Experiential differences between voluntary and involuntary job redundancy on depression, job-search activity, affective employee outcomes and re-employment quality.
The current paper reports the findings of two studies. Study 1 uses a cross-sectional design to compare levels of depression and job-search activity during unemployment between people who volunteered for their job redundancy and employees who were made involuntarily redundant. Study 2 is follow-up research that examines the subset of participants who gained re-employment in the 3 months following Study 1. The voluntary and involuntary groups are compared on depression, perceived job insecurity, organizational commitment and perceived re-employment quality. Predictions for differences in depression and job search during unemployment in Study 1 will be made using the learned helplessness theory (Abramson, Seligman, & Teasdale, 1978; Peterson & Seligman, 1983) (Hypotheses 1 and 2). In addition, learned helplessness theory will be used to make predictions about depression and quality of re-employment in Study 2 (Hypotheses 3 and 6). Predicted differences in depression, organizational commitment and perceived job insecurity upon re-employment in Study 2 are based upon Fineman's (1983) unemployment legacy (Hypotheses 4 and 5).
This paper contributes to the job loss literature by following a group of employees, who experience job redundancy, into their next job to see if there is any carry-over effect in the new job from being made redundant. Although there is a vast body of literature on the effects of job loss, there is comparatively little research on the reemployment experiences of people who have been made redundant. Indeed, Pugh, Skarlicki, and Passell (2003) state that 'the relationship between the layoff experience and victims' attitudes at their new place of work remains relatively unexplored' (p. 201). Yet, as suggested by Kinicki, Prussia, and McKee-Ryan (2000) 'research on what happens to displaced workers after they become re-employed is clearly needed' (p. 98).
This paper makes a further contribution to the job loss literature by separating job redundancy into the categories of voluntary and involuntary. Involuntary redundancy programmes target employees for job loss despite their own wishes (Macken, O'Grady, & Sappideen, 1997). Voluntary redundancy programmes 'allow eligible employees to choose whether or not to participate' (DeWitt, Trevino, & Mollica, 1998, p. 594).
A review of the broader job loss and turnover literature shows that job redundancy is typically treated as a form of involuntary turnover, so that terms such as lay-off, downsizing and job redundancy are often automatically taken to mean that the employee's exit was involuntary and that the employee was a passive actor in the job loss process (e.g. Campion, 1991; Iverson & Pullman, 2000). However, there is growing evidence of the use of voluntary workforce reduction programmes (e.g. DeWitt et al., 1998; Morehead, Steele, Alexander, Stephen, & Duffin, 1997). For example, companies such as Nortell Networks, Australia Post, General Motors and American Airlines all allow employees the opportunity to volunteer for their own job redundancy. Yet, despite the growth in the use of voluntary job redundancy programmes, the phenomenon of voluntary job redundancy has received inadequate research attention.
Leana and Feldman (1992), Miller and Hoppe (1994) and Winefield, Tiggemann, and Winefield (1992) suggest that the experience of job-leavers (e.g. those that quit) is likely to differ from job-losers (e.g. those that are fired or made redundant). Despite the appeal of this proposition, Hanisch (1999) concluded that '[m]ost of the research treats unemployment as a state removed from the process of termination' (p. 192). In the rare instance where the reason for a participant's job loss is reported (e.g. Wanberg, 1995; Wanberg, Hough, & Song, 2002), it is either not used in the statistical analysis or it is used only as a control variable. The current study will address this gap by comparing the experiences of involuntarily redundant employees with voluntarily redundant employees. *
STUDY I: DIFFERENCES IN DEPRESSION AND JOB SEARCH DURING UNEMPLOYMENT BETWEEN VOLUNTARILY AND INVOLUNTARILY REDUNDANT GROUPS
This study follows suggestions made by Feather (1990), Rodriguez (1997) and Winefield (2002) that learned helplessness may play a role in people's psychological reactions during unemployment. Depression and passivity, two major aspects of the learned helpless theory, will be used to make predictions about levels of depression and low job-search activities (a manifestation of passivity) in unemployed people who experience voluntary and involuntary job redundancy.
According to the basic tenets of learned helplessness theory, when someone is exposed to a negative, uncontrollable event they may conclude that their efforts are unrelated to their outcomes. In the context of the current study, when people experience involuntary job loss they may conclude that their job skills and performance are unrelated to their ability to maintain a secure job. One of the major consequences of this 'response-outcome independence' (Peterson & Seligman, 1983, p. 103) is depression (Abramson et al., 1978). According to Abramson et al., depression results when an individual is confronted with an aversive event and where the individual believes that they cannot change the event (e.g. involuntary job loss). Latack, Kinicki, and Prussia (1995) argued that when unemployed people 'do not see a connection between their efforts and outcomes, they can become depressed' (p. 332). Hence, if people who experience involuntary job redundancy see a disconnection between their behaviour and outcomes, this may lead to depression. Learned helplessness theory has, therefore, been used to develop the following hypothesis.
Hypothesis I. People who experience involuntary job redundancy will report higher levels of depression during unemployment than those who experience voluntary job redundancy.
Feelings of 'response-outcome independence' (Peterson & Seligman, 1983, p. 103) may also create a response of passivity through an expectation that future actions in the broader-related domain (e.g. employment) will also be unsuccessful (Seligman, 1975). Feather (1990) argued that when people feel helpless during unemployment they may reduce their job-search and coping efforts 'because they expect that regardless of how hard they try they will remain unemployed' (p. 72). It is therefore argued in this paper that a person's feelings of helplessness over job loss, created via involuntary redundancy, may be transferred into passivity over job gain. In this situation, it is likely that these people may not engage in high levels of job-search activities due to a belief that their effort (job search) is unlikely to affect the outcome (re-employment). Learned helplessness theory has been used to inform Hypothesis 2.
Hypothesis 2. People who experience involuntary job redundancy will report lower levels of job-search activity during unemployment than those who experience voluntary job redundancy.
STUDY 2: DIFFERENCES BETWEEN VOLUNTARILY AND INVOLUNTARILY REDUNDANT RE-EMPLOYEES ON DEPRESSION, PERCEIVED JOB INSECURITY, ORGANIZATIONAL COMMITMENT, AND PERCEIVED RE-EMPLOYMENT QUALITY
It is now well-established that people moving from unemployment to re-employment generally report improvements in psychological health, particularly if the re-employment is of a high quality (Caplan, Vinokur, Price, & Van Ryn, 1989; Wanberg, 1997). Jahoda (1988) argued that improvements in psychological health upon re-employment occur because employment provides people with social status, professional identity and collective purpose, each of which are difficult to obtain during unemployment but are vital for psychological health.
However, it is not always the case that psychological health improves upon reemployment. For instance, Fineman (1983) found that, for some re-employed people, 'a sour legacy remained' (p. 7) and depression did not automatically lift. Other authors have also found evidence of a legacy effect where distress, coping and life satisfaction do not improve upon re-employment (Latack et al., 1995; Leana & Feldman, 1995).
The question remains as to why some people report reductions in distress upon reemployment while others do not. This paper posits that type of job redundancy may be one factor partly to influence whether people experience a 'sweet or sour' effect upon re-employment. Although Latack and Dozier (1986) suggested that the legacy effect may be influenced by the type of termination experienced, their proposition has not been empirically tested.
Learned helplessness theory was used to predict that people who lose their job involuntarily may suffer higher levels of depression than those who volunteer for redundancy. Deep depression associated with involuntary job loss may be experienced as a longer-term phenomenon, not lifting immediately upon re-employment. In contrast, if people who volunteer for redundancy experience only minor depression during unemployment they may follow the more typical pattern of reductions to distress upon re-employment.
Hypothesis 3. Employees who experience involuntary job redundancy will continue to report high levels of depression in their new job while employees who volunteered for their job redundancy will experience a drop in the level of depression in their new job.
As well as its influence on psychological health upon re-employment, the legacy effect has also been found to influence affective employee outcomes such as commitment, insecurity, distrust and cynicism (Burke, 1986; Feldman & Leana, 2000; King, 2002). Again, there has been no empirical test of whether a legacy effect on affective employee outcomes is influenced by the way in which a person exited their last job. However, indirect evidence of a relationship between type of redundancy and a legacy effect was found by Pugh et al. (2003). Pugh et al. traced the relationship between lay-off experience and cynicism and distrust towards one's new place of work, concluding that the 'context of a layoff can predict layoff victim's attitudes beyond the organization in which layoff occurred' (p. 208). It appears that those employees who experienced psychological contract violation also experienced a more prominent legacy effect.
Applying these findings to the current paper, it could be argued that those employees who experience involuntary redundancy are more likely to perceive a psychological contract violation than those employees who volunteer for redundancy. This is because an involuntary job redundancy contravenes two key aspects of the psychological contract that of 'voluntary choice' and 'mutual agreement' (Rousseau, 2004) as the person loses their job against their own wishes and the decision of job loss is made solely by the employer. In contrast, although voluntary job redundancy signals an end to the employer-employee contract, it does so without violating the two principles listed above. It is suggested, then, that the legacy effect may be stronger in employees made involuntarily redundant.
Two factors that have been highlighted as being particularly negatively affected by downsizing are job insecurity and organizational commitment (Brockner, Grover, Reed, & DeWitt, 1992; Brockner, Wiesenfeld, Reed, & Grover, 1993). To date, tests of how downsizing affects job insecurity and organizational commitment have focused on survivors of downsizing. Research consistently shows that these employees report feelings of high perceived job insecurity and low commitment to their organization, even though they survived job redundancy.
Although there has been no research testing the effect of voluntary and involuntary job redundancy on insecurity and organizational commitment in re-employed layoff victims, it seems reasonable to suggest that victims are as likely, if not more likely, than survivors to report the same. Job insecurity is a perceptual phenomenon that reflects, in part, the degree to which an individual believes they can maintain their current employment (Hartley, Jacobson, Klandermans, & Van Vuuren, 1991). It is reasonable to argue that the experience of involuntary redundancy in one's previous job may affect the person's feelings of security, or lack thereof, in their next job. Certainly, King (2002) argued that
anything that might lower one's confidence in keeping their job is a potential antecedent to job insecurity. Being a previous victim of a layoff is precisely the kind of experience that could make for lower confidence in the stability of one's employment, because it personalizes a fear that others only experience in abstract terms (p. 28).
Roskies and Louis-Guerin (1990) sampled 1,055 Canadian managers and found that the experience of a previous job dislocation significantly predicted perceived job insecurity in their current position, and in fact, was a stronger predictor of current job insecurity than an index measuring the stability of the current employer. The question remains open whether the same legacy effect will be present for people who volunteer to leave their job. Roskies and Louis-Guerin suggest that insecurity occurs when job loss is viewed as a threatening event. They also found that unpredictability of prior job loss fed into insecurity in a new job. Presumably, the act of volunteering for one's redundancy negates a sense of unpredictability. Moreover, it is likely that people who volunteer for redundancy do so as they do not see job loss as a highly threatening event. Thus, it could be suggested that those people who volunteer for job redundancy may not face a legacy effect of high perceived job insecurity upon re-employment.
Hypothesis 4. Re-employed participants who experienced involuntary job redundancy will report higher levels of job insecurity in their new jobs than participants who volunteered for their own redundancy.
To date, there has been no research tracing the effect of job redundancy on organizational commitment once re-employed. However, the consistent negative relationship found between insecurity and organizational commitment in the broader management literature (Hartley et al., 1991; Rosenblatt & Ruvio, 1996) can be used to suggest that as involuntarily redundant employees feel more insecure, they may also feel less committed in their new job. In support of this, Roskies and Louis-Guerin's (1990) sample of managers reported a significant, negative relationship between perceived job insecurity and organizational commitment. King (2002) suggests that the insecurity felt by previously redundant employees may make them unwilling to commit to their new workplace for fear that they will again suffer the same fate.
Hypothesis 5. Re-employed participants who experienced voluntary job redundancy will report higher levels of organizational commitment in their new jobs than participants who experienced involuntary job redundancy.
Hypothesis 6. Perceived job insecurity will mediate the relationship between type of redundancy and organizational commitment.
Finally, learned helplessness can also be considered in terms of the perceived quality of re-employment obtained. Research has found that some unemployed people are reemployed into lower quality jobs assessed via self-reports on factors such as pay rate, level of seniority, skill use and employment terms (e.g. part-time vs. full-time hours and short-term contracts vs. tenured jobs; Burke, 1986; Mallinckrodt, 1990). While the move to lower quality employment may be a function of changes in the external labour market (Romeyn, 1992), research also suggests that that the perceived quality of re-employment obtained may be influenced by the job seeker's psychological state (Schaufeli & Vanyperen, 1993). Certainly, Leana and Feldman (1995) found that unemployed people with low levels of optimism and high levels of depression (both of which are aspects of learned helplessness) 'may jump at less-than-satisfactory opportunities because they fear nothing will come along' (p. 1386). Hence, if involuntary job redundancy is associated with higher levels of learned helplessness, it may also be associated with a perception of lower quality re-employment.
Hypothesis 7. Employees who experience involuntary job redundancy will report lower quality re-employment than employees who volunteer for their own job redundancy.
The sample used in Study 1 consisted of 102 job seekers who had voluntarily taken a job redundancy and 114 job seekers who had been made involuntarily redundant from their last job. The mean age of the voluntarily redundant group was 36.55 [+ or -] 8.39 years, 48% were female, 61% were blue-collar workers and 39% were white-collar workers, 24% had completed secondary level education or its equivalent while 76% were tertiary educated. Thirty-seven per cent received outplacement assistance.
The mean age of the involuntarily redundant group was 37.66 [+ or -] 8.75 years, 43% were female, 50% were blue-collar workers and 50% were white-collar workers, 46% had completed secondary level education or its equivalent whilst 54% were tertiary educated. Thirty-three per cent received outplacement assistance.
Chi-squared analysis and t tests were conducted in order to determine whether there were any differences in the demographic profile of the two groups. There was a significant association between type of redundancy and educational level, [chi square] (1) = 11.03, p = .001, where a higher number of people who were tertiary educated were in the voluntarily redundant group.
There was no association found between type of job redundancy with gender [chi square] (1) = 0.02, p = .879, outplacement assistance [chi square] (1) = 1.44, p = .230 and occupational status [chi square] (1) = 3.04, p = .081. There was also no significant difference between the two groups on age t(214) = -0.95, p = .346. No significant differences were found by t-test analysis on the amount of welfare payment provided to the two groups on a fortnightly basis by the Australian Government. This final analysis was performed by staff at Centrelink, the Government agency that provides welfare payments to job seekers. Owing to issues of confidentiality, staff from Centrelink did not divulge the amount of fortnightly financial assistance received by the two groups to the researcher.
Job seeker identification
The Centrelink database was used to identify and classify job seekers as voluntarily or involuntarily redundant. The Centrelink system records job loss as voluntary or involuntary based upon a letter received from the job seeker's previous employer outlining the type of job exit.
Kinicki and Latack's (1990) proactive job-search subscale, from their Coping with Job Loss scale, was used to measure job-search activity. This scale comprises five items (e.g. 'Get together with job contacts and people who can find me a job'), measured using a five-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). Higher scores indicate higher job-search activities. The alpha coefficient was .73 for the voluntarily redundant group and .81 for the involuntarily redundant group.
The depression subscale, from the Profile of Mood States (McNair, Lorr, & Droppelmann, 1981) was used to measure the level of depression experienced in the past week. Participants responded to an adjective checklist of 15 items (e.g. blue, unhappy, miserable), using a five-point Likert scale ranging from 1 (not at all) to 5 (extremely). Higher scores indicate higher levels of depression. The alpha reliabilities were .94 and .96, respectively, for voluntarily and involuntarily redundant participants at baseline.
The focus of the baseline study was to see whether levels of depression and job search differed in employees who volunteered for redundancy compared with those made involuntarily redundant. However, it must be recognized that variations between the voluntary and involuntary groups in depression and job search may be a function of variables other than the type of redundancy. For instance, the following seven variables have been shown to be significantly related to depression and job search during unemployment: self-efficacy, satisfaction with previous job, age, sex, outplacement assistance, employment commitment and occupational status (Feldman & Leana, 2000; Latack et al., 1995; Waters & Moore, 2002). It may be that group variances found between the voluntary and involuntary groups are due to differences in the variables listed above rather than the way in which the person exited their last job. As such, the bivariate relationship between these potential covariates and the dependent variables will be tested through correlation analyses. Following the recommendation of Tabachnick and Fidell (1996), any of the variables that are significantly correlated with the dependent variables will be entered as covariates into the subsequent statistical analyses.
Generalized self-efficacy The Generalized Self-Efficacy scale (Sherer et al., 1982) was used to assess generalized self-efficacy. This scale comprises 17 items (e.g. 'Failure just makes me try harder'), measured using a five-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). Higher scores indicate higher levels of generalized self-efficacy. The alpha coefficient was .71 for the voluntarily redundant group and .71 for the involuntarily redundant group.
Stafford, Jackson, and Bank's (1980) scale was used to assess employment commitment. This scale has six items (e.g. 'If I had enough money I would still want to work'), measured using a five-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). Higher scores indicate higher levels of employment commitment. The alpha coefficient was .88 for the voluntarily redundant group and .85 for the involuntarily redundant group.
Retrospective job satisfaction
Retrospective job satisfaction was measured using a single item where participants were asked to think back to their previous job and rate their satisfaction from 1 (extremely dissatisfied) to 5 (extremely satisfied). Participants were asked to specify how many days their employer allowed them to remain working after they had been told about, or volunteered for, their job loss. Participants were asked to record whether their employer provided them with outplacement assistance on a yes or no response format.
Data on age, gender, educational level (secondary/tertiary), length of unemployment and occupational status (blue-collar/white-collar) were collected. Occupations were classified using the Australian Standard Classification of Occupations (Department of Employment, Education, Training and Youth Affairs, 1997).
Job seekers who were receiving government unemployment benefits from Centrelink were recruited through 'Job Network', which is a national Australian service made up of private, government and community agencies that provide job referrals to job seekers. Five hundred surveys were sent via the Job Network to job seekers who had recently become unemployed via job redundancy (typically because their company had downsized) and a total of 222 people (44%) responded.
Both groups had been unemployed, on average, for 3 months. Participants were targeted at this stage in their unemployment to ensure that some people became re-employed before the second study. This is based on Wanberg et al.'s (2002) research which shows that most individuals find new work after 3 months of unemployment.
Tests of scale validity
Before testing the study hypotheses, it was necessary to establish the validity of the study variables. The two dependent variables (depression and job-search activities) and two of the potential covariates (generalized self-efficacy and employment commitment) were assessed using multi-item scales and were able to be submitted to tests of factor structure. A confirmatory factor analyses was run on the scale items using Judge, Erez, Bono, and Thoresen's (2002) parcelling approach. Missing values were replaced on 19 cases via the EM procedure using information available from all of the parcel variables to predict every other parcel variable. Following the recommendations of Hu and Bentler (1999) and Marsh, Balla, and McDonald (1988), the ensuing measures of model fit were checked: root mean square error of approximation (RMSEA) with upper and lower 90% confidence intervals, standardized root mean residual (SRMR), Tucker-Lewis index (TLI) and comparative fit index (CFI). Table 1 reports the fit indicators and shows that the model statistics suggested a reasonable fit for the four-factor model.
In a further attempt to assess the validity of the scales, a one-factor and a two-factor model were tested as competing tests to the four-factor model. The two-factor model combined employment commitment, self-efficacy and job-search activities on to the first factor. Depression was the second single factor. Both the one-factor and two-factor models showed a poor fit. Since these models were nested, chi-square comparisons could be performed. These chi-squared comparisons showed that the two-factor model was a better fit than the one-factor model, [chi square] change (1) = 641.0, p < .001. The four-factor model was also a better fit than the one-factor model, [chi square] change (6) = 1531.6, p < .001. Finally, the four-factor model was a better fit than the two-factor model, [chi square] change (5) = 512.6, p < .001. Hence, the four-factor model was the best fit to the data.
One hundred and twenty-two people participated in the 3-month retest. The response rate was comparable across the two groups (62% voluntarily redundant; 57% involuntarily redundant). Of the voluntarily redundant group, 38 people remained unemployed (58%) and 27 had become re-employed (42%). Of the involuntarily redundant group, 39 remained unemployed (60%) and 26 had become re-employed (40%). The proportion of participants who gained re-employment was similar to that found by Gowan, Riodan, and Gatewood (1999) and Wanberg, Watt, and Rumsey (1996).
The mean age of the participants in the voluntarily redundant group who had become re-employed at the 3-month retest was 35.63 [+ or -] 7.12 years. Of these, 37% were female, 63% were blue-collar workers and 76% were tertiary educated. The mean age of the participants in the involuntarily redundant group who had become re-employed at the 3-month retest was 34.04 [+ or -] 6.57 years. Of these, 50% were female, 69% were blue-collar workers, 80% were white-collar workers and 80% were tertiary educated.
Chi-squared analysis and t tests were conducted in order to determine whether there were any differences in the demographic profile of the two subgroups of re-employees. There was a significant difference in the length of advance notification given to voluntarily redundant employees ([bar.X] = 25.8 days, 95% CI = [+ or -] 2.2) and involuntarily redundant employees ([bar.X] = 4.9 days, 95% CI = [+ or -] 2.2), t(51) = 15.55, p = .001. However, there was no association found between type of job redundancy and gender [chi square] (1) = 0.91, p = .341, occupational status [chi square] (1) = 0.23, p = .630, receipt of outplacement assistance [chi square] (1) = 3.57, p = .059, educational level [chi square] (1) = 3.57, p = .059 and age t(51) = 0.85, p = .40.
At the 3-month retest, the same scale was used to assess depression. The alpha scores for the depression subscale of the Profile of Mood States (McNair et al., 1981) were .97 and .86, respectively, for voluntarily and involuntarily redundant participants. Three new dependent variables were added: perceived job insecurity, organizational commitment and perceived re-employment quality.
Perceived job insecurity
Hellgren, Sverke, and Isaksson's (1999) Quantitative Perceived job insecurity scale assessed the perceived threat participants felt to the continuity of their new job. This scale has three items (e.g. 'I feel uneasy about losing my job in the near future'), measured along a five-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). The alpha coefficient was .83 for the voluntarily redundant group and .84 for the involuntarily redundant group.
Organizational commitment was operationalized using the nine-item version of the scale developed by Porter, Steers, Mowday, and Boulian (1974) (e.g. 'I am willing to put in a great deal of effort beyond that normally required in order to help my employer be successful'). Items were measured using a five-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). In the present study, the alpha coefficient for voluntarily redundant participants was .89 and for the involuntarily redundant participants it was .84.
In order to assess perceived re-employment quality, respondents were asked to answer the following question, 'Taking into consideration factors such pay rate, seniority, skill use and employment terms, do you consider your new job to be of a higher, lower, or similar quality compared to your previous job'. Higher quality was scored as 3, similar as 2 and lower as 1.
At the baseline study and the 3-month retest participants were invited to make comments about their experience of job redundancy on a blank page at the end of the questionnaire. A selection of comments made by participants will be incorporated into the discussion section of this paper.
Each participant was given a code upon return of their baseline survey. This code was then written on the Study 2 surveys before they were posted out to each participant. All participants were recontacted via mail 3 months after Study 1. Participants were sent a questionnaire with their personalized identity code on it, a consent form and a plain language statement. Respondents were asked to return the questionnaire and consent form to the researchers' university address, using the reply-paid envelope provided. Upon return of the survey, each participant's code was used to match their Study 2 survey responses to their Study 1 survey responses in the SPSS data set.
Tests for response bias
A check of response bias within the voluntarily redundant group found no associations between responders and non-responders on gender, [chi square] (1) = 3.48, p = .062, occupational status, [chi square] (1) = 2.23, p = .095, education level, [chi square] (1) = 2.69, p = .101 or outplacement assistance, [chi square] (1) = 2.26, p = .103. In the involuntarily redundant group, there was no association between response/non-response with gender, [chi square] (1) = 0.01, p = .936, occupational status, [chi square] (1) = 0.48, p = .826, education level, [chi square] (1) = 1.27, p = .258 and outplacement assistance, [chi square] (1) = 0.14, p = .708.
The baseline scores for depression, job-search activity, self-efficacy, employment commitment, satisfaction with previous job, unemployment duration, age and advance notification were then compared between responders and non-responders within each group using multivariate analysis of variance (MANOVA). Within the voluntarily redundant group, the people who continued on in the study did not differ on these baseline measures to those who discontinued, F(2, 92) = 2.02, p = .139. There were also no differences between responders and non-responders on these baseline measures within the involuntarily redundant sample F(2, 106) = 126, p = .289.
Table 2 presents descriptive statistics for variables tested in Study 1 within the voluntarily and involuntarily redundant groups. In comparison to the involuntarily redundant sample, the voluntarily redundant group reported significantly less depression, significantly higher levels of job-search activity and significantly longer advance notification. Table 3 presents correlations among Study 1 variables. The upper quadrant of Table 3 shows the correlations found within the voluntarily redundant sample while the lower quadrant shows the correlations found within the involuntarily redundant sample.
Study I: Hypothesis testing
One-way, between-subjects analysis of covariance (ANCOVA) was used to assess Hypothesis 1. The independent variable was type of redundancy (voluntary vs. involuntary) and the dependent variable was depression. The assumptions of normality, linearity and homosedasticity were met. Based on results from the correlation analyses, generalized self-efficacy, employment commitment, advance notification and provision of outplacement were initially entered as covariates. However, when the ANCOVA was analysed, generalized self-efficacy was the only significant covariate F(1, 214) = 14.90, p = .000, [[eta].sup.2] = .07. Hence, employment commitment, advance notification and provision of outplacement were removed in order to create more degrees of freedom and the analysis was rerun. After adjusting for generalized serf-efficacy, the mean depression levels indicated that voluntary redundant participants ([bar.X] = 33.8, 95% CI = [+ or -] 1.1) were less depressed than involuntary redundant participants ([bar.X] = 36.8, 95% CI = [+ or -] 1.0) F(1, 214) = 12.64, p = .000, [[eta].sup.2] = .06.
One-way, between-subjects ANCOVA was also used to assess Hypothesis 2. The independent variable was type of redundancy (voluntary vs. involuntary) and the dependent variable was job search. The assumptions of normality, linearity and homosedasticity were met. When the ANCOVA was initially analysed, outplacement assistance was the only significant covariate F(1, 214) = 26.29, p = .000, [[eta].sup.2] = . 11. As such, generalized self-efficacy, employment commitment and advance notification were removed and the analysis was rerun. After adjusting for outplacement assistance, the mean job search levels indicated that voluntarily redundant participants ([bar.X] = 15.0, 95% CI = [+ or -] 1.1) were engaged in higher levels of job-search activity than the involuntarily redundant participants ([bar.X] = 10.6, 95% CI = [+ or -] 1.0) F(1, 214) = 33.43, p = .000, [[eta].sup.2] = .14.
Study 2: Hypothesis testing
Study 2 captured re-employed participants 3 months after the baseline study and was concerned with testing whether levels of depression, organizational commitment, perceived job insecurity and perceived quality of re-employment differed between those people who were made involuntarily redundant from their last job to those who volunteered for their own redundancy. Table 4 present the descriptive statistics and intercorrelations for Study 2 variables.
Hypothesis 3 was tested via two-way, within-subjects MANCOVA. Hypothesis 3 predicted that employees who experienced involuntary job redundancy from their previous job would maintain higher levels of depression in their new job compared with employees who volunteered for job redundancy. Given that generalized self-efficacy, employment commitment and advance notification were correlated with depression at Time 1, they were entered as covariates. However, as none of these variables were significant in the first rtm of the MANCOVA, they were removed from the analysis. The assumptions of normality, linearity and homosedasticity were met.
There was a significant main effect for type of redundancy where the involuntary group displayed higher levels of depression, F(1, 51) = 44.48, p = .000, [[eta].sup.2] = .46, as well as a significant main effect for employment status (unemployed to re-employed), F(1, 51) = 18.15, p = .000, [[eta].sup.2] = .26. More importantly for Hypothesis 3, there was a significant interaction between type of redundancy and employment status, showing that the effect of re-employment status upon depression was influenced by the way in which a person had lost their previous job, F(1, 51) = 9.43, p = .003, [[eta].sup.2] = .16.
Repeated measures t test found a significant change in depression over time for the voluntary redundancy group, t(26) = p = .001, [[eta].sup.2] = .16. Depression levels went down over time from Time 1 (mean = 32.9) to Time 2 (30.5), 95% CI of change (lower: 1.2; upper: 3.6). In contrast, repeated measures t test found no significant change in depression over time for the involuntary redundancy group, t(26) = 1.59, p = .12. These results support Hypothesis 3 as they show that those people who were made involuntarily redundant experienced a legacy effect in that depression stayed at high levels despite being re-employed. In contrast, depression was significantly lowered for the voluntarily redundant group.
An additional test of Hypothesis 3 was performed using ANCOVA to examine differential change in depression over time as a function of redundancy status. Redundancy status was the independent variable, depression at Time 1 was the covariate, and depression at Time 2 was the dependent variable. A main effect for redundancy status, F(1, 50) = 61.8, p < .001, and a main effect for depression at Time 1 was observed, F(1, 50) = 199.9, p < .001. Examination of parameter estimates made it clear that higher depression at Time 1 was strongly positively related to depression at Time 2 ([[eta].sup.2] = .80). Examination of the marginal means for Time 2 depression controlling for Time 1 depression were higher for involuntary (36.9, 95% CI = [+ or -] 0.7) than voluntary redundant (32.7, 95% CI = [+ or -] 0.7) individuals. Thus, individuals made involuntarily redundant became more depressed over time relative to the voluntary redundant individuals. The effect of treating depression as a covariate was to reduce the absolute difference between the two groups at Time 2 (difference without covariate = 8.6; difference after inclusion of covariate = 4.2).
A one-way, between-subjects MANOVA was used to investigate whether there were any differences in perceived job insecurity, organizational commitment and perceived quality of re-employment between re-employed participants who had previously experienced voluntarily or involuntarily job redundancy (Hypotheses 4, 5 and 7). Observed power was .89.
A global difference was found between the two groups, F(3, 49) = 9.51, p < .000, [[eta].sup.2] = .37. Univariate tests showed that voluntary redundant participants had higher levels of organizational commitment (95% CI = [+ or -] 1.6) than involuntary redundant participants (95% CI = [+ or -] 1.6) F(1, 52)= 12.76, p = .001, [[eta].sup.2] = .20. The voluntary redundant participants had lower levels of job insecurity (95% CI = [+ or -] 1.2) than involuntary redundant participants (95% CI = [+ or -] 1.3) F(1, 52) = 17.54, p = .001, [[eta].sup.2] = .26. The voluntary redundant participants had higher perceived quality of re-employment (95% CI = [+ or -] .30) than involuntary redundant participants (95% CI = [+ or -] .29) F(1, 52) = 4.88, p = .032, [[eta].sup.2] = .09. These results support Hypotheses 4, 5 and 7.
A mediational model (Baron & Kenny, 1986) was used to test Hypothesis 6 which predicted that the effect of type of job loss on organizational commitment would be mediated by job insecurity. This model was tested using the multiple criteria set out by Baron and Kenny. First, type of job loss did predict organizational commitment (involuntary predicting lower commitment), r = -.45, p < .01. Second, type of job loss did predict job insecurity (involuntary predicting greater insecurity), r = .51, p < .01. However, the third test failed because job insecurity did not predict organizational commitment when it was put into a regression that also included type of job loss, standardized [beta] = 0.04, p > 05. Finally, the regression weight for job loss on organizational commitment was not reduced when job insecurity was included in the regression, r = -.43, p < .01. The Sobel test showed that there was no significant partial mediation, Sobel Test = .28, p = .78.
The current paper is the first in the literature to separate job redundancy into voluntary and involuntary categories and then compare people in these two categories on depression and job search during unemployment as well as depression, perceived job insecurity, organizational commitment and perceived quality of re-employment once a new job is obtained. The results of this paper can be used to question the common assumption that job redundancy is universally experienced as an involuntary form of turnover. Instead, these studies show that some employees experience redundancy as a voluntary event. Moreover, the studies' findings suggest that voluntary and involuntary redundancy programmes may be associated with variations in employee reactions upon unemployment and re-employment.
The results of Study 1 show that, once unemployed, the people who volunteered for redundancy experienced lower levels of depression and engaged in more job-search activity than those who lost their job involuntarily. Indeed, type of redundancy accounted for 6% of the variance in depression and 14% of the variance in job search. This is an informative finding and suggests that the common practice of researchers in the field, to neglect the way in which a person has exited their last job, should be avoided. Importantly, this study indicates that depression and job search are influenced not only by the individual but also by the way in which the organization handles the job exit process. In particular, the results suggest that organizations may help to minimize depression of their redundant employees by offering voluntary redundancy options.
The current study proposes that people who lose their job involuntarily may experience aspects of learned helplessness, specifically the occurrence of response-outcome independence. It is proposed that helplessness experienced due to involuntary job redundancy may be transferred into feelings of passiveness in attempts to find new work. This phenomenon was reflected in the comments made by a number of involuntarily redundant participants, 'I didn't want to leave my last job, it was my boss who made the decision. Now it seems that recruiters and prospective employers make the decisions about my next job and my life' (Male, 32 years). 'My outplacement advisor keeps showing me ways to improve my resume. What is the point? I'll probably get the boot from the next job anyway' (Female, 28 years).
Study 2 responds to the calls of Feather (1990), King (2002) and Pugh et al. (2003) for a greater understanding of the re-employment experiences following job loss. The current results support the view of these authors in that, for victims of redundancy, the event of re-employment does not wipe the slate clean. This was especially the case for people who experienced involuntary job redundancy as they reported high levels of depression, high perceived job insecurity, low organizational commitment and low perceived job quality upon re-employment.
While authors such as Kinicki et al. (2000), King (2002), Pugh et al. (2003) and Fineman (1983) all suggest that job loss may be associated with a negative 'scarring' effect upon re-employment, they did not consider whether type of job loss plays a role. The results of this study suggest that type of redundancy is associated with perceived job insecurity, organizational commitment and depression upon re-employment. Specifically, the results of Study 2 show that, once re-employed, voluntarily redundant participants reported a decline in depression while involuntarily redundant employees experienced no change. These results suggest that type of job redundancy provides an important qualification to the legacy theory. The presence of a legacy effect, at least in terms of depression, was influenced in part by the way in which the employee lost, or left, their previous job. Only employees who had experienced involuntary redundancy showed a scarring effect in relation to depression. These results suggest that researchers and managers need to consider the way in which a person exited their previous job in order to understand his/her full reaction upon re-employment.
In addition to group differences in depression upon re-employment, involuntarily redundant employees reported significantly higher levels of perceived job insecurity in their new job compared with those who left their last job voluntarily. These results are in line with Hellgren et al.'s (1999) arguments that '(1) the experience of perceived job insecurity is individual and subjective and (2) that it is based on a change, against the wishes of the individual, from a secure to a non-secure position' (p. 13). These results are also in line with Roskies and Louis-Guerin's (1990) claim that unpredictability associated with involuntary redundancy leads people to fear that such job loss will reoccur.
Involuntarily redundant employees were also less committed to their new organization (supporting Hypothesis 5). However, Hypothesis 6 was refuted, as the lower levels of organizational commitment reported by the involuntary group were not due to the fact that this group felt more insecure in their new jobs. The qualitative data provide some information about why it may be that the involuntarily redundant employees were less committed to their new organization. A number of participants suggested that low organizational commitment may have arisen from feelings of misplaced trust in their previous employer. A 42-year-old male stated 'after what happened to me in my last job, my work is no longer my focus. I am here simply to earn money so I can enjoy the other aspects of my life. I leave at 5.30 and I won't be doing anything extra at work'.
The significant correlation between organizational commitment and perceived quality of re-employment may also provide a partial explanation for why the involuntarily redundant employees reported lower levels of organizational commitment than the voluntarily redundant employees. It could be that the involuntarily redundant group report lower levels of organizational commitment as they see themselves to have moved into lower quality re-employment than the voluntarily redundant group. This finding also supports Hypothesis 7 that the employees who experienced involuntary job redundancy would report lower perceived quality of re-employment than employees who volunteered for their redundancy.
It was suggested in the introduction that feelings of learned helplessness, more likely to be experienced by the involuntarily redundant group, may lead people to accept lower quality re-employment. The current results may be used to support such an interpretation. However, as there was no objective measure of re-employment quality taken in the current study it cannot be determined whether learned helplessness leads to poorer quality re-employment or whether the depression associated with learned helpless means that the involuntarily redundant group is more likely to perceive their reemployment quality as poor irrespective of the objective quality of the job.
Limitations and future research
Following the calls of Feather (1990), Rodriguez (1997) and Winefield (2002), this paper applies learned helplessness as a theoretical lens to predict and explain the current findings. More particularly, the phenomenon of 'response-outcome independence' (Peterson & Seligman, 1983, p. 103) is used to look at two aspects of learned helplessness that may be evident during unemployment: (1) depression and (2) passivity.
It must be noted that perceived control, a central construct in the learned helplessness theory, was not directly measured in the current study. This is because there are no existing measures to capture the degree of situational control employees feel they have over their job exit. Moreover, because the existing research has not adequately considered the possibility that job redundancy can be a voluntary event there are no existing measures to capture degree of control in this specific type of job exit. In the current study, the researcher has assumed that voluntary redundancies are associated with a higher degree of control than involuntary redundancies. However, this assumption requires validation through empirical testing. Future researchers who chose to test this assumption directly may benefit from designing a scale that assesses the degree of control, or lack of it, that employees report to have over their job exits.
Given that participants were tested only after they had lost their jobs, the study was unable to test for the possible effects that depression, organizational commitment and job insecurity may have had upon type of redundancy experienced in the first place. For example, it could be that depressed, insecure and uncommitted employees were more likely to be targeted for involuntary redundancy. However, the qualitative comments provided by study participants suggest that it was involuntary redundancy that led to a sense of helplessness (as evidenced by high depression, high insecurity, low job search and low commitment) rather than the other way around. Moreover, by controlling for a number of important covariates, this study has helped to minimize the possibility that the current results reflect pre-existing group differences (see Winefield (1997) for a discussion of the evidence pertaining to the 'selection vs. exposure' debate in the unemployment literature).
Caution is also required when drawing conclusions from the re-employed samples too, owing to their small size. Concern over the small sample size can be partly allayed, however, considering that the re-employment rate was similar to that reported in other unemployment research (Gowan et al., 1999; Wanberg et al., 1996). Also, the similarity in the demographic profiles of the current re-employed samples compared with their equivalent groups in the Australian labour force suggests that they are adequately representative of the larger re-employed population (Australian Bureau of Statistics, 2003).
As these employees were surveyed within 3 months of losing their last job, conclusions about the longer-lasting scarring effects of previous redundancy cannot be drawn. It may be that, over time, the trauma of involuntary job redundancy wears off. Future research is required to study the longer-term effects of job redundancy upon depression, perceived job insecurity and organizational commitment in employees after they settle into their new jobs (King, 2002; Pugh et al., 2003).
Implications and conclusions
An implication that may unfortunately be drawn from this study is that organizations would be best to avoid hiring employees who have experienced involuntary redundancy. It would be unwise for companies to do so as this subgroup of employees now make up a sizeable proportion of the labour market. Indeed, King has argued that in an increasing number of cases 'one company's layoff victim becomes another companies new employee' (2002; p. 35). Rather than avoid this issue, companies would do best to address it. Pugh et al. (2003) suggest that managers should engage in trust-building exercises to gain the confidence of new employees who have previously bad negative lay-off experiences.
If, as indicated in this study, learned helplessness is a factor in explaining psychological reaction to unemployment, then Seligman's (1991) learned optimism guidelines may be helpful in reducing depression and encouraging job-search activity. Counselling techniques such as distraction, evidence seeking, alternative generation and disputation used in a careers context may help people to diminish the depression associated with involuntary job redundancy.
The insecure employment conditions that many people now work within have led to substantial career discontinuity in the forms of underemployment and unemployment. In fact, labour market statistics in most western countries would suggest that unemployment is a common, albeit largely unwanted, aspect of the modern career. With this in mind, researchers and practitioners have a continued obligation to improve the understanding of reactions to unemployment and facilitators of quality re-employment. The results of this study suggest that type of job redundancy may be one useful explanatory factor in this quest.
Received 8 September 2004; revised version received 2 February 2006
Abramson, L., Seligman, M., & Teasdale, J. (1978). Learned helplessness in humans: Critique and reformulation. Journal of Abnormal Psychology, 87, 49-74.
Australian Bureau of Statistics. (2003). Labour force Australia (Catalogue No. 6202.0). Canberra: Australian Bureau of Statistics.
Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51, 1173-1182.
Brockner, J., Grover, S., Reed, T., & DeWitt, R. (1992). Layoffs, perceived job insecurity, and survivors' work effort: Evidence of an inverted-U relationship. Academy of Management Journal, 35, 413-425.
Brockner, J., Wiesenfeld, B., Reed, T., & Grover, S. (1993). Interactive effect of job content and context on the reactions of layoff survivors. Journal of Personality and Social Psychology, 64, 187-197.
Bureau of Labor Statistics. (2004). Extended mass layoffs in the first quarter of 2004 (USDL 04895). Washington: United States Department of Labor News.
Burke, R. (1986). Re-employment on a poorer job after a plant closing. Psychological Reports, 58, 559-570.
Campbell, I., & Webber, M. (1996). Redundancy and labour market flows in Australia. Economic and Labour Relations Review, 7, 89-119.
Campion, M. (1991). Meaning and measurement of turnover: A comparison of alternative measures and recommendations for research. Journal of Applied Psychology, 762, 100-212.
Caplan, R., Vinokur, D., Price, R., & Van Ryn, M. (1989). Job seeking, reemployment, and mental health: A randomized field experiment in coping with job loss. Journal of Applied Psychology, 74, 759-769.
Creed, P., & Reynolds, J. (2001). Economic deprivation, experiential deprivation and social loneliness in unemployed and employed youth. Journal of Community and Applied Psychology, 11, 167-178.
Crespo, S., Boisjoly, J., & Bernard, P. (1998). What do people do when they are laid off? Income and labour dynamics working paper, Catalogue 98-08. Ottowa: Statistics Canada product number 75F002M.
Department of Employment, Education, Training and Youth Affairs. (1997). ASCO: Australian Standard Classification of Occupations Second Edition. Canberra: Author.
DeWitt, R., Trevino, L., & Mollica, K. (1998). The influence of eligibility on employees' reactions to voluntary workforce reductions. Journal of Management, 24, 593-613.
Feather, N. (1990). The psychological impact of unemployment. New York: Springer-Verlag.
Feldman, D., & Leana, C. (2000). Whatever happened to laid-off executives? A study of reemployment challenges after downsizing. Organisational Dynamics, 29, 64-75.
Fineman, S. (1983). Counselling the unemployed: Help and helplessness. British Journal of Guidance and Counselling, 1, 1-9.
Fryer, D., & Fagan, R. (2003). Toward a critical community psychological perspective on unemployment and mental health research. American Journal of Community Psychology, 32, 89-96.
Gowan, M., Riordan, C., & Gatewood, R. (1999). Test of a model of coping with involuntary job loss following a company closing. Journal of Applied Psychology, 84, 75-86.
Hanisch, K. (1999). Job loss and unemployment research from 1994 to 1998: A review and recommendations for research and intervention. Journal of Vocational Behaviour, 55, 188-220.
Hardy, C. (1986). Strategies for redundancy: Reconciling individual and organisational needs. Canadian Journal of Administrative Science, 3, 275-289.
Hartley, J., Jacobsen, D., Klandermans, B., & Van Vuuren, T (1991). Perceived job insecurity: Coping with jobs at risk. London: Sage.
Hellgren, J., Sverke, M., & Isaksson, K. (1999). A two-dimensional approach to perceived job insecurity: Consequences for employee attitudes and well-being. European Journal of Work and Organisational Psychology, 8, 179-195.
Huang, T. (2003). Unemployment and family behavior in Taiwan. Journal of Family and Economic Issues, 24, 27-48.
Hu, L., & Bentler, P. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modelling, 6, 1-55.
Iverson, R., & Pullman, J. (2000). Determinants of voluntary turnover and layoffs in an environment of repeated downsizing following a merger: An event history analysis. Journal of Management, 26, 977-1003.
Jahoda, M. (1988). Economic recession and mental health: Some conceptual issues. Journal of Social Issues, 44, 13-23.
Judge, T., Erez, A., Bono, J., & Thorenson, C. (2002). Are measures of self-esteem, neuroticism, locus of control and generalized self efficacy indicators of a common core construct? Journal of Personality and Social Psychology, 83, 693-710.
King, J. (2002). Re-employed victims and multiple exposure: The continual accumulating impact of layoffs. Journal of Applied Management and Entrepreneurship, 7, 25-42.
Kinicki, A. J., & Latack, J. C. (1990). Explication of the construct of coping with involuntary job loss. Journal of Vocational Behavior, 36, 339-360.
Kinicki, A. J., Prussia, G. R., & McKee-Ryan, F. M. (2000). A panel study of coping with involuntary job loss. Academy of Management Journal, 43, 90-100.
Latack, J. C., & Dozier, J. (1986). After the axe falls: Job loss as a career transition. Academy of Management Review, 11, 377-385.
Latack, J. C., Kinicki, A.J., & Prussia, G. E. (1995). An integrative process model of coping with job loss. Academy of Management Review, 20, 311-342.
Leana, C. R., & Feldman, D. C. (1992). Coping with job loss. New York: Lexington Books.
Leana, C. R., & Feldman, D. C. (1995). Finding new jobs after a plant closing: Antecedents and outcomes of the occurrence and quality of re-employment. Human Relations, 48, 1381-1401.
Macken, J., O'Grady, P., & Sappideen, C. (1997). The law of employment (4th ed.). North Ryde: Commonwealth Government Printer.
Mallinckrodt, B. (1990). Satisfaction with a new job after unemployment: Consequences of job loss for older professionals. Journal of Counseling Psychology, 37, 149-152.
Marsh, H., Balla, J., & McDonald, R. (1988). Goodness-of-fit indexes in confirmatory factor analysis: The effect of sample size. Psychological Bulletin, 103(3), 391-410.
McNair, D., Lorr, M., & Droppelman, L. (1981). Manual: Profile of mood states. San Diego, CA: Educational and Industrial Testing Service.
Miller, M. V., & Hoppe, S. K. (1994). Attributions for job termination and psychological distress. Human Relations, 47, 307-327.
Morehead, A., Steele, M., Alexander, M., Stephen, K., & Duffin, L. (1997). Changes at work: The 1995 Australian workplace industrial relations survey. Melbourne: Addison Wesley Longman.
Peterson, C., & Seligman, M. (1983). Learned helplessness and victimization. Journal of Social Issues, 2, 103-116.
Porter, L. M., Steers, R. M., Mowday, R. T., & Boulian, P. V. (1974). Organisational commitment, job satisfaction, and turnover among psychiatric technicians. Journal of Applied Psychology, 59, 603-609.
Pugh, D., Skarlicki, D., & Passell, B. (2003). After the fall: Layoff victims' trust and cynicism in reemployment. Journal of Occupational and Organisational Psychology, 76, 201-212.
Rodriguez, Y. (1997). Learned helplessness or expectancy-value? A psychological model for describing the experiences of different categories of unemployed people. Journal of Adolescence, 20, 321-332.
Romeyn, J. (1992). Flexible working time: Part time and casual employment. Canberra: Australian Government Publishing Service.
Rosenblatt, Z., & Ruvio, A. (1996). A test of a multidimensional model of perceived job insecurity: The case of Israeli teachers. Journal of Organisational Behavior, 17, 587-605.
Roskies, E., & Louis-Guerin, C. (1990). Perceived job insecurity in managers: Antecedents and consequences. Journal of Organisational Behaviour, 11, 345-359.
Rousseau, D. (2004). Psychological contracts in the workplace: Ties that motivate. Academy of Management of Executive, 18, 120-129.
Seligman, M. (1975). Helplessness: On depression, development, and death. San Francisco: W.H. Freeman.
Seligman, M. (1991). Learned optimism. New York: Pocket Books.
Schaufeli, W., & Vanyperen, N. (1993). Success and failure in the labour market. Journal of Organisational Behavior, 14, 559-572.
Sherer, M., Maddux, J., Mercandante, B., Prentice-Dunn, S., Jacobs, B., & Rogers, R. (1982). The self-efficacy scale: Construction and validation. Psychological Reports, 51, 663-671.
Stafford, E., Jackson, P., & Banks, M. (1980). Employment, work involvement and mental health in less qualified young people. Journal of Occupational Psychology, 53, 291-304.
Tabachnick, B., & Fidell, L. (1996). Using multivariate statistics (3rd ed.). New York: Harper Collins. Wanberg, C. (1995). A longitudinal study of the effects of unemployment and quality of reemployment. Journal of Vocational Behavior, 46, 40-54.
Wanberg, C. (1997). Antecedents and outcomes of coping behaviors among unemployed and re-employed individuals. Journal of Applied Psychology, 82, 731-744.
Wanberg, C., Hough, L., & Song, Z. (2002). Predictive validity of a multidisciplinary model of re-employment success. Journal of Applied Psychology, 87, 1100-1120.
Wanberg, C., Watt, J. D., & Rumsey, D. J. (1996). Individuals without jobs: An empirical study of job-seeking behavior and re-employment. Journal of Applied Psychology, 81, 76-87.
Waters, L., & Moore, K. (2002). Predicting self-esteem during unemployment: The effect of gender, financial deprivation, alternate roles and social support. Journal of Employment Counseling, 39, 171-189.
Weir, G. (2003). Job separations. Labour Market Trends, 111(3), 121-132.
Winefield, A. (1997). Introduction to the psychological effects of youth unemployment: International perspectives. Journal of Adolescence, 20, 237-241.
Winefield, A. (2002). Unemployment, underemployment, occupational stress and psychological well-being. Australian Journal of Management, 27, 137-151.
Winefield, A., Tiggemann, M., & Winefield, H. (1992). Unemployment distress, reasons for job loss and causal attributions for unemployment in young people. Journal of Occupational and Organisational Psychology, 65, 213-218.
(1) While perceived job insecurity and organizational commitment were also tested via multi-item scales the samples sizes in Study 2 were too small to run CFA.
Lea Waters *
University of Melbourne, Australia
* Correspondence should be addressed to Lea Waters, Department of Management, The University of Melbourne, Parkville 3010, Australia (e-mail: I.firstname.lastname@example.org).
Table 1. Fit statistics for the one-factor, two-factor and four-factor models Model type Model Chi-squared df name One-factor model M1 1278.9 54 Parcelled 2 factor, M2 637.9 53 correlated, depression one factor, other three variables one factor Parcelled 4 factor, M3 125.3 48 correlated Model type p RMSEA (90%CI in brackets) One-factor model <.001 .325 (.310; .340) Parcelled 2 factor, <.001 .227 (.211; .242) correlated, depression one factor, other three variables one factor Parcelled 4 factor, <.001 .087 (.068; .105) correlated Model type SRMR CFI TLI One-factor model .266 .38 .24 Parcelled 2 factor, .202 .63 .70 correlated, depression one factor, other three variables one factor Parcelled 4 factor, .048 .96 .95 correlated Table 2. Means and standard deviations for voluntarily redundant employees and involuntarily redundant employees Voluntarily Involuntarily redundant redundant Variable (N = 102) (N = 114) Depression Mean 34.23 36.68 SD 6.26 4.96 Job search Mean 14.99 10.57 SD 5.72 6.23 Generalized self-efficacy Mean 15.62 15.22 SD 2.95 3.54 Employment commitment Mean 34.14 33.95 SD 8.79 8.25 Retrospective job Mean 4.25 4.24 satisfaction SD 1.63 1.73 Unemployment duration Mean 12.70 12.40 (weeks) SD 2.30 2.70 Advance notification Mean 26.31 5.98 for lay-off (days) SD 6.93 5.03 Variable 95% Cl F(2,213) Depression Mean [+ or -] 1.1 10.01 SD Job search Mean [+ or -] 1.7 32.74 SD Generalized self-efficacy Mean [+ or -] 0.6 .91 SD Employment commitment Mean [+ or -] 1.7 .03 SD Retrospective job Mean [+ or -] 0.3 .03 satisfaction SD Unemployment duration Mean [+ or -] 4.9 .10 (weeks) SD Advance notification Mean [+ or -] 1.2 65.62 for lay-off (days) SD Variable p [[eta].sup.2] Depression Mean .002 .045 SD Job search Mean .000 .134 SD Generalized self-efficacy Mean .341 .004 SD Employment commitment Mean .870 .000 SD Retrospective job Mean .855 .000 satisfaction SD Unemployment duration Mean .749 .000 (weeks) SD Advance notification Mean .000 .744 for lay-off (days) SD Table 3. Intercorrelations between study variables for voluntarily redundant employees and involuntarily redundant employees Variables 1 2 3 1. Depression 1.00# -.19 .33 ** 2. Job search -.27 ** 1.00# -.02 3. Generalized self-efficacy .20 * .05 1.00# 4. Employment commitment .20 * .03 .67 ** 5. Retrospective job -.15 -.09 -.17 satisfaction 6. Unemployment duration .07 -.07 -.02 (weeks) 7. Gender -.05 .01 -.13 8. Age -.02 -.17 -.18 9. Occupational status .02 .01 .30 ** (0 = blue collar, 1 = white collar) 10. Educational level .03 -.07 -.12 (0 = secondary, 1 = tertiary) 11. Outplacement provided .07 .36 ** .14 (0 = no, 1 = yes) 12. Advance notice (days) -.09 .19 * .05 Variables 4 5 6 1. Depression .21 * -.02 -.11 2. Job search -.05 -.06 -.13 3. Generalized self-efficacy .45 ** -.02 .03 4. Employment commitment 1.00# -.12 .13 5. Retrospective job .05 1.00# .03 satisfaction 6. Unemployment duration .03 .15 1.00# (weeks) 7. Gender -.15 -.06 .05 8. Age -.20 * .08 .27 ** 9. Occupational status .35 * .12 -.15 (0 = blue collar, 1 = white collar) 10. Educational level -.22 * .07 .09 (0 = secondary, 1 = tertiary) 11. Outplacement provided .21 .08 .06 (0 = no, 1 = yes) 12. Advance notice (days) .04 -.08 .21 * Variables 7 8 9 1. Depression .08 .01 -.12 2. Job search -.06 -.03 .12 3. Generalized self-efficacy .03 .10 .15 4. Employment commitment .15 .15 -.05 5. Retrospective job -.11 .18 .01 satisfaction 6. Unemployment duration -.10 .11 .03 (weeks) 7. Gender 1.00# -.16 -.22 * 8. Age -.08 1.00# -.04 9. Occupational status -.41 ** -.12 1.00# (0 = blue collar, 1 = white collar) 10. Educational level .27 ** .02 .44 ** (0 = secondary, 1 = tertiary) 11. Outplacement provided -.04 -.09 .14 (0 = no, 1 = yes) 12. Advance notice (days) -.09 .25 * -.09 Variables 10 11 12 1. Depression -.03 -.06 -.20 * 2. Job search -.01 .33 ** .08 3. Generalized self-efficacy -.02 .11 .12 4. Employment commitment .02 -.16 .08 5. Retrospective job .08 .03 -.02 satisfaction 6. Unemployment duration .06 .05 .26 ** (weeks) 7. Gender .11 .12 -.03 8. Age -.05 -.03 -.05 9. Occupational status -.21 * .01 .02 (0 = blue collar, 1 = white collar) 10. Educational level 1.00# .10 .02 (0 = secondary, 1 = tertiary) 11. Outplacement provided -.08 1.00# .02 (0 = no, 1 = yes) 12. Advance notice (days) .02 -.12 1.00# Note. Values above the bold line are correlations from the voluntarily redundant sample; N = 102. Values below the bold line are correlations from the involuntarily redundant sample; N = 114. * p < .05; ** p < .01. Values below the bold line are correlations from the involuntarily redundant sample indicated by #. Table 4. Intercorrelations between study variables for voluntarily redundant and involuntarily redundant groups at 6-month retest Variables Voluntarily 1 redundant (N = 28) 1. Depression Mean 32.88 1.00# SD 6.04 2. Organisational commitment Mean 29.37 .51 ** SD 5.1 3. Job insecurity Mean 7.61 .01 SD 2.8 4. Perceived re-employment Mean 2.04 .49 ** quality SD 0.85 Variables 2 3 4 1. Depression .68 ** .10 .76 ** 2. Organisational commitment 1.00# -0.29 .73 ** 3. Job insecurity -.19 1.00# -.01 4. Perceived re-employment .76 ** .01 1.00# quality Variables Involuntarily redundant (N = 26) 1. Depression Mean 39.09 SD 3.24 2. Organisational commitment Mean 25.23 SD 3.03 3. Job insecurity Mean 11.29 SD 3.53 4. Perceived re-employment Mean 1.58 quality SD 0.64 Note. Values above the bold line are correlations from the voluntarily redundant sample; N = 102. Values below the bold line are correlations from the voluntarily redundant sample involuntarily redundant sample; N = 114. * p < .05; ** p <.01. Values below the bold line are correlations from the voluntarily redundant sample involuntarily redundant sample indicated by #.
|Printer friendly Cite/link Email Feedback|
|Publication:||Journal of Occupational and Organizational Psychology|
|Article Type:||Author abstract|
|Date:||Jun 1, 2007|
|Previous Article:||On the relations among work value orientations, psychological need satisfaction and job outcomes: a self-determination theory approach.|
|Next Article:||A typology of the part-time workforce: differences on job attitudes and turnover.|