Documente Academic
Documente Profesional
Documente Cultură
OF THE
ALFRED
P.
An Empirical Investigation of the Motivational Determinants of Task Perf or^iance Interactive Effects between Ins~rumentality-Valence| and Motivation-Ability by
J. R, Galbraith" and L.
L.
Cu-mings>'"
261-67
An Empirical Investigation of the Motivational Determinar.ts of Task Perf orir.ance Interactive Effects between Instrumentality-Valence and Motivation-Ability
by
J. R. Galbraith" and L. L. Cuxmings*"
261-67
May 26
1967
Abstract
f(M'A)],
be postulated to vary with the valences (V) of outcomes associated with the
of performance for
as well as M and A,
A modified analysis
J UN 9.a 1967
"All the studies of human relations and supervision tell little about
50.)
195^5. p
6.)
2 -
outcome 1
alternative 1
role
alternative
^^
outcome 2
Figure 1
why the role occupant will choose one of the alternatives and reject xhe
other.
Approaches to
-:'ne
upon the assumption that behavior is directed toward pleasure and away from
pain.
put
The history of inquiry into this problem has been one of trying to
;o~.e
3 -
after the fact, by noting the probable sources of pleasure and pain.
which outcomes are pleasurable and which are painful and links the
concepts with empirically observable events.
The Vroom model attempts to predict which outcomes are pleasurable and
The
alternative 1
role
jk
alternative
First Level Outcome
Figure
Choice Situation Depicting First and Second Level Outcomes Which Result from Alternative Acts
u -
are viewed as events to which the first level outcomes are expected to lead.
In other words the first level outcomes are viewed as means to the second
level outcomes.
Preferences
mentality.
which the first level outcome acquires a valence is stated by the proposition:
The valence of an outcome to a person is a monotonically increasing
function of the algebraic sum of the products of the valences of all other
outcomes and his conceptions of its instrumentality for the attainment of
these other outcomes (Vroom, 196U, p. 17).
n
V.
J
Thus,
f.(j:
1
V,
'^
.1,
1.
2,
.,,n
(1)
L=l
i^'
6f_.
6f.
- 5 -
Valence of outcome
V
k
Valence of outcome k
.,
The use of equation (1) for explaining choices among alternative levels
All of the
Some employees
predicted a priori which outcomes are positively valent and which are
negatively valent.
K-1 6f
6v
o
"
choice approach stems from the fact that all voluntary behavior is the result of a conscious or unconscious selection of particular actions out of
all those which are physically possible.
or deliberate process.
follows one particular course of action, there are other courses of action
that he foregoes.
may be undertaken.
The Vroom model provides a vehicle for explaining the choice situation
Since organiza-
tions vary in the activities determined by formal authority, the Vroom model
choice among alternative levels of task related effort and is able to act
out his choice.
The range is
bounded on the high side by the abilities of the worker and the constraints
of the technology.
- 8
y to r
h.as
operative worker.
Manufacturing
firms with tangible goals are thus able to measure an individual's contribu-
are measured and that is in terms of their relative attractiveness for the
worker.
Alternatively,
The first
The second
level outcomes are rewards and penalties which are given on the basis of
performance.
The variations in valences, with which the variations in productivity are hypothesized to be correlated, are to result from differences in the
8/4-
can be conceived as rewards and penalties deriving from those groups with
work, can affect the goals desired by other? thereby creating a situation
of interdependence.
If this scheme is useful, the valence of performance
penalties.
those who would have an interest in his performance and ability to supply
They are
Thus, for
On this basis, a
Incorporating themiRole Variablesm i > miw the Model into n mi ! i !! waiii II^wCMi iiiiiipi^i iiTBM mil w
I
III
II
II
ai
ii
ii
Each of the variables can be cast into the model presented earlier.
For each group a reward can be specified which carries a valence and will
with the product of the valence of the reward and the perceived instrumen-
Variations in
For example, the instrumentality of money will vary from one to zero for
a wage incentive system versus a union negotiated contract.
Of course it
instrumentality.
10 -
the ability of the supervisor to exercise influence over the workers depends
subordinate's goals have been variously described as supportive, employeecentered, and considerate.
This is to communicate
directly with the product of the valence of supportive behavior and the
It differs
Previous
- 11 -
While there
have been many research studies concerning the work group, our interest is only in the influence of the work group on the worker's choice of levels of
performance.
dependence between the worker and the people with whom he works.
group in the individual's performance and the individual's preference for group acceptance.
perception of the group's interest in his performance or perceived instrumentality and the valence of group acceptance.
Group acceptance can be
directly with the product of the valence of group acceptance and the
The proposition can be interpreted, as have the previous ones, in terms of the multiplicative interaction between valence and instrumentality.
Performance will be high when the individual desires group acceptance and
believes that high performance is a behavior necessary for acceptance.
Performance will be low when the worker desires acceptance and believes
that low performance is a behavior necessary for acceptance.
Similar
Performance
performance
Thus
model as follows:
V=VI pm m
where
V
+V^I
^
pf
+VI
P pp
+VI ps +VI s
g pg
valence of money valence of fringe benefits valence of promotion valence of supportive behavior valence of group acceptance
m
V
f
=
V p V
s
=
=
V
g
pm
perceived instrumentality of performance for money perceived instrumentality of performance for fringe benefits
'
perceived instrumentality of performance for promotion perceived instrumentality of performance for supportiveness perceived instrumentality of performance for group acceptance
PP
ps pg
means to other ends but quite possibly can be perceived as an end itself.
The rewards which result as a natural consequence of the behavior were
13
individual.
performance, this performance will vary directly with the amount of egoinvolvement in the task.
The complete model can now be presented.
VrV+VI pm +V^I^+VI PP e m f pf P
where:
V
e
=
+V^I ps s
+ VI^ g PR
(5)
determining performance.
(Lawler, 1966),
P = f(MA).
(5)
Equations (5) and (6) make up the model for which an empirical test
will now be illustrated.
The presentation will proceed from sample
lU -
Research pesign
Indiana city.
objective measures of production and held jobs where they controlled the
pace of work.
In other words, the sample was selected so as to provide
One
The
The implicaLet
15 -
that choice may be different from preference when expectancy does not equal
one.
While the worker does not depend on others in his task performance
(after the machine is set up and materials are available), there are other
An
aries, the worker notifies the foreman who authorizes the reporting of the
non-standard condition.
16
attainment of that outcome, he must have control over his pace of work. Under this condition the worker's choice of the amount of effort he will
put forth will determine his output and therefore the performance measure.
It must also be pointed out that work cannot be classified into the simple
machine-paced.
The
subjects were also selected from those departments where the man
On the
basis of the above discussion, the assumptions seem reasonable for the
Measurements
17 -
ments.
The
variation to be explained by the motivational variables will be the variation about the production standards.
The approach used was to use
one month's performance since data in this form is available from the
company.
other variables.
following method.
instrumentality:
performance?
18
I feel that the more I produce, the better are my chances for a promotion,
I feel that my production is important but management looks at other things too*
performance-promotion
don't think my chances for a promotion -r^re affected one way or another by the amount of my production,
I I feel that large amounts of production could hurt my chances for a promotion, I feel that large amounts of production would definitely hurt my chances for a
j^romotion.
The graphic rating technique used above, and also to be used for valence, has been shown to be effective in developing criteria for defense system
The outcomes
After
completing the ranking, the subjects were asked to indicate the relative
degree of liking or disliking by drawing lines from the ranked outcomes to
the rating scale.
Listed below are several things which you could receive in connection
You could be popular and accepted by the men with whom you work,
immediate supervisor.
19 -
d)
e)
pensions, etc.
f)
1,
Rank the above items in the order which you would prefer them.
Rank
Draw lines from the word to the scale indicating how much you prefer
these outcomes.
10
extremely desirable
very desirable
1.
2,
desirable
3.
receive it or not
undesirable
5,
very undesirable
6.
extremely undesirable
The
study of job
20
If a problem comes up in your work and it isn't all settled by the time
you go home, how likely is it that you will find yourself thinking about
it after work?
I
vs^.
reliability measurement.
A total of eight usable questionnaires resulted for valences and nine for
instrumentality.
Likewise
21 -
Valence
r r
.50 (n = U8)
.80 (n = U5)
Instrumentality
Ego- Involvement
r=
,81
(n-
8)
Statistical Jests
The task remaining is to devise the necessary statistical tests for the above hypotheses and also to assess their relative importance.
The
variable individually.
For ability, those subjects holding their present job for greater
22 -
To
high conditions.
The computation needed to generate the test statistics now involves matrix
Performance
'
(7)
where:
X,
23 -
X
2
when instrumentality is
"!,-,
The last term (X.X^) picks up the hypothesized interaction between valence
and instrumentality.
relationship.
If
b^^,
b., or b
and b
Let
(8)
where:
X
o
Now if valence and instrumentality interact but motivation and ability exert
h^^
2^ -
car.
be extended to simultaneously
There will be
ability there are thirty-eight -- seven for each extrinsic reward and
three for ego-involvement.
Due to the small sample size (n = 32) and the attempts to select the
This is an excellent
We now have a
has an F-ratio which has fallen to less than 1,0 it is removed and the
25
process continues.
of the fact that one always gets perfect correlation in a two variable
Results
When
the linear model given by Equation (7) was run for each variable, none of the valences alone, instrumentalities alone or the two in combination
and instrumentality (b
ability (b_).
26 -
Table
Significant Variables from Regression Considering Variables Alone with Ability Both Controlled and Not Controlled All Others Are Not Significant
p <
.OS
Supportiveness b
p <
.01
variation at the research site, or that in fact they are not related.
are possible but cannot be determined in this research.
All
Therefore we will
and 3.
Table
reports results
Table
Coeffi cient
b
*
-1.57
,20.
.57
27 -
The
Since it is
coefficient was 0,57 meaning about one-third of the variation was explained
by the significant variables in the model.
Table
Table
Variable
Supportive Behavior Interaction (high valence, high instrumentality, high ability)
Money Interaction (high valence, high instrumentality, high ability)
value
3,70
Significance Level
,001
3,01
,01
-2.0U
.10
,68
28 -
constituted six months and the sample was split at the median.
While
there are certainly other ability dimensions on which the subjects may
vary, time on the job captures some of the substance of the ability
variable.
For supportive behavior and money, the variable represents the three-way
For
A
Discussion
Lawler, 1966),
29
The present study also offers support for the Vroom (1964) model of the
can be made.
within the company is largely contingent upon seniority, the low instru-
fringe levels.
power of the union versus the company -- not from the level of performance
exhibited by any given operative.
is that the independent variable
30 -
interaction.
operative performance over the time span of measurement used in the present
study. The present study suggests that at least three conditions are
effectiveness of the reward system must be such that the organization can
vary the magnitude of the reward component sufficiently to evoke variations
- 31
in employee behavior.
32 -
REFERENCES
1,
Ando A., and Kaufman, G. M, "Evaluation of an Ad Hoc Procedure for Estimating Parameters of Some Linear Models", Sloan School (To of Management, M.I.T., Working Paper No, 9U-6U, 196U. tics appear in The Review of Economics and Statis
.
2,
Atkinson, J. W,, "The Achievement Motive and Recall of Interrupted and Completed Tasks", Journal of Ex perimental Psychology , 68 (1953), pp. 381-390.
Blau. P. M,, and Scott, W. R., Formal Organizations, San Francisco: Chandler, 1962.
3,
U,
Dubin, R., "Supervision and Productivity: Empirical Findings and Theoretical Considerations", in Dubin, R., Homans, G. C, Mann, F, C, and Miller, D. C, Leadershio an d Productivity , San Francisco: Chandler, 1965.
5,
Eckenrode, R. T., "Weighing Multiple Criteria", Managemen t Science 12, 1965, pp. 180-192.
French, Elizabeth G., "Effects of Interaction of Achievement Motivation and Intelligence on Problem Solving Success", ^'^* 1957, pp. 399-UOO. ^^f5."JJ!gZ5JlJ-- iA?I.*
6,
7,
Fleishman, E. A., Harris, E, F., and Burtt, H. E., LeadershiD and S_upervisi on in In dustry , Columbus: Ohio State University, Bureau of Educational Research, 1955, Fleishman, E. A,, "A Relationship Between Incentive Motivation and Ability Level in Psychomotor Performance", Journal ^_^?^ ^ f]'.^,j".Q,?.^.^,f P sychology , 56, 1958, pp, 78-81,
^
,
8,
9,
Georgopoulos , B, S., Mahoney, G, M,, and Jones, N. W., "A Path-Goal Approach to Productivity", Jo ur na^l of _^pj-/^j,./,g^_'=,^,f,,?,y '*^ 1957, pp. 3U5-353.
Goldberger, A. S., Econometric^^^Theoi^, Wiley, 196U,
Halpin, A. W., and Winer, B. J., "A Factorial Study of the Leader Behavior Description Questionnaire, in Stogdill, R. M. , and Coons, A. E. (Editors), Leader__^Behav^i orj^ ^"^_g.?5r.f.E."li,Il 5"fl ass Measurement, Columbus: 0^'o~St1ate~Tj1niveVsTty^'feurVau Mangr. No 88, 1957, pp. 52-6U, ResearcKT^es.
10.
11,
33 -
12,
Johnston, J,, Econometric Methods. New York: McGraw-Hill, 1963, pp. 221-228.
Lawler, E. E., "Ability as a Moderator of the Relationship Between Job Attitudes and Job Performance", Personnel Psychology, 19, 1966, pp. 153-16U,
Lewin, K,, "The Conceptual Representation and the Measurement of Psychological Forces", Con_tr .__ Ps^ychol Theo ry , Durham, North Carolina; Duke University Press, 1, No, u, 1938,
13,
14,
15,
Slater, C. J. W,, "Some Factors Associated with Internalization of Motivation Towards Occupational Role Performance", Unpublished Doctoral Dissertation, Ann Arbor: University of Michigan, 1959,
16,
Vroom, V, H,, "Ego-Involvement, Job Satisfaction, and Job Performance", Personnel Psychology , 15, 1962, pp. 159-177.
Vroom, V, H,, ^ork and Motivation, New York: Wiley, 1964,
17,
18,
J'^^'^^^^?^^^'^ t
3U
FOOTNOTES
statistical section.
However, the
BASEMEN r^
Date Due
'ft?l
^_____^
\z;A
NOV
TO
7
la
!<}
Lib-26-67
.,..,.M.m|,|><i>jnij;'M
3""=i'Dfib'QQ3
MIT LIBRARIES
67 D
3M'=i
^5f'(^r
3
2'y^-'-o7
3
TOAD DO 3
fl?D
31
TD6D 003
TOl.
441
'^'5Z~Gy
I5^'(c7
ll!
lililllll
TDflD D
fl?
3b4
25^ -^"^^
TOAD
fl7D
353
MIT LIBRARIES
t!'|iT'||'!|l|l'M|'!!
{"I
Toa D
1^0-^^
2fc('&7
illllll
'IDfl
003
fl70
Sflfl
MiT LIBRARIES
1^2- (-'7
3
TOflO
003 67
bE
Z6Vt7
3
=1060
003
01 SbS