Journal of Vocational Behavior 71 (2007) 247264 www.elsevier.

com/locate/jvb

Employability during unemployment: Adaptability, career identity and human and social capital
Sarah McArdle a, Lea Waters a,*, Jon P. Briscoe b, Douglas T. (Tim) Hall c
a

Department of Management, Faculty of Economics and Commerce, The University of Melbourne, Parkville, Vic. 3101, Australia b Department of Management, College of Business, Northern Illinois University, Barsema Hall, 245 H, DeKalb, IL 60115-2897, USA School of Management, Boston University, 595 Commonwealth Avenue, Boston, MA 02215, USA Received 17 November 2006 Available online 10 July 2007

Abstract Recently, Fugate et al. [Fugate, M., Kinicki, A. J., & Ashforth, B. E. (2004). Employability: A psycho-social construct, its dimensions, and applications. Journal of Vocational Behavior, 65(1), 14] dened employability as a psycho-social construct comprised of three dimensions: (i) adaptability; (ii) career identity; and (iii) human and social capital. The aim of the current paper was to empirically test Fugate et al.s model in a sample of 416 unemployed Australians (n = 126 for longitudinal sample). Specically, this research explored employability in relation to three aspects of unemployment: (1) self-esteem during unemployment; (2) job search during unemployment; and (3) re-employment (at a 6-month follow-up). Overall, the results of this longitudinal study provide broad support for the psycho-social construct of employability and demonstrate its applicability to the unemployment context. 2007 Elsevier Inc. All rights reserved.
Keywords: Employability; Unemployment; Job loss; Career; Employment; Adaptability; Career identity; Job search; Re-employment

Corresponding author. Fax: +61 3 9439 42393. E-mail address: l.waters@unimelb.edu.au (L. Waters).

0001-8791/$ - see front matter 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.jvb.2007.06.003

248

S. McArdle et al. / Journal of Vocational Behavior 71 (2007) 247264

1. Introduction Careers once characterised by stability, vertical progression and job security, are now likely to involve lateral movements across organisations, increased instability, and periods of unemployment for most workers (Brown, Hesketh, & Williams, 2003; Eby, Butts, & Lockwood, 2003; Rousseau, 1997). In this insecure environment, the concept of employability has emerged as a key contributor to career success (Bagshaw, 1997; Fugate, Kinicki, & Ashforth, 2004; Garavan, 1999). Traditionally, employability has been broadly dened as the ability to gain and maintain employment, both within and across organisations (Finn, 2000). In discussing the concept, many authors use uni-dimensional, outcomebased denitions, labelling an individual as employable based on their ability to gain employment (e.g. Finn, 2000; Ritchie, 2000; Van der Heijden, 2002). Such denitions are tautological because people who are employed are dened employable. As a consequence, the investigation of employability in unemployed populations has been limited. Moreover, as such denitions are based on the objective outcome of being employed they provide little indication of the personal factors that contribute to ones employability. For these reasons, the current paper adopts a more comprehensive model of employability. Fugate et al. (2004) present employability as a person-centred, psycho-social construct, decoupled from ones employment status. This means that one can be employable without necessarily being in employment. In contrast to the outcome-based denitions presented above, the psycho-social approach showcases the role of the individual in determining, in part, their own employability. Given its applicability to individuals both in and out of employment, we contend that the psycho-social model is particularly useful in helping individuals to cope with unemployment as it can be used to assist them to recognise that their employability can be self-improved, despite the temporary absence of paid employment. 2. The psycho-social model of employability According to Fugate et al., employability comprises three separate, yet inter-related, dimensions: (i) adaptability; (ii) career identity; and (iii) human and social capital. Adaptability refers to the willingness and ability to change behaviours, feelings and thoughts in response to environmental demands (Fugate et al., 2004). According to Hall (1996, 2004) adaptability is essential to success in the current era of career insecurity. Savickas (1997) has linked adaptability to planfulness, a readiness to cope and a willingness to explore oneself and ones environment. Adaptable individuals have a high tolerance for uncertainty and ambiguity and are comfortable in novel situations and across organisational boundaries (OConnell (in Press)). Encompassed in the dimension of adaptability is the construct of proactive personality. According to Seibert, Crant, and Kraimer (1999) and Bateman and Crant (1993) proactive individuals are able to aect environmental change, remaining relatively unimpeded by situational constraints. Proactive personality has been linked to identifying and acting on opportunities, feelings of control, perseverance, self-ecacy, self-direction, coping, and information-seeking (Bateman & Crant, 1993; Crant, 2000; Seibert et al., 1999; Seibert, Kraimer, & Crant, 2001; Thompson, 2005). Job loss research has shown that factors such as an internal locus of control, self-ecacy and problem-focussed coping play important roles in gaining re-employment.

S. McArdle et al. / Journal of Vocational Behavior 71 (2007) 247264

249

The second employability dimension, career identity, represents the way individuals dene themselves in the career context, and can be conceptualised as a cognitive compass used to navigate career opportunities (Fugate et al., 2004). Career identity reects the knowing-why competencies identied by Dellippi and Arthur (1994). Knowing-why competencies encompass attributes such as career motivation, personal meaning and individual values. Given that career trajectories are less externally dened in the new career environment (McGreevy, 2003), the use of an internal career compass has become critically important in providing direction, especially when the individual nds him/herself outside of the boundaries of an employing organisation. Hall, Briscoe, and Kram (1997) suggest that, in todays turbulent career environment, identity needs to be decoupled from a specic job or organisation, instead representing an individuals personal values, motivations and broader career interests. In support of this, Anakwe, Hall, and Schor (2000) found that teaching MBA students to reect on their career-related values, needs and motivations (assisting the formation or consolidation of career identity) facilitated eective goal setting and decision making. In periods of career transition, such as unemployment, the ability to harness ones career identity as a guide when establishing goals and making decisions may be crucial in identifying career opportunities. Human capital refers to the personal variables that may aect ones career advancement, including education, work experience, training, skills, and knowledge. Human capital encompasses the knowing-how competencies outlined by Dellippi and Arthur (1994). Knowing-how competencies refer to career-related knowledge and skills built via occupational learning and professional development activities. By investing in continuous learning, individuals can develop their human capital, thereby building employability. Social capital reects the interpersonal aspect of employability, incorporating Dellippi and Arthurs (1994) knowing-whom competencies concerning formal and informal career-related networks. Empirical work has demonstrated that interpersonal connections (who one knows) are crucial in shaping individuals self-perceptions (Eby et al., 2003), and providing access to career-related information and resources (Seibert, Kraimer, & Liden, 2001). Social networks can also be a source of social support, serving to ameliorate the destructive consequences of stressful events, such as unemployment (Lazarus & Folkman, 1984; McKee-Ryan, Song, Wanberg, & Kinicki, 2005). 3. Employability and unemployment Fugate et al. (2004) suggest that employability is particularly applicable to the unemployment context. Specically, they propose that employable individuals are (a) less likely to be psychologically harmed by job loss, (b) more likely to engage in greater job search, and (c) more likely to gain high quality re-employment. However, to date, these assertions have not been empirically tested. The current studies address this by empirically investigating: (i) the relationship between employability, self-esteem and job search during unemployment (study 1); and (ii) the relationship between employability, self-esteem, job search and re-employment (study 2). 3.1. Employability and self-esteem during unemployment Research has consistently shown that self-esteem is a key psychological factor that is negatively aected by unemployment (see, e.g. Hanisch, 1999; Jahoda, 1982; Wineeld,

250

S. McArdle et al. / Journal of Vocational Behavior 71 (2007) 247264

Tiggeman, & Wineeld, 1992; Winegardner, Simonetti, & Nykodym, 1984). However, this paper proposes that high employability may act as a buer and prevent people from suering from low self-esteem during unemployment. By being proactive and adaptable and maintaining a strong career identity, employable individuals may be more likely to see the positives that can come from unemployment and, as a consequence, may use job loss as a time for critical reection on career identity and direction (Eby & Buch, 1995; Fugate et al., 2004; Hall et al., 1997; Latack & Dozier, 1986). Using unemployment as an opportunity to aect change and strengthen ones career identity may mean that people high on employability maintain a positive self-esteem despite being unemployed. Career identity may be particularly important in determining ones self-esteem during unemployment. According to Eden and Aviram (1993) unemployment can be a signicant blow to an individuals self-concept. This is likely to be especially so if the individual has invested much of their identity in a particular job or organisation. As self-esteem is determined by an individuals evaluation of the salient aspects of their self-concept (Harter, 1990; Rosenberg, 1965; Shaer, 2002), individuals who have dened themselves in relation to a specic job or rm may be more likely to suer decreases in self-esteem. However, the self-esteem of those individuals who are high on employability, and thus have a strong career identity, dening themselves more broadly in terms of their career (rather than a specic job or employer), may not suer to such an extent. This is because an identity that is centred on career may be more easily decoupled from a particular rm or position. By dening themselves independently of organisations or jobs, employable individuals can protect their self-concept and self-esteem from the impact of unemployment and use their career identity as a guide in times of uncertainty (Hall et al., 1997). Social and human capital may also serve to buer the self-esteem of employable individuals from the impact of unemployment. As Lazarus and Folkman (1984) argue, social capital (and social support in particular) can help individuals cope with stressful situations. According to McIntosh (1991) individuals who have strong social support are more likely to feel valued and, as a result, have higher self-esteem. McKee-Ryan et al. (2005) suggest that human capital variables can also contribute to ones wellbeing during unemployment. For example, highly educated individuals may have higher re-employment expectations, limiting their anxiety during unemployment (Price & Fang, 2002). The cultural capital associated with ones education status may also reinforce the self-esteem of unemployed individuals. Taking the above ndings into account, we predict that the combined aect of career identity, adaptability, human, and social capital in concert is likely to help a person retain a positive self-esteem despite the absence of employment. Accordingly, the rst hypothesis addressed in this research is as follows: Hypothesis 1. Employability (adaptability, career identity, human, and social capital) relates positively to self-esteem during unemployment. 3.2. Employability and job search This paper further contends that employability will associate positively with job search. Past research has shown that adaptability and proactive personality are associated with

S. McArdle et al. / Journal of Vocational Behavior 71 (2007) 247264

251

active engagement in the labour market, feelings of control, optimism, information-seeking, initiative, and self-ecacy, which are likely to encourage the individual to undertake job search (Crant, 2000; Fugate et al., 2004; Savickas, 1997). Furthermore, OConnell (in Press) suggest that adaptability may be linked to a persons sense of condence and marketability, which may further encourage individuals to pursue re-employment via job search. Moreover, career identity may provide the individual with direction during unemployment, preventing them from becoming overwhelmed with the myriad ways to conduct job search and allowing them to take a more directed and problem-focused approach. The self-awareness and motivation associated with career identity can also help structure and focus job search activity. Additionally, social capital provides individuals with access to career-related networks, informational resources and social support during the job search process (Fugate et al., 2004; Seibert et al., 2001). According to Kanfer, Wanberg, and Kantrowitz (2001) social supporters can inuence individuals to see job search as a worthwhile pursuit, as well as providing advice and practical assistance throughout the process. Furthermore, social networks can be an important source of encouragement and reassurance when the individual is rejected (Vinokur & Caplan, 1987). Taking the above into account, the second hypothesis is as follows: Hypothesis 2. Employability (adaptability, career identity, human and social capital) relates positively to job search during unemployment. Self-esteem has also been linked to job search. Kanfer et al. (2001) found that higher self-esteem was linked to increased job search intensity in an unemployed sample. Furthermore, Shamir (1986) found that individuals with high self-esteem were more likely to show initiative and assertiveness during unemployment, obtaining greater information from their environment. Interestingly, however, low self-esteem may also motivate job search by encouraging individuals to pursue re-employment in an attempt to alleviate feelings of low self-esteem or psychological distress (Crossley & Stanton, 2005). Nonetheless, in accordance with Kanfer et al. (2001) this paper takes the former perspective. Hence, assuming Hypothesis 2 holds, employability may also be indirectly related to job search, via self-esteem: Hypothesis 3. Employability relates indirectly and positively to job search during unemployment, via its positive eect on self-esteem. 3.3. Employability and re-employment Employable individuals may also be more successful than their less employable counterparts at achieving re-employment (Kanfer et al., 2001). Certainly, research by Eby et al. (2003), Prussia, Fugate, and Kinicki (2001), and Wanberg, Kammeyer-Mueller, and Shi (2001) indicates that human capital, networking (social capital) and job search intensity are all positively associated with re-employment. Furthermore, proactive personality, optimism, openness, self-ecacy, willingness to explore, and tolerance for uncertainty (as encompassed by adaptability; Crant, 2000; Fugate et al., 2004; Savickas, 1997) and the direction provided by career identity may also prevent feelings of paralysis during times of turbulence, encouraging employable individuals to actively pursue reemployment. Consequently, it is hypothesised that employability may enhance the ability of employable individuals to regain employment:

252

S. McArdle et al. / Journal of Vocational Behavior 71 (2007) 247264

Hypothesis 4. Employability (adaptability, career identity, human, and social capital) relates positively to obtaining re-employment. According to Kasl (1982) self-esteem may also play a role in re-employment. Kasls reverse-causation hypothesis proposes that positive psychological health (in general) and high self-esteem (in particular) facilitates re-employment. As mentioned earlier, unemployment is typically associated with low self-esteem; however, Kasl proposes that having low self-esteem may be detrimental to the unemployed, by inhibiting their ability to gain re-employment. Kasls reverse-causation hypothesis has been supported by authors such as Wineeld and Tiggeman (1985) and Vinokur and Schul (1997). However, other researchers (see for instance Waters & Moore, 2002a) have found no link between selfesteem and re-employment. To investigate this relationship further, the current paper adopts Kasls reverse-causation hypothesis, proposing a positive relationship between self-esteem and re-employment: Hypothesis 5. Self-esteem will be positively related to obtaining re-employment. Wanberg (1997), Wanberg, Kanfer, and Rotundo (1999), Wanberg, Kammeyer-Mueller, and Shi (2001), and Kanfer et al. (2001) have also found a positive relationship between job search and re-employment. This is intuitively appealing, as one would expect that individuals who spend more time looking for a job would be more likely to nd one. Individuals who are engaged in job search are likely to have a strong re-employment goal and commitment to employment guiding their behaviour (Wanberg et al., 1999). It is likely that this focus and direction will predispose job searchers to pursue re-employment specifically, rather than considering other options such as further education. As such, one would expect that those engaged in high levels of job search would be more likely to obtain re-employment. Accordingly, the nal hypothesis is as follows: Hypothesis 6. Job search intensity will be positively related to obtaining re-employment. The six hypotheses outlined above are visually represented in Fig. 1:

Adaptability H1

Self-esteem H5 H3

Career Identity

Employability H4 H2

Re-employment

H6

Human and Social Capital

Job Search

Fig. 1. An illustration of the hypothesised relationships between employability, self-esteem, job search and reemployment during unemployment (H, hypothesis).

S. McArdle et al. / Journal of Vocational Behavior 71 (2007) 247264

253

4. Methods and results The current research was conducted as two studies following the baseline sample over a 6-month period using a longitudinal, survey design. 4.1. Study 1: baseline sample 4.1.1. Sample The study 1 sample consisted of 416 unemployed people (63% male and 37% female) with a mean age of 33.63 years (sd = 11.17). Respondents reported a variety of education levels: 43.1% had nished high school, 23.5% had completed university, 14.3% had obtained a TAFE (Training and Further Education) qualication, 11.8% had completed part of a university course, and 7.3% had completed an apprenticeship. 4.1.2. Measures Adaptability, career identity, human capital, social capital, self-esteem, and job search were all measured in study 1. With the exception of education, all items were assessed using a 5-point Likert scale from (1) strongly disagree to (5) strongly agree. In all scales, a higher score indicated a higher presence of the construct. Proactive personality and boundaryless mindset were used as proxy measures for adaptability. Proactive personality was measured using Bateman and Crants (1993) 10item scale. This scale includes items such as: I excel at identifying opportunities. The reliability for this scale in the current study was a = 0.87. The boundaryless mindset scale (Briscoe, Hall, & DeMuth, 2006) assesses ones attitude towards working across organisational boundaries and includes items such as: I enjoy jobs that require me to interact with people in many dierent organisations and I seek job assignments that allow me to learn something new. The reliability for this scale was a = 0.86. Stumpf, Colarelli, and Hartmans (1983) Identity Awareness scale and Kossek, Roberts, and Demarrs (1998) career self-ecacy scale were used to assess career identity. The Identity Awareness scale includes eight items, such as: I have reected on how my past career and activities match my future career. The reliability of this scale was a = 0.82. The Career self-ecacy scale consists of four items, including: I know the qualities needed for my career path. The reliability of this scale was a = 0.80. Education was used to assess human capital. Additionally, age and gender were used as control variables in the study. Participants provided their age in years and were asked to indicate their gender (M/F) and highest level of education. Gender and education were coded dichotomously as follows: males were coded as 1, and females as 0; those who had engaged in further education following high school were coded as 1 and those who had not gone beyond high school were coded as 0. Age was treated as a continuous variable. Social capital was measured via two networking items: Over the last 6 months I have attended various networking meetings and workshops to learn about new career options and Over the last 6 months I have initiated conversations with knowledgeable individuals in new career areas of interest to me; and one social support item: Over the last 6 months I have been supported by my family and friends. Self-esteem was measured using three items from Messer and Harters (1986) selfesteem scale. Sample items include: I am very happy being the way that I am. Messer

254

S. McArdle et al. / Journal of Vocational Behavior 71 (2007) 247264

and Harter reported strong internal reliability for the scale (a = 0.92) and this scale has also been used validly in unemployment research by Waters and Moore (2002b) and Waters and Muller (2003). In the current study, the reliability of the scale was a = 0.78. Kinicki and Latacks (1990) job search behaviour scale was used to measure job search intensity. This scale consists of ve items, such as: Since being unemployed I have focused my time and energy on job search activities. This scale was found to have a reliability of a = 0.84. Kinicki and Latacks (1990) scale has been used successfully in the job loss literature by researchers such as Lai and Chan (2002) and Wanberg (1997). 4.1.3. Procedure Participants were accessed via the JobNetwork system. JobNetwork is an Australian Government initiative which provides unemployed individuals with access to a number of private and community organisations that can assist in the job search process (JobNetwork, 2005). One thousand surveys were placed at the front counter of JobNetwork oces by the researchers, along with a plain language statement (PLS), and pre-paid returnaddressed envelope. Surveys were left in the oces for a 1-month period. Of the 1000 surveys distributed, 752 were taken by JobNetwork clients. Four hundred and fty-ve surveys were returned to the researcher (of which 416 contained sucient data for analysis), resulting in an overall response rate of approximately 60%. On receipt of the initial survey, participants were given a study code and their contact details were recorded. On completion and return of the survey, respondents were paid $10. 4.1.4. Results Prior to hypothesis testing, missing values for proactive personality, boundaryless mindset, career self-ecacy, proactive personality, networking, social support, job search, and self-esteem were replaced using Estimation Maximisation (EM). In EM information from all the variables of interest is used to predict the missing items. No more than 19 values (4%) were missing on any one item. Missing values on the education measure, age and sex were not replaced, as they were unable to be estimated by scores on other items. This left a total of 39 cases (approximately 8%) with missing values. These cases were removed for data analysis purposes, resulting in the nal sample size of 416. Table 1 presents descriptive statistics and correlations for all variables assessed in the study. As Table 1 shows, age was not signicantly correlated with any other variable whilst gender was signicantly correlated with identity awareness and job search. Education was negatively correlated with social support. Proactive personality, boundaryless mindset, career self-ecacy, identity awareness, social support, job search, and networking were all positively correlated with each other. Self-esteem was positively correlated with all the preceding variables, except for job search. Re-employment was positively correlated with career self-ecacy, boundaryless mindset and proactive personality. Structural Equation Modelling (SEM) was used to test hypotheses in study 1. Prior to entering the data, scale items for proactive personality, boundaryless mindset, career selfecacy, identity awareness, and job search were grouped into three parcels. By parcelling scale items this way model t is facilitated by reducing the number of parameters (Landis, Beal, & Tesluk, 2000). As there were only three self-esteem items and two networking items, these items were not parcelled. Factor analysis was used to parcel items appropriately, by creating a 3 factor structure for each scale.

S. McArdle et al. / Journal of Vocational Behavior 71 (2007) 247264

Table 1 Means, standard deviations, and correlations of the study variables Variable 1. Age 2. Education 0 = up to high school 1 = past high school 3. Sex 0 = female 1 = male 4. Proactive personality 5. Boundaryless mindset 6. Career self-ecacy 7. Identity awareness 8. Social support 9. Networking 10. Self-esteem 11. Job search 12. Re-employment 0 = unemployed 1 = re-employed
* **

Mean (sd) 33.63 (11.69) 0.57 (0.49)

1 1.00 0.02

2 1.00

10

11

12

0.62 (0.49)

0.01

0.02

1.00

3.68 3.62 3.61 3.60 3.55 3.40 3.34 3.76 0.30

(0.59) (0.56) (0.77) (0.69) (1.26) (0.94) (0.87) (0.77) (0.46)

0.01 0.03 0.05 0.05 0.09 0.05 0.06 0.04 0.07

0.06 0.07 0.09 0.01 0.11* 0.06 0.04 0.05 0.17

0.03 0.05 0.09 0.16** 0.04 0.07 0.04 0.11* 0.06

1.00 0.50** 0.42** 0.37** 0.14** 0.34** 0.36** 0.32** 0.24**

1.00 0.28** 0.36** 0.19** 0.21** 0.26** 0.26** 0.23*

1.00 0.27** 0.18** 0.23** 0.30** 0.28** 0.27**

1.00 0.22** 0.37** 0.15** 0.30** 0.11

1.00 0.21** 0.13** 0.21** 0.09

1.00 0.18** 0.35** 0.02

1.00 0.10 0.12

1.00 0.08

1.00

Correlation is signicant at the 0.05 level (2-tailed). Correlation is signicant at the 0.01 level (2-tailed).

255

256

S. McArdle et al. / Journal of Vocational Behavior 71 (2007) 247264

Study variables were entered into AMOS (Version 4) for modelling. Prior to assessing the structural model, the measurement model for the latent constructs (proactive personality, boundaryless mindset, career self-ecacy, identity awareness, networking, job search, and self-esteem) was assessed in order to ensure the validity of the measures. Given the positive correlations between each construct (as identied in the preliminary analyses) covariance was assumed between the latent constructs. All parcelled items were signicantly related to the latent constructs, with standardised regression weights of 0.53 or higher. This indicates strong validity of the measures. The goodness of t indices for the measurement model were: v2(149) = 288.559, p < .05; Goodness of Fit Index (GFI) = 0.935; TuckerLewis coecient (TLI) = 0.930; Comparative Fit Index (CFI) = 0.945; Incremental Fit Index (IFI) = 0.946; Root Mean Square Error of Approximation (RMSEA) = 0.048. While the v2 test was signicant (which indicates poor t), Hox and Bechger (1998) argue that, when large sample sizes are used, the v2 test is almost always signicant. According to Raykov, Tomer, and Nesselroade (1991) large samples have high statistical power, resulting in almost all reasonable models being rejected when only the v2 value and its associated probability are considered. Consequently, it is important to examine the other values. Analysis of the other indicators suggests that the model ts the data well. In particular, as the GFI, TLI, CFI, and IFI exceed the .90 acceptance level, and the RMSEA score is below .05, this model is accepted as a good t for the data (Hox & Bechger, 1998; Hoyle, 1995). In testing the structural model, education and social support were entered as observed variables, while job search, self-esteem, networking, identity awareness, career self-ecacy, boundaryless mindset and proactive personality were entered as latent variables reected by the parcelled items. Age and sex were initially entered as control variables, but were not found to be related to the outcome variables and were, therefore removed. The nal model is presented in Fig. 2.

Proactive Personality Education Self-esteem .81 Career Selfefficacy .68 Identity Awareness .68 .65 Networking .31 Job Search 42% .70 Employabilty n.s .44 -.21 20%

Boundaryless Mindset

.71

Social Support

Fig. 2. The relationships between employability, self-esteem and job search during unemployment (study 1 SEM results) (n.s., non-signicant).

S. McArdle et al. / Journal of Vocational Behavior 71 (2007) 247264

257

All the pathways presented in the model (Fig. 2) were signicant at the 0.05 level, except for the pathway between education and employability, represented by a dotted line. The goodness of t indices for this model were as follows: GFI = 0.921; TLI = 0.915; CFI = 0.927; IFI = 0.928; RMSEA = 0.050. These indices suggest a strong t of the model to the data. Analysis of the standardised regression weights indicated that proactive personality contributed the most to employability (0.81), followed by boundaryless mindset (0.70), career self-ecacy (0.68), identity awareness (0.68), networking (0.65), and social support (0.31). Overall, 20% of the variance in self-esteem and 42% of the variance in job search was explained by employability. 4.2. Study 2: longitudinal sample 4.2.1. Sample In study 2, 126 of the original 416 people completed the second-wave survey. Thirtynine of these people had gained re-employment, while 87 remained unemployed. Of the re-employed group, 59% were male and 41% female with a mean age of 33.77 years (sd = 10.92). In terms of education, 38.5% had completed university, 28.2% had nished high school, 17.9% had a TAFE qualication, 10.3% had gained an apprenticeship, and 5.1% had completed part of a university course. Of the unemployed group, 65% were male and 35% female, with a mean age of 35.46 (sd = 12.21). Forty-six percent had completed high school, 18.4% had nished university, 14.9% had a TAFE qualication, 12.6% had undertaken part of a university course, and 8% had completed an apprenticeship. In order to check for possible response bias, a number of statistical comparisons were performed between those baseline participants who participated in both studies 1 and 2 and those who only participated in study 1. The two groups did not dier signicantly on age (t(414) = 1.57, p = .117), gender (v2(1) = 0.04, p = .834), or education (v2(1) = 0.39, p = .530). The baseline scores for proactive personality, boundaryless mindset, career self-ecacy, identity awareness, networking, social support, job search and self-esteem were also compared between the two groups using Multivariate Analysis of Variance (MANOVA) and no dierences were found (F(8, 407) = 1.142, p = .334). This pattern of results indicates that the sample did not suer from attrition bias between study 1 and study 2. 4.3. Measures In study 2, survey participants were asked to indicate their employment situation. Individuals were coded as re-employed (1) if they had obtained full-time re-employment and as unemployed (0) if they were unemployed with little to no work of any sort. 4.3.1. Procedure Six months after the initial survey, all study 1 participants were contacted by mail with the study 2 questionnaire. Respondents were paid $25 for completion of the follow-up survey. Upon return, the questionnaires were coded to match each participant to their study 1 data.

258

S. McArdle et al. / Journal of Vocational Behavior 71 (2007) 247264

Proactive Personality Education Boundaryless Mindset Career Selfefficacy .60 Identity Awareness .52 n.s Networking Social Support .31 Job Search 39% .84 Employabilty Self-esteem .77 n.s .27 .49 n.s n.s Re-employment 16% 7%

.64

n.s

Fig. 3. The relationships between employability, self-esteem, job search and re-employment (study 2 SEM results) (n.s., non-signicant).

4.3.2. Results SEM was again used to test the study 2 hypotheses. The same variables were entered with the addition of the binary measure of re-employment. The study 2 model is presented in Fig. 3. The pathways between networking and employability; education and employability; self-esteem and job search; self-esteem and re-employment; and job search and re-employment were not signicant, and are represented by dotted lines in Fig. 3. All other pathways were signicant at the 0.05 level. The goodness of t indices for this model were as follows: GFI = 0.841; TLI = 0.891; CFI = 0.906; IFI = 0.910; RMSEA = 0.052. While the GFI and TLI fail to reach the .90 acceptance level, this may be due to the smaller sample size used in study 2. The CFI, IFI, and RMSEA gures are, however, at an acceptable level, suggesting that this model does t the data. Analysis of the standardised regression weights indicated that boundaryless mindset contributed the most to employability in study 2 (0.84), followed by proactive personality (0.77), career self-ecacy (0.60), identity awareness (0.52), and social support (0.31). In total, 16% of the variance in re-employment was explained by employability.

5. Discussion The aim of the current paper was to apply Fugate et al.s (2004) psycho-social construct of employability to an unemployed sample. Specically, this paper investigated the relationship between employability and (i) self-esteem during unemployment; (ii) job search during unemployment; and (iii) re-employment. Overall, the results of this longitudinal research provide support for Fugate et al.s (2004) employability model and demonstrate the applicability of employability in an unemployment context. The majority of the employability indicators measured in the current study were found to contribute signicantly to the model, with the exception of education (in relation to all

S. McArdle et al. / Journal of Vocational Behavior 71 (2007) 247264

259

outcome measures) and networking (in relation to re-employment only). In total, 20% of the variance in self-esteem and 42% of the variance in job search was explained by employability, oering support for Hypotheses 1 and 2. Hypothesis 4 was also supported, with 16% of the variance in re-employment explained by the employability indicators. Hypothesis 3, that employability would be indirectly, positively related to job search via selfesteem was not supported. Instead, it was found that self-esteem was negatively related to job search. Additionally, Hypotheses 5 and 6, that self-esteem and job search would be positively related to obtaining re-employment, were not supported. The adaptability variables (boundaryless mindset and proactive personality), career identity variables (identity awareness and career self-ecacy), and social support were all found to signicantly contribute to the employability model in both studies 1 and 2. Networking was also found to contribute signicantly to employability in study 1, but not in study 2. It is surprising that networking did not contribute signicantly to employability in study 2. It may be that the eects of networking show up more strongly during the initial months of unemployment and that the benets of this social capital may have been expended by the 6-month point. Qualitative data from interviews conducted in a separate part of this project indicate that participants often reduce their connections with friends and family as the period of unemployment lengthens. Wethington and Kessler (1986) found that with continuous unemployment, social support declined as the unemployed individual burnt out supporters in their core network. This suggests the value of interventions that encourage ongoing networking as part of an eective career assistance program. The human capital measure of education did not signicantly contribute to the employability model in either study. This does not mean that human capital is not an important component of employability, however, as this nding may be a result of the way human capital was measured in the current studies. Firstly, education was assessed quite crudely, as a binary construct (high school or less vs. post-high school), and hence, some relationships may have been distorted or obscured. Drawing ner distinctions between levels of education may produce dierent ndings in the future. Additionally, education was the only measure of human capital utilised in the research. The use of a variety of human capital indicators would be preferable, and is a limitation of this study. The signicant relationships between employability, self-esteem, job search and reemployment identied in this research demonstrate the applicability of a psycho-social construct of employability to individuals outside organisational boundaries (i.e. the unemployed). These ndings support Fugate et al.s (2004) contention that employability can be determined by more than ones employment status, and that it has utility across varied career contexts. This study indicates that employability can signicantly impact ones mental wellbeing during unemployment as the self-esteem of individuals higher on employability is less likely to suer during unemployment than those who have low employability. This may be because employable individuals take a more positive approach to unemployment, seeing it as an opportunity rather than a threat. Also, individuals high on employability may be more comfortable with uncertainty and more adaptable in novel situations. Additionally, their identity is not tied up in a particular job or organisation and, hence, the absence of a job does not lead to absence of identity. The support available from social networks may further buer self-esteem by providing a source of reassurance or guidance during unemployment.

260

S. McArdle et al. / Journal of Vocational Behavior 71 (2007) 247264

The signicant relationship between employability and job search suggests that employable individuals take a proactive approach to engaging with the labour market. Employable individuals may be likely to adjust to unemployment well due to their adaptability and associated comfort outside of organisational boundaries. As a consequence, they may be responsive to the situation by engaging in goal-directed behaviour, such as job search. The strong career identity associated with employability could also contribute to this by providing guidance and direction in an otherwise unstructured period. Additionally, social networks and social support may facilitate job search by providing information, motivation, and reassurance following rejection. Whilst it was hypothesised that self-esteem would be positively related to job search, study 1 found a negative relationship between the two variables. This nding contradicts research by Kanfer et al. (2001) who found that individuals higher on self-esteem engaged in greater job search. Crossley and Stanton (2005) recently found, however, that distress was also linked to job search intensity. Taken together, Kanfer et al.s (2001) and Crossley and Stantons (2005) results suggest that individuals undertake job search for dierent reasons, associated with both high and low self-esteem. This may be explained in terms of Higgins (1997) Regulatory Focus Theory, which proposes that individuals will approach important goals through either a prevention or promotion focus. A prevention focus centres on alleviating or avoiding negative outcomes, while a promotion focus is associated with attaining positive outcomes (Higgins, 1997; Idson & Higgins, 2000). An individual low on self-esteem may engage in job search as part of a prevention focus, in order to alleviate the distress associated with unemployment. Another, simpler, possibility might be that high self-esteem individuals have higher expectations for the kind of job that would be a good t, and if they are pessimistic about the likelihood of obtaining such a job in their particular job market, these high expectations could discourage job search. To explore all of these possibilities, in the future, it would be useful to investigate the persons specic employment goals and expectations and the role of regulatory focus in mediating the relationship between self-esteem and job search to better understand this nding. Employability was signicantly related to re-employment, suggesting that employability can assist an unemployed individual in nding a new job. The adaptability and career identity associated with employability are likely to be particularly important in obtaining re-employment. Firstly, adaptability is both benecial to the individual and attractive to potential employers (McLaughlin, 1995). Secondly, career identity may assist in reemployment by guiding career-related decisions and providing direction and focus in seeking a new role. Additionally, individuals with high career identity are typically aware of the skills and qualities they require to be successful in their career. This may provide them with insight into the types of jobs that can help them achieve their career goals. Social support could also contribute to ones re-employment by providing access to potential employers and by providing encouragement to the unemployed individual, boosting their condence during the recruitment process. Self-esteem, as an independent variable, was not related to re-employment. This does not support Kasls (1982) reverse-causation hypothesis but is, instead, consistent with Waters and Moores (2002a) nding that the two variables are not related. According to Shamir (1986) individuals who are higher on self-esteem are more discerning in terms of re-employment and are unwilling to accept less-than-ideal jobs. This suggests that individuals who are higher on self-esteem may only accept a job if it is congruent with their

S. McArdle et al. / Journal of Vocational Behavior 71 (2007) 247264

261

desires and expectations, while individuals who are lower on self-esteem may be more likely to accept the rst job they are oered.1 This would, therefore, be unlikely to lead to a clear-cut association between self-esteem and re-employment. Perhaps future researchers should consider assessing not only whether re-employment was or was not obtained, but also the quality of the re-employment and the personjob t obtained in the new job. It could be that self-esteem plays a stronger role in these latter two aspects of re-employment. Job search was not signicantly related to re-employment. This nding is surprising and contradicts Wanberg (1997), Wanberg et al. (1999), Wanberg et al. (2001) and Kanfer et al.s (2001) ndings of an association between the two. The current nding may reect the fact that job search is internally driven by the individual, while re-employment is controlled by a third party (the employer). Those engaging in high job search intensity may not necessarily be the most attractive to employers. Additionally, given that all study participants were accessed via the JobNetwork agency (an organisation set up specically to assist in the job search process) it is likely that the base level of job search for all participants was relatively high, which may have impacted the results. Moreover, Wanberg et al. (1999) suggest that unemployed individuals may underestimate the amount of job search required to be successful in gaining re-employment. Consequently, some respondents may believe they are engaged in a high level of job search, when in reality, they are engaging in relatively lower job search in comparison to other respondents. In addition, it is important to note that only 39 people gained re-employment in this study, less than 10% of the original respondents. Also, there was only a 6-month time period that had elapsed between study 1 and study 2. There may have been other factors at such an early stage of the unemployment process that facilitated re-employment for these individuals other than job search. For example, they may have knowledge or skills that are highly valued and were immediately noticed in the labour market and were not captured by the survey. Alternatively, there may have been people who accepted less-than-ideal jobs in order to gain re-employment, while other individuals may have rejected re-employment opportunities, instead engaging in greater job search in order to nd better roles. This proposition is supported by Wanberg et al.s (1999) nding that job search was related to gaining re-employment but was not predictive of the quality of that re-employment. Furthermore, in study 2, baseline job search was used to predict re-employment at a 6month follow-up. It is possible that the level of job search may have changed over the 6 months between surveys. Individuals low on job search at time one may have upped the ante in the interim and increased their job search intensity prior to nding re-employment, while others may have had high job search at time one but become disillusioned and withdrawn from job search activities in the intervening period. In the future, it could also be useful to investigate job search strategy or quality in addition to the quantity/intensity. It may be that those individuals who gained re-employment were more focused in their job search activity, taking greater time to adapt each job application to the employers specic needs, rather than using a scatter-gun approach, applying for anything and everything on oer. Given that this study was the rst to operationalise the employability dimensions proposed by Fugate et al. (2004), it drew on existing scales and, subsequently, serves

Our interview data, collected for another part of this project, provide support for this explanation.

262

S. McArdle et al. / Journal of Vocational Behavior 71 (2007) 247264

somewhat of an exploratory function. Rather than tting existing scales to Fugate et al.s (2004) model, however, it would be preferable to develop an employability inventory specically designed to measure the psycho-social employability construct. As mentioned earlier, a key limitation of this research was the measurement of human capital. One of the key problems in operationalising human capital is a lack of consensus in the wider literature regarding what it actually comprises. Measuring the construct is likely to be problematic until more denitional research is conducted. Additionally, a more nuanced measure of social capital could be fruitful. Measuring the type of social networks, their strength and size could contribute to a better understanding of this variable. Similarly, measuring the dierent types of social support (e.g. informational, instrumental, and emotional) could also prove benecial. Despite these limitations, however, the relationships identied thus far encourage further investigation into employability. As the rst study to empirically test Fugate et al.s (2004) psycho-social model of employability, this research provides broad support for the model, and demonstrates the signicant role employability plays in relation to self-esteem, job search, and reemployment. The ndings of this study provide both theoretical and practical tools to help understand and assist individuals during unemployment. It is hoped that the relationships identied here will encourage further research in this area. References
Anakwe, U., Hall, J., & Schor, S. (2000). Knowledge-related skills and eective career management. International Journal of Manpower, 21(7), 566579. Bagshaw, M. (1997). Employabilitycreating a contract of mutual investment. Industrial and Commercial Training, 29(6), 187189. Bateman, T., & Crant, J. (1993). The proactive component of organizational behavior. Journal of Organizational Behavior, 14, 103118. Briscoe, J., Hall, D., & DeMuth, R. (2006). Protean and boundaryless careers: An empirical exploration. Journal of Vocational Behavior, 69, 3047. Brown, P., Hesketh, A., & Williams, S. (2003). Employability in a knowledge-driven economy. Journal of Education and Work, 16(2), 107126. Crant, J. (2000). Proactive behavior in organizations. Journal of Management, 26, 435462. Crossley, C., & Stanton, J. (2005). Negative aect and job search: Further examination of the reverse causation hypothesis. Journal of Vocational Behavior, 66, 549560. Dellippi, R., & Arthur, M. (1994). The boundaryless career: A competency-based perspective. Journal of Organizational Behavior, 15(5), 307. Eby, L. T., & Buch, K. (1995). Job loss as career growth: Responses to involuntary career transitions. The Career Development Quarterly, 44(1). Eby, L. T., Butts, M., & Lockwood, A. (2003). Predictors of success in the era of the boundaryless career. Journal of Organizational Behavior, 24(6), 689. Eden, D., & Aviram, A. (1993). Self-ecacy training to speed reemployment: Helping people to help themselves. Journal of Applied Psychology, 78(3), 352360. Finn, D. (2000). From full employment to employability: A new deal for Britains unemployed? International Journal of Manpower, 21(5), 384399. Fugate, M., Kinicki, A. J., & Ashforth, B. E. (2004). Employability: A psycho-social construct, its dimensions, and applications. Journal of Vocational Behavior, 65(1), 14. Garavan, T. (1999). Employability, the emerging new deal? Journal of European Industrial Training, 23(1). Hall, D. T. (1996). Protean careers of the 21st century. The Academy of Management Executive, 10(4), 816. Hall, D. T. (2004). The protean career: A quarter-century journey. Journal of Vocational Behavior, 65(1), 113. Hall, D. T., Briscoe, J. P., & Kram, K. E. (1997). Identity, values and learning in the protean career. In C. L. Cooper & S. E. Jackson (Eds.), Creating tomorrows organizations (pp. 321332). Chichester: John Wiley & Sons Ltd.

S. McArdle et al. / Journal of Vocational Behavior 71 (2007) 247264

263

Hanisch, K. (1999). Job loss and unemployment research from 1994 to 1998: A review and recommendations for research and intervention. Journal of Vocational Behavior, 55(2), 188220. Harter, S. (1990). Processes underlying adolescent self-concept formation. In R. Montemayor, G. R. Adams, & T. P. Gullota (Eds.), From childhood to adolescence (pp. 205239). Sage, CA: Newbury Park. Higgins, E. (1997). Beyond pleasure and pain. American Psychologist, 52, 12801300. Hox, J., & Bechger, T. (1998). An Introduction to Structural Equation Modelling. Family Science Review, 11, 354373. Hoyle, R. (Ed.). (1995). Structural equation modelling: Concepts, issues, and applications. Sage, CA: Thousand Oaks. Idson, L., & Higgins, E. (2000). How current feedback and chronic eectiveness inuence motivation: Everything to gain versus everything to lose. European Journal of Social Psychology, 30, 583592. Jahoda, M. (1982). Employment and unemployment: A social-psychological analysis. New York: Cambridge University Press. JobNetwork (2005). JobNetwork: An Australian Government Initiative. Retrieved August 11. Available from http://workplace.gov.au/workplace/jobnetwork. Kanfer, R., Wanberg, C. R., & Kantrowitz, T. M. (2001). Job search and employment: A personalitymotivational analysis and meta-analytic review. Journal of Applied Psychology, 86(5), 837855. Kasl, S. V. (1982). Strategies of research on economic instability and health. Psychological Medicine, 12, 637649. Kinicki, A., & Latack, J. (1990). Explication of the construct of coping with involuntary job loss. Journal of Vocational Behavior, 36, 339360. Kossek, E., Roberts, K., & Demarr, B. (1998). Career self-management: A quasi-experimental assessment of the eects of training intervention. Personnel Psychology, 51(4), 935962. Lai, J., & Chan, K. (2002). The eects of job-search motives and coping on psychological health and reemployment: A study of unemployed Hong Kong Chinese. International Journal of Human Resource Management, 13(3), 465483. Landis, R. S., Beal, D. J., & Tesluk, P. E. (2000). A comparison of approaches to forming composite measures in structural equation models. Organization Research Methods, 3, 186207. Latack, J. C., & Dozier, J. B. (1986). After the ax falls: Job loss as a career transition. Academy of Management Review, 11, 375392. Lazarus, R., & Folkman, S. (1984). Stress, appraisal and coping. New York: Springer. McGreevy, M. (2003). The changing nature of work. Industrial and Commercial Training, 35(5), 191195. McIntosh, N. (1991). Identication and investigation of properties of social support. Journal of Organizational Behavior, 12(3), 201217. McKee-Ryan, F. M., Song, Z., Wanberg, C. R., & Kinicki, A. J. (2005). Psychological and physical well-being during unemployment: A meta-analytic study. Journal of Applied Psychology, 90(1), 5376. McLaughlin, M. (1995). Employability skills prole: What are employers looking for? Eric Digests the Educational Resources Information Center (ERIC). Available from www.eric.ed.gov. Messer, B., & Harter, S. (1986). Adult self-perception prole: Manual. Denver: Denver University Press. OConnell, D., McNeely, E., Hall, D. (in press). Unpacking personal adaptability at work. Journal of Leadership and Organizational Studies. Price, R., & Fang, L. (2002). Unemployed Chinese workers: The survivors, the worried young and the discouraged old. International Journal of Human Resource Management, 13, 416430. Prussia, G. E., Fugate, M., & Kinicki, A. J. (2001). Explication of the coping goal construct: Implications for coping and reemployment. Journal of Applied Psychology, 86(6), 11791190. Raykov, T., Tomer, A., & Nesselroade, J. (1991). Reporting structural equation modelling results in psychology and aging: Some proposed guidelines. Psychology and Aging, 6(4), 499503. Ritchie, J. (2000). New deal for young people: Participants perspectives. Policy Studies, 21(4), 301312. Rosenberg, M. (1965). Society and the adolescent self-image. Princeton: Princeton University Press. Rousseau, D. (1997). Organizational behavior in the new organizational era. Annual Review of Psychology, 48, 515546. Savickas, M. (1997). Career adaptability: An integrative construct for life-span, life-space theory. Career Development Quarterly, 45(3), 247259. Seibert, S., Crant, J., & Kraimer, M. (1999). Proactive personality and career success. Journal of Applied Psychology, 84, 416427. Seibert, S., Kraimer, M., & Crant, J. (2001). What do proactive people do? A longitudinal model linking proactive personality and career success. Personnel Psychology, 54(4), 845.

264

S. McArdle et al. / Journal of Vocational Behavior 71 (2007) 247264

Seibert, S., Kraimer, M., & Liden, R. (2001). A social capital theory of career success. Academy of Management Journal, 44, 219237. Shaer, D. (2002). Developmental psychology: Childhood and adolescence. CA: Wadsworth Thomson Learning. Shamir, B. (1986). Self-esteem and the psychological impact of unemployment. Social Psychology Quarterly, 49(1), 6172. Stumpf, S., Colarelli, S., & Hartman, K. (1983). Development of the career exploration survey. Journal of Vocational Behavior, 22(2), 191226. Thompson, J. (2005). Proactive personality and job performance: A social capital perspective. Journal of Applied Psychology, 90(5), 10111017. Van der Heijden, B. (2002). Prerequisites to guarantee life-long employability. Personnel Review, 31(1), 4461. Vinokur, A., & Caplan, R. (1987). Attitudes and social support: Determinants of job-seeking behavior and wellbeing among the unemployed. Journal of Applied Social Psychology, 17, 1007. Vinokur, A., & Schul, Y. (1997). Mastery and inoculation against setbacks as active ingredients in the JOBS intervention for the unemployed. Journal of Consulting and Clinical Psychology, 65, 867877. Wanberg, C. R. (1997). Antecedents and outcomes of coping behaviours among unemployed and reemployed individuals. Journal of Applied Psychology, 82(5), 731744. Wanberg, C. R., Kammeyer-Mueller, J. D., & Shi, K. (2001). Job loss and the experience of unemployment: International research and perspectives. In N. Anderson, D. Ones, H. Sinangil, & C. Visuresvaran (Eds.). The handbook of industrial work and organizational psychology (Vol. 2, pp. 253269). London: Sage Publications Inc. Wanberg, C. R., Kanfer, R., & Rotundo, M. (1999). Unemployed individuals: Motives, job-search competencies, and job-search constraints as predictors of job seeking and reemployment. Journal of Applied Psychology, 84(6), 897910. Waters, L., & Moore, K. (2002b). Psychological health and coping: A comparison of unemployed and reemployed people. Journal of Organizational Behavior, 23, 593604. Waters, L., & Moore, K. (2002a). Reducing latent deprivation during unemployment: The role of meaningful leisure activity. Journal of Occupational and Organizational Psychology, 75, 1532. Waters, L., & Muller, J. (2003). Money or time? An investigation of the deprivation and agency theories in explaining psychological well-being of unemployed adults. Australian Journal of Psychology, 55(3), 166175. Wethington, E., & Kessler, R. C. (1986). Perceived support, received support, and adjustment to stressful life events. Journal of Health and Social Behavior, 27(1), 7889. Wineeld, A., & Tiggeman, M. (1985). Psychological correlates of employment and unemployment: Eects, predisposing factors and sex dierences. Journal of Occupational Psychology, 58, 229242. Wineeld, A., Tiggeman, M., & Wineeld, H. (1992). Unemployment distress, reasons for job loss and causal attributions for job loss in young people. Journal of Occupational and Organizational Psychology, 65, 213218. Winegardner, D., Simonetti, J., & Nykodym, N. (1984). Unemployment: The living death. Journal of Employment Counselling, 21, 149155.