposium
conducted at the meeting of Organization Name,
Location.
Presenter, A. A. (Year, Month). Title of paper
or poster. Paper or poster session presented at the
meeting of Organization Name, Location.
5. Unpublished works
(selective examples)
Author, A.A. (Year). Title of manuscript.
preparation"].
For a detailed description of the procedure
related to the citation of other types of work
than those listed above, consult the 6th APA Publication Manual.
Footnotes
Footnotes are used to provide additional
content or to acknowledge copyright permission
status.
Appendices
The appendices of the manuscript (labeled
APPENDIX A, APPENDIX B etc.) contain
materials that supplements article content such
as lengthy methodological procedures, calculations of measures, scales etc.
Tables and Figures

The author should number all tables and
figures with Arabic numerals in the order in
which they are first mentioned in the text,
4/17/2012
1:10 PM
Page 135
Guide for authors
regardless of whether a more detailed discussion

of the table or figure occurs later in the paper.
The author should label them as Table 1, Table
2, and so on or Figure 1, Figure 2, and so on.
List all tables first followed by figures. Place
tables and figures after appendices at the end of
the manuscript, and indicate the position of each
in the text as follows:
-----------------------------------Insert Table 1 about here
-----------------------------------Each table or figure needs an introductory
sentence in your text. The format agreed is the
standard (canonical) one. Each table should
report one type of analysis (which is identified
in the title), and each vertical column and horizontal row should contain only one type of data.
Citation
It is important to put in the Reference section
every work you have cited throughout the
manuscript. The author can cite in-text as follows:
1. One author
Name and year: It has been found that X is
associated with Y (Author, year)
Year only: Author (year) has found that
2. Two authors
When a work has two authors, the author
should cite both names every time the reference
occurs in the text.
When a work has three, four, or five
authors, you should cite all authors the first time
the reference occurs but in the subsequent citations, include only the surname of the first
author followed by et al., (not Italicized and
with a period after al.) and the year.
3. Two or more cited works

The author should order citations alphabetically. Designate two or more works by one
author (or by an identical group of authors) published in the same year by adding a, b, and
so forth, after the year.
4. Works with no identified author
or with an Anonymus author
When a work has no identified author, the
author should cite in text the first few words of
the reference list entry (usually the title) and the
year. Use double quotation marks around the
title of an article, a chapter, or a web page and
italicize the title of a periodical, a book, a
brochure, or a report:
on organizational commitment (Study
Report, 2011)
the book Motivational Outcomes (2011)
5. Page numbers in citations
To cite a specific part of a source, the author
should indicate the page, chapter, figure, table,
or equation at the appropriate point in text.
Always give page numbers for quotations.
(Johnny, 2011, p. 13)
6. Secondary sources
When the original work is out of print,
unavailable through usual sources, the author
should give the secondary source in the reference list and in the text you should name the
original work and give a citation for the secondary source
Minnies report (as cited in Smith, 2011).
Thank you for paying attention to the conventions outlined in this guide it will help the
work of everyone involved in the publication of
this journal.
135
4/17/2012
1:10 PM
Page 136
PSYCHOLOGIE DES RESSOURCES HUMAINES

Guide pour les auteurs
Les diteurs
Ce document reprsente le Guide pour les

auteurs. Il couvre le format et la langue utiliser pour les manuscrits soumis la Psychologie
des Ressources Humaines. En outre, ce document peut tre trouv sur le Web ladresse
suivante: ... ...
Ce Guide pour les auteurs suit le Manuel
de Publication APA, 6me dition.
Soumission de manuscrit et Format

Tous les manuscrits pour la revue de Psychologie des Ressources Humaines doivent tre
soumis la suivante adresse de-mail:
revista@apio.ro.
Pour diter le manuscrit sil vous plat utiliser Times New Roman de 12 points, interligne 1,5
et la mise en page A4. Chaque page sera
numrote dans le coin suprieur droit. Les
marges suprieures et latrales doivent tre laisses au moins 1inch ou 2.54 cm. Un exemple
complet dun manuscrit peut tre trouv dans le
Manuel de Publication APA, 6me dition.
Premire page
La premire page du manuscrit doit comporter les informations suivantes:
1. Titre
Un titre doit tre un nonc concis du sujet
principal et doit identifier les variables ou les
questions thoriques de lenqute et la relation
entre eux. Il doit tre dactylographi en lettres
majuscules et minuscules, centr entre les
marges de la page et positionn dans la moiti
suprieure de la page.
2. Nom de lauteur(s)
et affiliation(s) institutionnelle(s)
Nom de lauteur(s) sera prsent sous la
forme suivante: prnom, initiale(s) des autres
prnoms et le nome.
Laffiliation institutionnelle doit reflter
linstitution / lemplacement o lauteur(s) a t
quand la recherche a t mene. Quand un
auteur na aucune affiliation institutionnelle, la
ville et ltat de rsidence dessous le nom de
lauteur doit tre indiqu. Laffiliation institutionnelle doit tre centre sous le nom de lauteur, sur la ligne suivante.
Publications
Les articles accepts sont copis-dits et
retaps. Les auteurs ont examiner les modifications et relire leurs travaux. Lditeur de la
revue Psychologie des Ressources Humaines va
contacter lauteur correspondant aprs lditeur
assigne son travail un numro de la revue. Si
votre travail est accept, sil vous plat garder
lditeur inform des changements dans votre
information de contact et les longues absences.
136
3. Note de lauteur
Cette section doit inclure les lments suivants:
Premier paragraphe doit inclure laffiliation(s) au dpartement au moment de ltude pour tous les auteurs comme suit: nom
de lauteur tel quil apparat dans la ligne,
une virgule, nom du dpartement, une virgule, le nom de luniversit, point-virgule,
nom de lauteur suivant, et ainsi de suite, et
se termine par un point.
Deuxime paragraphe doit inclure tous les
changements dans les affiliations de lauteur la suite de lpoque de ltude comme
4/17/2012
1:10 PM
Page 137
suit: [nom de lauteur] est maintenant

[affiliation].
Le troisime paragraphe doit inclure des
remerciements (uniquement pour les subventions ou autres aides financires, des
accords spciaux concernant les droites
dauteur, merci pour laide personnel) et
des circonstances particulires (les
divulguer avant que les remerciements dans
ce paragraphe).
Le quatrime paragraphe doit inclure des
informations sur la personne contacter en
prcisant ladresse postale et de-mail.
Placez la note de lauteur sur la page de
titre, en dessous du titre, nome de lauteur et
laffiliation, et centrez ltiquette Note de lauteur. Commencez chaque paragraphe de la
note avec un retrait et ditez des paragraphes
distincts pour les noms des auteurs et les affiliations actuelles, les changements dans les affiliations, les remerciements, et des circonstances
particulires (sil y a unes), avec la personne
contacter. La note de lauteur nest pas
numrote ou cite dans le texte.
interruption, ne commencez pas une nouvelle

page lorsque survient une nouvelle rubrique.
Cette section doit inclure les lments suivants:
Introduction du problme. Cette section
prsente le problme spcifique sous linvestigation et dcrie la stratgie de
recherche. Il nest pas ncessaire dtiqueter cette section Introduction.
Explorez limportance du problme. Cette
section stipule pourquoi le problme mrite
de nouvelles recherches. Explicitez ce
problme en fonction du type de ltude
(tude empirique, revue de la littrature et
une mta-analyse, tude mthodologique et
tude de cas).
Dcrivez la thorie pertinente en discutant
de la littrature pertinente lie et en dmontrant la continuit logique entre les travaux
antrieurs et prsents
Etat chaque hypothse teste clairement et
fournissez un argument thorique pour la
faon dont elle est drive de la thorie ou
est logiquement lie des donnes
antrieures et argumentation.
Page du rsum
Mthode
Le rsum ainsi que le titre vont sur la
page 2.
Le rsum ne doit pas dpasser 150 mots.
Ltiquette Rsum doit apparatre en lettres
majuscules et minuscules, centre, en haut de la
page. Editez le rsum lui-mme comme un seul
paragraphe, sans retrait de paragraphe. Placez
un titre courant (titre abrg).
Le rsum sera crit en Anglais, France et
Roumaine. Il est ncessaire dinclure 3-5 mots
cls aprs chaque rsum, dans ces langues.
Pages principales du texte

Dans la prparation de votre manuscrit, commencez lintroduction la page 3. Editez le titre
du manuscrit en lettres majuscules et minuscules,
centrez-le en haut de la page, puis tapez le texte.
Les autres sections de larticle se succdent sans
Cette section dcrit en dtail comment ltude a t mene en prcisant les dfinitions
conceptuelles et oprationnelles des variables
utilises dans ltude. Les auteurs doivent
inclure les lments suivants:
Caractristiques de lchantillon en
dcrivant ses caractristiques majeures
avec un accent particulier sur les caractristiques qui peuvent avoir une incidence sur
linterprtation des rsultats
La procdure dchantillonnage en
dcrivant les procdures de slection des
participants en termes de mthode dchantillonnage, le pourcentage de lchantillon
approch qui a particip, le nombre de participants qui sa lui-mme slectionn dans
lchantillon
La taille de lchantillon, la puissance et la
prcision
137
4/17/2012
1:10 PM
Page 138
Les mesures et covariables en dcrivant les

mthodes utilises pour recueillir des donnes et amliorer la qualit des mesures
La conception de la recherche
Les manipulations exprimentales ou les
procdures
La description des tches
Rsultats
Cette section rsume les donnes recueillies
et lanalyse effectue sur les donnes pour tester
les hypothses proposes. Signaler lanalyse des
donnes de faon suffisamment dtaille pour
justifier vos conclusions. Pour plus dinformations sil vous plat consulter le Manuel de Publication APA, 6me dition.
Discussion
Cette section value et interprte les implications des rsultats, surtout en ce qui concerne
les hypothses originales. Examinez, interprtez, qualifiez les rsultats et tirez des
infrences et des conclusions. Insistez sur les
consquences thoriques ou pratiques de ces
rsultats. En outre, les limites de ltude et les
tudes futures doivent tre considres dans
cette section.
Rfrences
Les rfrences sont les entres dans la liste
alphabtique la fin de larticle ou note de
recherche. Cette liste doit inclure tous les
travaux qui ont t cits tout au long du
manuscrit. Les rfrences doivent tre formates comme suit:
1. Priodiques (exemples slectives)
Auteur, A.A., Auteur, B.B., & Auteur, C.C.
(anne). Titre de larticle. Titre du priodique,
xx, pp-pp. doi: xx.xxxxxxxxxx
Auteur, A.A., Auteur, B.B., Auteur, C.C.,
Auteur, D.D., Auteur, E.E., Auteur, F.F.,
Auteur, Y.Y. (anne). Titre de larticle. Titre du
priodique, xx, pp-pp. doi: xx.xxxxxxxxxx
138
Auteur, A.A, Auteur, B.B., & Auteur, C.C.

(anne). Titre de larticle. Titre du priodique,
xx, pp-pp.
Auteur, A.A., & Auteur, B.B. (in press). Titre
de larticle. Titre du priodique. Retrieved from
2. Livres
Auteur, A.A. (anne). Titre de luvre.
Lieu: Editeur.
Auteur, A.A. (anne). Titre de luvre.
Retrieved from http://www.xxxxxxx
Auteur, A.A. (anne). Titre de luvre. doi:
xxxxx
Editeur, A.A. (d.) (anne). Titre de luvre. Lieu: Editeur.
3. Pour les chapitres dun livre ou une
inscription dans un livre de rfrence
(exemples slectives)
Auteur, A.A., & Auteur, B.B. (anne). Titre
du chapitre ou de lentre. In A. diteur, B. Editor, & C. Editeur (Eds.), Titre du livre (pp. xxxxxx). Lieu: Editeur.
du chapitre ou de lentre. In A. diteur, B. Editor, & C. Editeur (Eds.), Titre du livre (pp. xxxxxx). Retrieved from http://www.xxxxxxx
du chapitre ou de lentre. In A. diteur, B. Editor, & C. Editeur (Eds.), Titre du livre (pp. xxxxxx). Lieu: Editeur. doi: xxxxxxxx
4. Runion et de symposium
Contributeur, A.A., Contributeur, B.B.,
Contributeur, C.C., & Contributeur, D.D.
(anne, mois). Titre de la contribution. In E.E.
Chairmen (Chair), Titre du symposium. Symposium conducted at the meeting of [nome de lorganization], Lieu.
Prsentateur, A.A. (anne, mois). Titre de la
communication ou un poster. Paper or poster
session presented at the meeting of [nome de
lorganization], Lieu.
5. Les uvres non publies
Auteur, A.A. (anne). Titre du manuscrit.
preparation"].
4/17/2012
1:10 PM
Page 139
Pour une description dtaille de la procdure lie la citation des autres types de travaux
que ceux numrs ci-dessus, consulter le
Manuel de Publication APA, 6me dition.
Notes
Les notes sont utilises pour fournir un contenu supplmentaire ou pour reconnatre le
statut dautorisation de droits dauteur.
Annexes
Les annexes du manuscrit (tiquetes
ANNEXE A, ANNEXE B, etc.) contiennent des
matriaux qui compltent le contenu de larticle
tels que les longues procdures mthodologiques,
les calculs de mesures, les chelles etc.
Citation
Il est important de mettre dans la section
Rfrences tous les travaux qui ont t cits tout
au long du manuscrit. Lauteur peut citer dans le
texte comme suit:
1. Un auteur
Nom dauteur et anne: Il a t constat que
X est associ Y (Auteur, anne)
Anne seulement: Auteur (anne) a constat que .
2. Deux auteurs
Quand un travail a deux auteurs, lauteur
doit citer deux noms chaque fois que se trouve
le renvoi dans le texte.
Quand une uvre a trois, quatre ou cinq
auteurs, vous devriez citer tous les auteurs la
premire fois le renvoi, mais dans les citations
suivantes, ne devriez pas inclure que le nom du
premier auteur suivi de et al. (pas en italique et
avec un point aprs al.) et lanne.
Tableaux et figures
Lauteur doit numroter tous les tables et
les figures en chiffres arabes dans lordre dans
laquelle elles sont cites dans le texte, indpendamment si une discussion plus dtaille du
tableau ou de la figure se produit plus tard dans
le document. Lauteur doit les tiqueter comme
le Tableau 1, Tableau 2, et ainsi de suite ou de la
Figure 1, Figure 2, et ainsi de suite. Liste toutes
les tables en premier, suivi par des figures. Liez
les tableaux et les figures aprs annexes la fin
du manuscrit, et indiquez la position de chacun
dans le texte comme suit:
-----------------------------------Insert Table 1 about here
-----------------------------------Chaque tableau ou figure a besoin dune
phrase dintroduction dans votre texte. Le format accept est le standard (canonique). Chaque
tableau doit dclarer un type danalyse (qui est
identifie dans le titre) et chaque colonne verticale et range horizontale doit contenir quun
seul type de donnes.
3. Deux ou plusieurs travaux cits

Lauteur doit ordonner citations par ordre
alphabtique. Lauteur doit dsigner deux ou
plusieurs uvres dun auteur (ou par un groupe
identique dauteurs) publies dans la mme
anne en ajoutant une lettre a, b et ainsi de
suite, aprs lanne de publication.
4. Travail avec aucun auteur identifi
ou avec un auteur Anonymus
Quand une uvre na pas un auteur identifi, lauteur doit citer dans le texte les premiers
mots de lentre de la liste de rfrence
(gnralement le titre) et lanne.Utilisez des
guillemets doubles autour du titre dun article,
un chapitre ou une page web et en italique le
titre dun priodique, un livre, une brochure ou
un rapport:
sur lengagement organisationnel (Rapport dtude, 2011)
le livre Les rsultats de motivation (2011)
5. Les numros de page dans
les citations
Pour citer une partie spcifique dune
source, lauteur doit indiquer la page, le
chapitre, la figure, le tableau ou lquation
139
4/17/2012
1:10 PM
Page 140
lendroit appropri dans le texte. Toujours donnez les numros de pages pour les citations.
(Johnny, 2011, p. 13)
6. Les sources secondaires
Lorsque luvre originale est puise,
indisponible par des sources habituelles, lauteur doit donner la source secondaire dans la
liste de rfrences et, dans le texte, il doit don-
140
ner le nom de luvre originale et la citation de

la source secondaire:
le rapport de Minnie (as cited in Smith,
2011).
Merci pour lattention porte aux conventions dcrites dans ce guide il aidera le travail
de toutes les personnes impliques dans la publication de ce journal.
4/17/2012
1:10 PM
Page 141
4/17/2012
1:10 PM
Page 142

Organizationala - Grup - 2010 Curseu

Încărcat de

Informații document

Titlu original

Drepturi de autor

Formate disponibile

Partajați acest document

Partajați sau inserați document

Opțiuni de partajare

Vi se pare util acest document?

Este necorespunzător acest conținut?

Drepturi de autor:

Formate disponibile

Organizationala - Grup - 2010 Curseu

Încărcat de

Drepturi de autor:

Formate disponibile

Volumul_10_nr_01.

Psihologia Resurselor Umane

Asociaia de Psihologie Industrial i Organizaional

Asociaia de Psihologie Industrial i Organizaional

Copyright 2012 Asociaia de Psihologie Industrial i Organizaional

Publicarea de articole n Revista Psihologia Resurselor Umane

Reflecii asupra sinergiei cognitive n grupurile sociale restrnse

Practica managerial sugereaz faptul c

Cu alte cuvinte, am ncercat s descopr cum

Department of Organisation Studies, Tilburg University, NL.

Reflecii asupra sinergiei cognitive n grupurile sociale restrnse

Calitatea muncii n echip

baza studiului prezentat n Cureu, Schruijer i

Reflecii asupra sinergiei cognitive n grupurile sociale restrnse

Aa cum se arat n Figura 1, grupurile n

creativitii i complexitii cognitive a acestuia.

Preferinele relaionale reciproce,

Reflecii asupra sinergiei cognitive n grupurile sociale restrnse

Reflecii asupra sinergiei cognitive n grupurile sociale restrnse

proce sunt antecedente ale sinergei cognitive de

Reflecii asupra sinergiei cognitive n grupurile sociale restrnse

Figura 3. Comparaia dintre grupurile cu i fr intervenii normative (bazat pe datele raportate n

ambele scoruri sunt negative, dup cum rezult

Reflecii asupra sinergiei cognitive n grupurile sociale restrnse

tiv n cadrul grupului. Mai mult dect att, am

Cureu, P. L. & Schruijer, S. G. L. & Boro, S. (in

Reflections on group cognitive synergy

Common managerial wisdom suggests that

marize some of my research attempts to find out

Group cognitive synergy defined

Department of Organisation Studies, Tilburg University, NL.

Reflections on group cognitive synergy

individual cognitive structures (Cureu, 2006a;

bers as they were part of groups with no synergy

Reflections on group cognitive synergy

2010). Similarly we show in another study

Reflections on group cognitive synergy

effective groups has no unequivocal or definite

Reflections on group cognitive synergy

opinion as natural and helpful, (6) view initial

reported in the original paper). Weak cognitive

Reflections on group cognitive synergy

Cureu, P. L. (2006a). Emergent states in virtual teams:

Reflections on group cognitive synergy

Karau, S. J., & Williams, K. D. (1993). Social loafing:

Larson, J. R. Jr. (2010). In search of synergy in small

The relationship between emotional labor and burnout

College of Communication and Public Relations, NSPSAS.

outcomes, among which the most frequently

rules: surface acting (regulating the emotional

scale development (1998), viewed emotional

(bill collectors or law enforcement officers, for

and those experienced in private life revealed that

burnout. In this respect, Bakker and Heuven

in Wharton & Ericksons (1993) terms). In a

stress at work is due to threats with resource

HI.II. Surface acting will be positively

As the main interest in the current study

Demerouti, Verbeke and Bakker (2005)

was related positively to personal accomplishment.

with job performance (Wright & Bonett, 1997;

burnout (Green et al., 1991; Lee & Ashforth,

to actually experience the emotions that you