Affective Job Insecurity

Transcription

1 Int. Studies of Mgt. & Org., vol. 40, no. 1, Spring 2010, pp M.E. Sharpe, Inc. All rights reserved. ISSN / 2010 $ DOI /IMO Guo-Hua Huang, Cynthia Lee, Susan Ashford, Zhenxiong Chen, and Xiaopeng Ren Affective Job Insecurity A Mediator of Cognitive Job Insecurity and Employee Outcomes Relationships Abstract: Researchers who work on job insecurity (JI) have largely ignored the differences between cognitive job insecurity and affective job insecurity. In this study, we argue that it is conceptually important to study affective JI and cognitive JI as distinct constructs. Based on the conceptualization of stress and affective event theory, we propose that affective JI is an outcome of cognitive JI and that affective JI partially mediates the relationship between cognitive JI and employee outcomes. In two samples of working people, we found that affective JI partially explains the effect of cognitive JI on employees job Guo-hua Huang is an assistant professor in the Department of Management, Hong Kong Baptist University, Kowloon, Hong Kong; tel.: (852) ; fax: (852) ; mnhgh@hkbu.edu.hk. Cynthia Lee is a chaired professor of management, Hong Kong Polytechnic University, and professor of management and organizational development, College of Business Administration, Northeastern University, Boston, MA 02115; tel.: (617) ; fax: (617) ; c.lee@neu.edu. Susan Ashford is the associate dean for leadership programming and the executive MBA program, Michael and Susan Jandernoa Professor of Management and Organizations, Ross School of Business, University of Michigan, Ann Arbor, MI ; tel.: (734) ; fax: (734) ; sja@bus.umich.edu. Zhenxiong Chen is a professor of management, School of Management, Marketing, and International Business, Australian National University, Canberra, ACT 0200, Australia; tel.: ; fax: ; george.chen@anu.edu.au. Xiaopeng Ren is an associate professor at the Center for Social and Economic Behavior, Institute of Psychology, Chinese Academy of Sciences, Beijing, China; tel.: ; fax: ; renxp@ psych.ac.cn. The work described in this article was supported by Cynthia Lee s Walsh Professorship, College of Business Administration, Northeastern University. 20

2 Affective Job Insecurity 21 satisfaction, organizational commitment, and somatic well-being. Implications and recommendations for future research are discussed. The past three decades have seen a dramatic change in the nature of work (Ashford, George, and Blatt 2007; Schabracq and Cooper 2000) and in an organization s psychological contract with its workers (Rousseau 1995). As organizations restructure and right-size, not only in industrialized countries but also in rapidly developing economies such as China and India, an increasingly occurring phenomenon is job insecurity (JI) (Klandermans and van Vuuren 1999; Lee, Bobko, and Chen 2006). Job insecurity is important as studies have documented its significant impact on important individual and organizational outcomes, such as employee psychological well-being, job satisfaction, and job performance (Ashford, Lee, and Bobko 1989; Sverke, Hellgren, and Näswall 2002). The eight consequences reviewed in Sverke, Hellgren, and Näswall s (2002) recent meta-analysis indicate that JI has detrimental consequences for both the individual and the organization. Our study adds to the conceptualization of JI. Most researchers agree that JI is a subjective phenomenon (Ashford, Lee, and Bobko 1989; Greenhalgh and Rosenblatt 1984; Hartley et al. 1991; Klandermans and van Vuuren 1999; Sverke, Hellgren, and Näswall 2002). Job insecurity is thought to reflect the degree to which employees perceive their jobs, or important features of their jobs, to be threatened and to which they perceive themselves to be powerless to do anything about it (Ashford, Lee, and Bobko 1989; Greenhalgh and Rosenblatt 1984). Job insecurity reflects the subjective anticipation of an involuntary event. As such, this anticipation has two components: one that is cognitive (beliefs) and one that is affective (emotional state). However, the literature to date has treated JI as an essentially cognitive phenomenon. Predominant measures tap employees perceptions of various aspects of their work settings (see Table 1) or subjective probabilities of various events affecting the job or its various features (Ashford, Lee, and Bobko 1989). Weiss and Cropanzano s (1996) Affective Event Theory (AET), however, emphasizes the unique role of affective reactions in the workplace. They offered their theory as a counterbalance to theories that exclusively focus on judgment or cognitive processes. According to AET, the experience of a positive or negative work event (e.g., cognitive JI) can elicit affective reactions (e.g., concern or worry about job insecurity) that contribute to the formation of work attitudes and behaviors (Mignonac and Herrbach 2004; Rupp and Spencer 2006). Hartley et al. (1991) proposed a model in which cognitive JI as stressor predicts emotional reactions. Given the theoretical importance of affect in a domain like job insecurity and given the scant empirical attention, we examine affective JI reaction as an outcome of cognitive JI and also as a

3 22 Huang (China) et al. mediator between cognitive JI and work-related outcomes. We contribute to the literature by studying the JI phenomenon in China. Most existing studies have been conducted in well-developed North American and European countries. Considering the increasing magnitude of JI in China due to the significant economic reforms during the past two decades, studying JI in Chinese organizations may be necessary and interesting (Wong et al. 2005). In the following sections, we first review the literature on JI, focusing on the nature of the construct. Based on the review, we propose three hypotheses regarding the mediating role of affective JI in the relationship between cognitive JI and work-related outcomes. We then report the results of a study we conducted in China, including two samples from four Chinese organizations. The contributions, implications, and limitations of our study are discussed at the end. Cognitive and affective job insecurity Researchers have defined JI as perceived powerlessness to maintain desired continuity in a threatened job situation (Greenhalgh and Rosenblatt 1984, 438), concern about the future permanence of the job (Van Vuuren and Klandermans 1990, 134), and expectations about continuity in a job situation (Davy, Kinicki, and Scheck 1997, 323). An important distinction among the various conceptualizations of JI is the relative emphasis on cognitive versus affective elements of JI (König and Staufenbiel 2006; Reisel and Banai 2002). Portrayals that emphasize cognitive JI refer to the perception of the likelihood of negative changes to one s job, (e.g., losing the job or losing attractive job features) while affective JI captures the affective elements of the JI experience, such as being concerned, worried, or anxious about losing the job or particular job features. We summarize existing approaches to the JI construct in Table 1. As shown in Table 1, studies have focused heavily on a cognitive over an affective conceptualization of JI. Few studies have employed both conceptualizations and theory development. The fact that researchers have largely ignored the distinction between cognitive and affective JI has limited the conceptual development of JI (König and Staufenbiel 2006). Fortunately, a few studies have begun to examine this distinction (e.g., Borg and Elizur 1992; König and Staufenbiel 2006; Reisel and Banai 2002). For instance, Borg and Elizur (1992) differentiated two dimensions of JI: JI as cognitive appraisal and JI as worry (affective) and found that the two dimensions were correlated at a low level. Adopting their measure in German organizations, König and Staufenbiel (2006) found that the two dimensions correlated differently with employee job attitudes (e.g., job satisfaction, commitment, or justice perceptions) and behaviors (e.g., performance). The correlations of affective JI are consistent with the general idea that people

4 Affective Job Insecurity 23 Table 1 Summary of different approaches to the study of job insecurity Author(s) Date (and sample) Conceptualization Operationalization Caplan et al (USA) Cognitive, global Additive (4 items) Greenhalgh and Rosenblatt 1984 (conceptual paper) Cognitive, multidimensional Multiplicative (conceptual) Johnson, Messe, and Crano 1984 (USA) Affective, global Additive (7 items) Ashford, Lee, and Bobko 1989 (USA) Cognitive, multidimensional Multiplicative (5 subscales, 60 items) Hartley et al (Israeli sample) Cognitive, global Additive (3 items) Hartley et al (Dutch and UK samples) Cognitive and affective, global Additive (3 items) Borg and Elizur 1992 (multinational sample from Europe) Cognitive and affective, global Additive (4 items for JI fear [cognitive] and 3 items for JI worry [affective]) Hellgren, Sverke, and Isaksson 1999 (Sweden) Unclear a Additive (3 items for total job and 4 items for job features) Probst 2003 (USA) Cognitive and affective Additive (18 items for JI perception and 20 items for JI satisfaction) Note : Studies are in order of the date of publication of the original scales. a Although the authors claim they followed Ashford, Lee, and Boblo s (1989) conceptualization, the wording of the items in their scale seems to involve both cognitive (e.g., I feel that [the organization] can provide me with a stimulating job content in the near future ) and emotional (e.g., I am worried about having to leave my job before I would like to ).

5 24 Huang (China) et al. that care about their job report high affective JI. Further, the correlations of cognitive JI are consistent with earlier research that has shown harmful effect of cognitive JI on performance, job attitudes, and health (cf. Sverke, Hellgren, and Näswall 2002). Sampling in Canada, Ito and Brotheridge (2007) studied the relationships between cognitive and affective JI. Using two items from Caplan et al. s (1975) scale to measure cognitive JI and Borg and Elizer s (1992) two-item scale to measure affective JI, they have found that cognitive JI and affective JI were associated with each other but differed in their associations with predictors and consequences. This literature stream, however, has its limitations. First, no consensus has been reached on the operationalization of cognitive and affective JI. As shown in Table 1, despite prevailing theory (e.g., Greenhalgh and Rosenblatt 1984), some researchers use global measures reflecting career and job loss dimensions of cognitive JI as in the Caplan s measure. Others use a multidimensional measure reflecting the job loss and job features dimensions of cognitive JI. Second, there is no consistency in the treatment of these two JI components. For instance, some studies combined cognitive and affective JI in a single construct and aggregated the two into one measure (cf. Reisel and Banai 2002). Third, with regard to the effect of affective JI on job attitudes, findings have been mixed. Studies have documented both negative (e.g., Ito and Brotheridge 2007) and positive (e.g., König and Staufenbiel 2006) effects. For example, while Ito and Brotheridge (2007) found affective JI to be negatively related to affective organizational commitment and job satisfaction but have no relationship with turnover intention, König and Staufenbiel (2006) reported positive associations between affective JI and organizational commitment and job satisfaction as well as a negative association between affective JI and turnover intent. Given these issues, more studies are needed to examine the conceptualization and impact of cognitive and affective JI. The most widely accepted definition of JI in the literature is perceived powerlessness to maintain desired continuity in a threatened job situation (Greenhalgh and Rosenblatt 1984, 438). This definition s content domain covers not only loss of the whole job, but also loss of job features. Moreover, it involves not only the threat to one s job and features, but also the sense of powerlessness to counteract the threat. Based on this conceptualization, Ashford, Lee, and Bobko (1989) developed a five-component job insecurity scale (JIS). Researchers using JIS have found evidence for the impact of JI on important individual and organizational outcomes (Lee, Bobko, and Chen 2006). However, this JIS captures only the cognitive aspect of JI. It comprises a cognitive appraisal of the importance of various job features (or the job itself) and the perceived probability of the loss of those features (the job itself) multiplied by perceptions of powerlessness. This theory and measure implicitly portray

6 Affective Job Insecurity 25 individuals as probability calculators. This calculative, cognitive portrayal seems quite distinct from the emotional experience of being worried or emotionally concerned about a change to or the loss of one s job. Most existing studies either do not recognize this distinction or, if they do, problematically combine both cognitive and affective JI into one construct. In contrast with these previous studies, we examine the cognitive and affective aspects of JI as two distinct constructs in examining the JI phenomenon. Two distinct measures are necessary because one s cognitive judgment of JI, although likely to be related to, is not the same as one s emotional reaction to that judgment. For example, two employees occupying identical positions may perceive identical levels of JI. However, they may not have identical affective reactions to these perceptions, depending on a number of variables ranging from their financial dependence on the job to the characteristics of their personalities, such as their dispositional optimism. Conceptually, this distinction is similar to that made in two other literatures, psychological contracts and the stress literature. Although these literatures recognize the importance of affect, the JI literature has largely ignored it. This distinction we are drawing is analogous to the conceptual development of the psychological contract model in the 1990s (Morrison and Robinson 1997; Rousseau 1995). Psychological contract breach refers to the employee s perception of the extent to which the organization has failed to fulfill its promises or obligations (Robinson and Rousseau 1994). In the psychological contract literature, researchers have used the terms violation and breach interchangeably. However, Morrison and Robinson (1997) distinguished between the two constructs by suggesting that breach is the cognitive evaluation that one s organization has failed to fulfill its obligations, whereas violation is the affective state that may follow from the breach. We borrow the distinction between contract breach and violation to examine cognitive and affective JI. We acknowledge that the two conceptualizations, namely, psychological contract breach and violation and cognitive and affective JI, are not completely analogous. While contract breach comes from the employer not fulfilling its obligations, cognitive JI can come from the employee as well through, for example, poor performance. As such, we examine affective JI as an outcome of cognitive JI. We believe that this conceptual clarification will make important contributions to JI research as well as to management practices. The stress literature employs a model that suggests various antecedent stressors cause stress, which, in turn, causes strain and physical, psychological, or behavioral outcomes. In this model, contextual factors, such as stressors interact with individual factors in affecting the subjective appraisal of the stressful situation, which, in turn, results in affective reactions. Folkman and Lazarus defined stress as the relationship between the person and the environment that

7 26 Huang (China) et al. is appraised by the person as relevant to his or her well being and in which the person s resources are taxed or exceeded (1985, 152). This definition suggests that a cognitive appraisal of environmental threats, such as organizational changes (Ashford, Lee, and Bobko 1989; Lee, Bobko, and Chen 2006), can be sources of stress in organizations. As Lazarus (1991, 1993, 1999) noted, emotions follow from appraisals. In our case, cognitive JI is the appraisal of a threat. Affective JI follows from cognitive JI. In turn, employees may be more or less satisfied with or less committed to their jobs and organizations and they may experience physical and/or psychological problems. Furthermore, according to AET, after a negative workplace event, individuals typically first respond in terms of negative affect arousals such as anger or fear (Zhao et al. 2007). Research on neural systems and brain functioning has offered supporting evidence for this proposition (LeDoux 1995). Once triggered, affective reactions may overwhelm the individual. The importance and relevance of the event to personal goals will determine the intensity of the affect. Thus, as the most proximal reaction to the appraisal of a significant event, affect plays a central mediating role in determining the effect of the event on other outcomes. Based on the above analysis, we propose that cognitive JI is the appraisal of a workplace event, such as a job threat resulting from organizational change or from an employee not meeting performance norms. That appraisal results in affective reactions that, in turn, contribute to the formation of work attitudes and other individual outcomes such as physical and psychological problems. Therefore, we propose the following hypotheses to be tested in this study. Hypothesis 1: Cognitive JI relates positively to affective JI. Hypothesis 2: Affective JI relates negatively to job satisfaction (H2a) and organizational commitment (H2b); it relates positively to somatic complaints (H2c). Hypothesis 3: Affective JI mediates the negative relationships between cognitive JI and job satisfaction (H3a) and organizational commitment (H3b); it mediates the positive relationship between cognitive JI and somatic complaints (H3c). Methods Samples and procedures We collected data for this study from two samples. Sample 1 consisted of female nurses from three hospitals in a large city in central China. These hos-

8 Affective Job Insecurity 27 pitals were all state-owned and of similar size. We therefore combined data from the three hospitals. We followed the same procedure in data collection in the three hospitals. All the nurses were invited to complete our survey during their regular monthly meeting at work. Of the nurses who attended their regular monthly meeting, after listening to our description regarding the general purpose of the survey and the voluntary nature of their participation and after we assured confidentiality of the results, only one nurse from hospital A and two nurses from both hospitals B and C left the meeting room. In each hospital, the questionnaires were distributed and were collected by a research assistant. Each participant received a gift of about RMB10 (US$1.30) after completing the survey. The final sample consisted of 130, 228, and 92 participants from the three hospitals, respectively, and 12 other participants whose hospital information was missing. All 462 participants were female nurses. Their mean age was 25. On average, they had years of education and 4.75 years of tenure with the hospitals. To check the differences among participants from three hospitals, we compared the demographics of the participants using one-way ANOVA and the chi-square test. Participants from the three hospitals differed significantly in age (F(2, 442) = 28.98, p <.001), education (F(2, 440) = 6.25, p <.01) and organizational tenure (F(2, 441) = 41.32, p <.001). To address individual differences across organizations in our sample, in all the regression analyses, age, education, and tenure were included as control variables. Participants from the three hospitals did not differ significantly in the mean levels of affective JI (F(2, 447) = 1.71, p >.05). To control for differences at the organizational level, dummy variables for the three organizations were created as control variables (Cohen and Cohen, 1983). Because sample 1 only included female respondents in one occupation, to ensure generalizability of the result, we collected another sample, which consisted of employees/managers in a large bank located in a major city in southern China. This sample was used in a larger study focused on the measurement of cognitive JI (Lee et al. 2008). We mailed the questionnaires with a cover letter to the respondents via the internal mailing system of the bank, describing the general purpose of the survey and the voluntary nature of their participation and assuring confidentiality of the results. The respondents were given self-addressed envelopes in which to return their completed responses to our research assistant. In all, 289 subordinate questionnaires were distributed, and 219 completed questionnaires were returned for a 76 percent response rate. Of the 219 respondents, 41 percent were female. The respondents reported an average age of years, an average of years of education, and an average organizational tenure of 9.61 years. Of these participants, 78.7 percent held nonmanagerial positions, 14 percent were from lower manage-

9 28 Huang (China) et al. ment, and 7.3 percent were from middle management. The job functions of the respondents included teller, accounting/finance, sales/marketing, computer, customer service, administration/human resource, and technician. Measures Unless noted, we used the same measures in both samples. All the scales used in this study were translated into Chinese and then translated independently back into English (Brislin 1980). Expert judges in the Chinese language examined the questionnaire to ensure that the items were interpretable in Chinese. Unless noted, all multiple-item measures used a five-point response scale ranging from strongly disagree (1) to strongly agree (5). Affective job insecurity. We used seven items to measure affective JI: I wish I had more job security in this company ; The lack of job security in this company makes me feel very nervous ; I feel very uneasy about my chances of remaining with this company ; I lose sleep worrying about my future with this company ; I am unhappy with the amount of job security that I have ; I am very tense about maintaining my current job employment status ; and I am not confident that I can remain employed with this company for as long as I choose. Our items were written based on the work of Borg and Elizur (1992) and Johnson, Meese, and Crano (1984), capturing the affective elements of the job insecurity experience such as being afraid of losing the job or being concerned about future status with the firm. (Coefficient alphas =.72 and.71 in the nurse and bank samples, respectively.) Cognitive job insecurity. We adopt Lee, Bobko, and Chen s (2006) revised JIS to measure cognitive JI in this study. These authors added two items on total job importance and likelihood of job loss of total job to Ashford et al s original JIS and validated it using the Chinese samples. In the bank sample, we used this revised full JIS, which has 59 perceptual items, including 17 items on the importance of job features, 17 items on the likelihood of losing job features, 11 items on the importance of the total job, 11 items on the likelihood of total job loss, and three items to assess powerlessness. In the nurse sample, we used a short form of JIS that included 16 items on the total job (8 items on its importance and 8 on the likelihood of its loss), 16 items on job features (8 items on their importance and 8 on the likelihood of their loss) and 3 items on powerlessness. Lee et al. (2008) validated this abridged form of JIS. In both cases, the total score of this multiplicative cognitive JI measure was calculated according to its underlying theoretical derivation (cf. Greenhalgh and Rosenblatt 1984) as a multiplicative function as follows: JI

10 Affective Job Insecurity 29 = [sum (importance of job features likelihood of losing job features) + sum (importance of negative changes in total job likelihood of negative changes in total job)] [perceived powerlessness to resist threat]. Job satisfaction. Participants overall job satisfaction was assessed by the fiveitem general satisfaction scale of the Job Diagnostic Survey (Hackman and Oldham 1975). This scale uses a seven-point, strongly agree (7) to strongly disagree (1) scale format. (Coefficient alphas = 0.72 and 0.91 in the nurse and bank samples, respectively.) Somatic complaints. We used a ten-item scale developed and validated by Caplan et al. (1975) to measured to assess symptoms of strain. The scale items asked how frequently in the past month (never = 1, once or twice = 2, three times or more = 3) that respondents had symptoms such as heart beating hard, dizzy spells, trouble sleeping, and so forth. (Coefficient alphas =.84 and.83 in the nurse and bank samples, respectively.) Organizational commitment. In the hospital survey, we included an additional attitudinal outcome, organizational commitment. It was measured with Mowday, Steers, and Porter s (1979) nine-item scale using a seven-point, strongly agree (7) to strongly disagree (1) scale format. An example for sample items is I feel a strong sense of belonging to my organization. (The internal reliability coefficient alpha measure for the construct was.85.) We included a set of control variables because of their potential impact on the mediator and outcomes. In both surveys, we included individuals years of age, years of education, and number of years of tenure with the organizations. We created two dummy variables to control for the effect of the three hospitals. For the bank sample, we also included gender (male = 0; female = 1) and position (1 = nonmanagerial positions; 2 = lower management; and 3 = middle management). Results The means, standard deviations, and correlations among all the variables for the two samples are presented in Tables 2 and 3. All the measures have acceptable internal consistency reliability estimates. It should be noted that cognitive JI is conceptualized as a multidimensional construct that employs a multiplicative model as specified above (Ashford, Lee, and Bobko 1989; Greenhalgh and Rosenblatt 1984). An overall internal consistency reliability alpha is, therefore, not an appropriate measure for its reliability. For the same reason, we did not conduct confirmatory factor analysis (CFA) to examine its discriminant validity with the measure of affective JI. To examine the factor

11 30 Huang (China) et al. structure of our measure of affective JI, we first conducted an exploratory factor analysis (EFA) using data from the hospital sample. The principal component analysis method with Varimax rotation was used in the EFA. We expected the seven items to load on one factor, but two factors were identified with an eigenvalue larger than one. The first factor explained 50 percent of the total variance while the second factor only explained 14 percent. We further conducted a CFA using data from the bank sample. Results show that the one-factor solution with seven items was acceptable (χ 2 = 34.86; df = 9; RMSEA = 0.10; CFI =.92; NNFI =.86). Therefore, our subsequent analyses were based on this one-factor solution. Correlation results show that affective JI correlated moderately with cognitive JI in both the bank sample (r =.26, p <.01) and the nurse sample (r =.11, p <.05). Further, cognitive and affective JI showed different patterns of correlation with demographic variables. For example, tenure was unrelated to cognitive JI, but it was significantly related to affective JI in both samples. Together, these results provide some support for our argument that the two are distinct constructs. Consistent with our expectation, affective JI significantly correlated with all the outcome variables in both samples. Controlling for the effects of demographics, our regression results were consistent with the correlations. That is, the main effect of cognitive JI on individual outcomes was significant (including job satisfaction in both samples and organizational commitment in the nurse sample) and individual psychosomatic complaints in both samples. Second, the main effect of cognitive JI on affective JI was also significant in both the bank and nurse samples (β =.24, p <.05; β =.14, p <.01, respectively), supporting H1. Our results also show that the main effect of affective JI on job satisfaction (β =.20, p <.05; β =.23, p <.01, respectively) and somatic complaints (β =.25, p <.01; β =.10, p <.10, respectively) is significant in both samples. Lastly, the main effect of affective JI was also significant on organizational commitment in the nurse sample β =.22, p <.01), supporting H2. (This was not measured in the bank sample.) To test the hypothesis that affective JI mediates the relationships between cognitive JI and individual outcomes, we used hierarchical regressions following Baron and Kenny s (1986) three-step approach. Results for the bank and nurse sample data are shown in Tables 4 and 5, respectively. Affective JI had an incremental effect on outcomes after controlling for cognitive JI. When the effects of affective JI were controlled, the association between cognitive JI and the outcomes decreased. However, the effects of cognitive JI were still significant at the same level for all the outcomes except for organizational commitment in the nurse sample in which the coefficient of cognitive JI dropped from.11 (p <.05) to.08 (p <.10). We also tested

12 Affective Job Insecurity 31 Table 2 Descriptive statistics, reliability coefficients, and correlations among variables from the bank sample Mean S.D Age Gender Education Tenure **.14***.18* Position ** * Cognitive JI 1, Affective JI *.03.27** (.71) Job satisfaction *** ***.10.43**.22** (.91) Somatic complaints *.24**.23**.25** (.83) * p <.05; ** p <.01; *** p <.001. Notes: N = 219. Internal consistency reliability coefficients (alphas) appear on the diagonal in parentheses. The table uses the full version of Ashford, Lee, and Bobko s (1989) JI measure.

13 32 Huang (China) et al. Table 3 Descriptive statistics, reliability coefficients, and correlations among variables from the nurse sample Mean S.D Age Education *** 3 Tenure ***.22*** 4 Dummy ***.10**.40*** 5 Dummy ***.04.27***.65*** 6 Cognitive JI ** 7 Affective JI ***.11**.20*** ** (.72) 8 Job satisfaction ***.26*** (.72) 9 Org ***.10**.12***.09.13***.11**.24***.53*** (.85) commitment 10 Somatic ***.09.19***.09*.24***.20*** (.84) complaints *p <.05; **p <.01; ***p <.001 Notes: N = 462. Internal consistency reliability coefficients (alphas) appear on the diagonal in parentheses. The table uses the abridged version of the Ashford, Lee, and Bobko s JI measure (see Lee et al. 2008).

14 Affective Job Insecurity 33 Table 4 Results of hierarchical regression for testing mediation with the bank sample data Control AJI Job satisfaction Somatic complaints Model 1 Model 2 Model 3 Model 4 Model 5 Age Gender.04.23**.22* Education Tenure Position *.20* Independent JI.24**.39***.35***.34***.29*** Mediator AJI.20*.25** DR 2 (note).05**.15***.04**.11***.06*** DF 6.07** 19.88*** 5.21** 13.68*** 7.46*** Adjusted R * p <.05; ** p <.01; *** p <.001. Notes: DR 2 is for the last variable entered into the model. N = 219. whether the mediation effects were significant using the Sobel test (Preacher and Leonardelli 2003; Sobel 1982). The Sobel test revealed that, in the bank sample, affective JI was a significant partial mediator between cognitive JI and job satisfaction (z = 2.39, p <.05) and between cognitive JI and somatic complaints (z = 2.49, p <.05). In the nurse sample, affective JI was a significant partial mediator between cognitive JI and job satisfaction (z = 2.14, p <.05) and between cognitive JI and organizational commitment (z = 2.12, p <.05), but not between cognitive JI and somatic complaints (z = 1.48, p >.10). Overall, the results show that affective JI partially mediated the relationship between affective JI and employee outcomes. In all, H1 and H2 were supported while the results supported a partial mediation instead of a full mediation relationship in H3.

15 34 Huang (China) et al. Table 5 Results of hierarchical regression for testing mediation with the nurse sample data AJI Job satisfaction Org. commitment Somatic complaints Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7 Control Age *.22.23** Education.09* Tenure.25** Dummy **.13* ***.17** Dummy *.14** Independent Cognitive JI.14***.21***.17***.11**.08*.19***.17*** Mediator Affective JI.23***.22***.09* DR 2 (note).02***.04***.05***.01**.04***.03***.01* DAdjusted R * p <.05; **p <.01; *** p <.001. Notes: DR 2 is for the last variable entered into the model. N = 462.

16 Affective Job Insecurity 35 Discussion Research on JI has grown considerably since the 1980s as a result of the fastchanging nature of organizations and the more competitive and dynamic organizational environment they face. Although researchers have found evidence for the impact of JI on important individual and organizational outcomes, several gaps in the literature still exist. In this study, we clarify the distinct conceptual meanings of cognitive and affective JI and highlight the importance of treating them as two different constructs. In support of Borg and Elizur (1992) and Reisel and Banai (2002), our results from two samples showed initial support for the distinctiveness of cognitive and affective JI. Affective JI is an outcome of cognitive JI that deserves additional attention in future longitudinal studies. Further, building on the conceptual framework established in stress research and the AET, we proposed and tested a mediating process of JI in which affective JI is an outcome of cognitive JI and in which affective JI mediates the relationship between cognitive JI and important employee outcomes. Results from a sample of bank employees and a sample of nurses from three hospitals show that affective JI partially explains the effect of cognitive JI on job satisfaction, organizational commitment and somatic complaints. Using two samples from different types of organizations and representing different occupational groups provide evidence for generalizability of our findings in Chinese settings. This study contributes to the literature in a number of ways. First, it suggests that to facilitate an appropriate interpretation of results, researchers should be careful not to combine cognitive and affective JI items into one scale (e.g., Johnson, Meese, and Crano 1984). Second, we have identified affective JI as a mediator between the cognitive JI and employee outcomes. Future studies can further examine the construct validity of affective JI by extending our nomological network and including other antecedents and outcomes. For example, having strong supportive relationships with one s supervisor may lower the negative effects of affective JI or buffer the effects of cognitive JI on affective JI. Moreover, research should also explore individual differences in the reaction to cognitive JI. The relationship between cognitive and affective JI may vary if the individual is optimistic, for example. Further, in time, people s reactions to cognitive JI may be weakened. Therefore, longitudinal studies are needed to explore the mechanism of affective JI. Although we took careful steps in developing the affective JI items, there were differences in results from the two samples. That is, the seven-item affective JI scale loaded on two factors in the hospital sample while this scale loaded on one factor in the bank sample. It is possible that there are differences in perception by our female sample (hospital) and a mixed gender sample

17 36 Huang (China) et al. (bank). Future studies should further refine our affective JI scale across different occupational groups in a variety of settings. This study has its limitations. Although we used two samples, our study is cross-sectional in nature, which did not enable us to investigate cause-effect relationships as our framework suggests. Future studies using a longitudinal design are needed to examine the discriminant and predictive validity of the affective JI construct. Because of the construct structure of cognitive JI, we could not demonstrate discriminant validity of affective JI using confirmatory factor analysis. It is important that future studies, using longitudinal designs, investigate the different antecedents and consequences of both the cognitive and affective JI constructs. The second limitation of our study is that common method bias may explain some of the positive associations among our variables. It was not plausible to ask respondents in independent samples to respond to the studied variables in the present study such as the subject s insecurity or satisfaction. Future studies are needed that involve different data sources for variables such as supervisor-rated behavior outcomes. Finally, the factor analysis results of the seven-item measure of affective JI were acceptable but not completely satisfactory. Future studies should improve the measure, especially by avoiding reverse-coded items and by paying attention to item wording if used with a Chinese- or other non-english-speaking samples. In conclusion, despite these limitations, we believe that our study contributes to the job insecurity literature by clarifying that cognitive and affective JI are two distinct constructs and by showing that affective JI partially mediates the relationships between cognitive JI and individual outcomes. Although still at an exploratory stage, this study has important implications for managers especially in Chinese contexts. First, consistent with results from other cultures, our results provide further evidence of the negative effects of both cognitive and affective JI on employee work attitudes and psychological well-being. Employees perceptions about the security of their jobs can lead to worries or concerns about their jobs. Managers should therefore employ measures to decrease both cognitive and affective JI. We believe that there may be some individual and organizational coping strategies and resources toward JI that is especially useful in Chinese context, such as having quality supervisor-subordinate relationships, impression management strategies, and supportive organizational practices, that deserve future studies. References Ashford, S.J., C.L. Lee, and P. Bobko Content, Causes, and Consequences of Job Insecurity: A Theory-Based Measure and Substantive Test. Academy of Management Journal 32:

18 Affective Job Insecurity 37 Ashford, S.J., E. George, and R. Blatt Chapter 2: Old Assumptions, New Work: The Opportunities and Challenges of Research on Nonstandard Employment. Academy of Management Annuals 1: Baron, R.M., and D.A. Kenny The Moderator-Mediator Variable Distinction in Social Psychological Research: Conceptual, Strategic and Statistical Considerations. Journal of Personality and Social Psychology 51: Borg, I., and D. Elizur Job insecurity: Correlates, moderators and measurement. International Journal of Manpower 13(2): Brislin, R.W Translation and Content Analysis of Oral and Written Material. In Handbook of Cross-Cultural Psychology, Volume 2, ed. H.C. Triandis and J.W. Berry, Boston: Allyn and Bacon. Caplan, R.D., S. Cobb, Jr., J.R.P. French, R. Van Harrison, Jr., and S.R. Pinneau Job Demands and Worker Health: Main Effects and Occupational Differences. Washington, DC: U.S. Department of Health, Education and Welfare. Cohen, J., and P. Cohen Applied Multiple Regression/Correlation Analysis for the Behavioral Sciences. Hillsdale, NJ: Lawrence Erlbaum Associates. Davy, J.A., A.J. Kinicki, and C.L. Scheck A Test of Job Security s Direct and Mediated Effects on Withdrawal Cognitions. Journal of Organizational Behavior 18: Folkman, S., and R.S. Lazarus If It Changes It Must Be a Process: Study of Emotions and Coping During Three Stages of a College Examination. Journal of Personal and Social Psychology 48(1): Greenhalgh, L., and Z. Rosenblatt Job Insecurity: Toward Conceptual Clarity. Academy of Management Review 9: Hackman, J.R., and G.R. Oldham Development of the Job Diagnostic Survey. Journal of Applied Psychology 60: Hartley, J., D. Jacobson, B. Klandermans, and T. Van Vuuren Job Insecurity: Coping with Jobs at Risk. London: Sage. Hellgren, J., M. Sverke, and K. Isaksson A Two-Dimensional Approach to Job Insecurity: Consequences for Employee Attitudes and Well-Being. European Journal of Work and Organization Psychology 8: Ito, J.K., and C.M. Brotheridge Exploring the Predictors and Consequences of Job Insecurity s Components. Journal of Managerial Psychology 22(1): Johnson, C.D., L.A. Messe, and W.D. Crano Predicting Job Performance of Low Income Workers: The Work Opinion Questionnaire. Personnel Psychology 37: Klandermans, B., and T. van Vuuren Job Insecurity: Introduction. European Journal of Work and Organizational Psychology 8(2): König, C. J., and T. Staufenbiel The Difference Between Cognitive and Affective Job Insecurity. Paper presented at the 21st annual conference of the Society for Industrial and Organizational Psychology, Dallas, TX. Lazarus, R.S Emotion and Adaptation. New York: Oxford University Press Coping Theory and Research: Past, Present and Future. Psychosomatic Medicine 55: Stress and Emotion: A New Synthesis. New York: Springer. LeDoux, J.E Emotion: Clues from the Brain. Annual Review of Psychology 46:

19 38 Huang (China) et al. Lee, C., P. Bobko, and Z.X. Chen Investigation of the Multidimensional Model of Job Insecurity in Two Countries. Applied Psychology: An International Review 55: Lee, C., P. Bobko, S.J. Ashford, Z.X. Chen, and X.P. Ren Crosscultural Development of an Abridged Job Insecurity Measure. Journal of Organizational Behavior 29: Mignonac, K., and O. Herrbach Linking Work Events, Affective States, and Attitudes: An Empirical Study of Managers Emotions. Journal of Business and Psychology 19: Morrison, E.W., and S.L. Robinson When Employees Feel Betrayed: A Model of How Psychological Contract Violation Develops. Academy of Management Review 22: Mowday, R.T., R. M. Steers, and L.W. Porter The Measurement of Organizational Commitment. Journal of Vocational Behavior 14: Preacher, K.J., and G.J. Leonardelli Calculation for the Sobel Test: An Interactive Calculation Tool for Mediation Tests. sobel/sobel.htm, accessed February 16, Probst, T.M Development and Validation of the Job Security Index and the Job Security Satisfaction Scale: A Classical Test Theory and IRT Approach. Journal of Occupational and Organizational Psychology 76: Reisel, W.D., and M. Banai Job Insecurity Revisited: Reformulating with Affect. Journal of Behavioral and Applied Management 4: Robinson, S.L, and D.M. Rousseau Breaching the Psychological Contract: Not the Exception but the Norm. Journal of Organizational Behavior 15: Rousseau, D.M Psychological Contracts in Organizations: Understanding Written and Unwritten Agreements. Thousand Oaks, CA: Sage. Rupp, D.E., and S. Spencer When Customers Lash Out: The Effects of Customer Interactional Injustice in Emotional Labor and the Mediating Role of Discrete Emotions. Journal of Applied Psychology 91: Schabracq, M.J., and C.L. Cooper The Changing Nature of Work and Stress. Journal of Managerial Psychology 15: Sobel, M.E Asymptotic Intervals for Indirect Effects in Structural Equations Models. In Sociological Methodology, ed. S. Leinhart, San Francisco: Jossey-Bass. Sverke, M., J. Hellgren, and K. Näswall No Security: A Meta-Analysis and Review of Job Insecurity and Its Consequences. Journal of Occupational Health Psychology 7: Van Vuuren, C., and P. G. Klandermans Individual Reactions to Job Insecurity: An Integrated Model. In European Perspectives in Psychology, vol. 3, ed. P.J.D. Drenth, J.A. Sergeant, and R.J. Takens, Oxford: John Wiley. Weiss, H.M., and R. Cropanzano Affective Events Theory: A Theoretical Discussion of the Structure, Causes and Consequences of Affective Experiences at Work. In Research in Organizational Behavior: An Annual Series of Analytical Essays and Critical Reviews, vol. 18, ed. B.M. Staw and L.L.Cummings, Greenwich, CT: Elsevier Science/JAI Press. Wong, Y.T., C.-S. Wong, H.-Y. Ngo, and S. Lui Different Responses to Job Insecurity of Chinese Workers in Joint Ventures and State-Owned Enterprises. Human Relations 58(11):

20 Affective Job Insecurity 39 Zhao, H., S.J. Wayne, B.C. Glibkowski, and J. Bravo, The Impact of Psychological Contract Breach in Work-Related Outcomes: A Meta-Analysis. Personnel Psychology 60: To order reprints, call ; outside the United States, call

21 Copyright of International Studies of Management & Organization is the property of M.E. Sharpe Inc. and its content may not be copied or ed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or articles for individual use.