The roles of shared perceptions of individual job insecurity and job insecurity climate for work- and health-related outcomes: A multilevel approach

Abstract

The aim of this study is to examine job insecurity from a multilevel perspective and to investigate the roles of two types of job insecurity – job insecurity climate and individual job insecurity – for work-related attitudes and health outcomes. It further explores the role of the workgroup – as a social context – in shaping job insecurity perceptions. Data were collected from white-collar employees in a Swedish organization, with 126 participants nested in 18 groups. The results show that 19% of the variance in job insecurity climate perceptions, and none of the variance in individual job insecurity perceptions, could be attributed to group membership. Further, compared to other members of their group, those perceiving a stronger job insecurity climate reported lower levels of negative self-rated health and higher burnout scores. These results imply that the workgroup is an important social context for job insecurity climate perceptions.

Keywords

Burnout job insecurity job insecurity climate job satisfaction multilevel analysis productivity self-rated health

In a changing world, many employees worry about potentially losing their job (Burgard et al., 2009; Heery, 2000; ILO, 2015). When studied from a psychological perspective, job insecurity is conceptualized as a subjective perception, and it has been defined as ‘the perception of a potential threat to the continuity of the current job’ (Sverke et al., 2002: 243). However, worrying about job loss is not necessarily just a personal and private concern; it can also exist at a social level as a job insecurity climate. As such, it may manifest itself through conversations and rumors about the future of the organization or as a general impression among employees when, for example, negative changes are expected. Job insecurity can therefore be conceptualized as an individual phenomenon, concerning an individual’s perceptions of insecurity regarding one’s own job, or as a social phenomenon, concerning a climate of job insecurity within an organization or a workgroup. This distinction between job insecurity as an individual perception and as a social climate has been shown in a number of previous studies (e.g., De Cuyper et al., 2009; Sora et al., 2009).

A large body of literature demonstrates the negative effects of individual job insecurity on work-related attitudes, health, and productivity (for meta-analyses, see Cheng and Chan, 2008; Sverke et al., 2002). In studies where the job insecurity climate is operationalized as individual job insecurity perceptions aggregated to the group level, the associations found between aggregated perceptions and various outcomes have been similar to those found between person-level individual job insecurity and the outcomes (Sora et al., 2009, 2013). These results suggest that job insecurity has certain group-level properties, but additional research into multilevel job insecurity and its outcomes is called for to disentangle the person- and group-level sources of the variance in job insecurity. What these previous studies lack is a measure of individuals’ perceptions of the climate of job insecurity at their workplace, i.e., data on how employees perceive their workgroup generally feels in regard to job insecurity (see e.g., Låstad et al., 2015; Mauno et al., 2013).

The present study investigates job insecurity both in terms of how individuals perceive their own job insecurity and how they perceive job insecurity levels among others in their workgroup, here referred to as job insecurity climate. These two aspects of job insecurity are also aggregated to the workgroup level. Thus, the present study aims to investigate how individual job insecurity and job insecurity climate, at both the person and group levels, relate to work attitudes, health, and productivity. The present study will contribute to a deeper understanding of the relations between job insecurity and its outcomes and between shared perceptions of job insecurity climate and both attitudes and health at work.

Job insecurity climate

Given that job insecurity perceptions arise in a context which is inherently social (the workplace), it is plausible that job insecurity perceptions arise not only as an individual experience but also at an interpersonal level. This line of reasoning also extends to other, similar constructs. For instance, both stress and coping have previously been described as partly collective phenomena, in that people as a group can perceive a situation to be stressful, and that the group can cope with the situation as a collective (e.g., Länsisalmi et al., 2000). Sora et al. (2009: 130) define job insecurity climate as ‘a set of shared perceptions of powerlessness to maintain the continuity of threatened jobs in an organization.’ In previous studies, job insecurity climate is represented by data based on individual job insecurity measured at the person level and then aggregated to the workgroup level. However, a limitation of this approach is that it measures job insecurity climate indirectly, since aggregated scales with an individual referent (in this case, asking specifically about the individual’s own job insecurity) does not necessarily reflect a social climate (see e.g., Baltes et al., 2009 for a general discussion; and Mauno et al., 2013 for a discussion specifically on job insecurity); this approach does not include any information about how individuals perceive the climate around them. A social climate is not constituted just by the combination of a group of individuals’ perceptions about themselves; it can also reflect individuals’ perceptions of their social surroundings (e.g., psychosocial work environment) (see e.g., Parker et al., 2003). In the present study, the conceptualization of climate focuses on individuals’ perceptions of their surroundings, and job insecurity climate is measured directly, by asking specifically about the situation in the workgroup. This approach means that the individual will be reporting on the climate they perceive at their workplace − which is in line with the referent-shift model (Chan, 1998), and follows recommendations offered by Mauno et al. (2013). These person-level climate perceptions can (given a sufficient level of within-group inter-rater agreement) in turn be aggregated to the workgroup level to reflect shared perceptions of the job insecurity climate at work. Thus, the two approaches differ in their conceptualization of job insecurity climate: the direct consensus model (used by Sora et al., 2009) considers job insecurity climate as individual job insecurity aggregated to the unit level (e.g., the workgroup or organization) and the alternative referent-shift model, used in this study, focuses on the perception of the job insecurity climate at the person and workgroup levels.

The relationship between job insecurity climate and outcomes

The process through which both individual job insecurity and job insecurity climate negatively affect psychological wellbeing can be explained by transactional stress theory: stress results from a process where certain stimuli are appraised as potential threats by an individual, threats which the individual does not perceive he or she has the necessary resources to cope with (Lazarus and Folkman, 1984). Anticipating potential job loss creates tension because of the possible negative consequences, such as long-time unemployment, financial problems, loss of social status, and loss of one’s social arena outside the family sphere (cf. Jahoda, 1982). The severity of these negative potential consequences, combined with a perceived powerlessness to counteract the perceived threat of job loss, contributes to the person experiencing job insecurity (cf. Greenhalgh and Rosenblatt, 1984).

Previous research is consistent with predictions based on transactional stress theory, and has shown that individual job insecurity may have health-related outcomes. Regarding physical health, job insecurity is generally found to have a negative effect on health (e.g., Kalil et al., 2009; László et al., 2010; Schreurs et al., 2010) and it has also been associated with higher sickness absence (Kinnunen et al., 1999). Previous research has also identified negative consequences of individual job insecurity for mental health (see e.g., Burchell, 2009; Huang et al., 2012; Näswall et al., 2005; Silla et al., 2009; Størseth, 2006), and in a similar vein, burnout has been identified as a consequence of job insecurity (see e.g., Hu and Schaufeli, 2011; Tilakdharee et al., 2010). Because individual job insecurity has been strongly linked to both physical and mental health, we included a measure of general subjective health and a measure of mental ill-health: burnout, which can be defined as ‘a syndrome of emotional exhaustion, depersonalization, and reduced personal accomplishment’ (Maslach and Jackson, 1981: 99).

For the work-related outcomes, job satisfaction, which can be defined as the ‘pleasurable emotional state resulting from the appraisal of one’s job as achieving or facilitating one’s job values’ (Locke, 1969: 316), is the most frequently investigated outcome of job insecurity. Previous research has repeatedly found a strong negative relationship between individual job insecurity and job satisfaction (see e.g., König et al., 2011; Mauno et al., 2005; Näswall et al., 2005; Sora et al., 2010a; Størseth, 2006). Research on job insecurity typically also focuses on work-related behaviors and we included self-rated productivity as a measure of performance. Concerning performance, previous findings are more mixed. For instance, in a study comparing individuals threatened with layoffs with a control group, the rate of production was higher for the job insecure individuals, but the assessed quality of their work was lower (Probst, 2002). Similarly, in a more recent study, job insecurity was related to more negative work attitudes, which in turn led to poorer performance, but there was also a simultaneous direct effect on performance, so that job insecurity predicted higher performance ratings (Staufenbiel and König, 2010).

Similar reasoning to why individual job insecurity is considered a work-related stressor (see e.g., De Witte, 1999) can also be applied to the job insecurity climate. Since individuals are part of social contexts (work environments), mechanisms such as the appraisal of situational threats and of potential coping strategies, as described in transactional stress theory (Lazarus and Folkman, 1984), are at play when social or contextual cues indicate that there is a climate of job insecurity at the workplace. There could for instance be rumors about organizational changes in a firm and these group-level phenomena result in appraisals of threat on the individual level. The few previous studies on job insecurity climate (where job insecurity climate was operationalized as job insecurity aggregated to the team or organizational level) have reported some negative effects of shared job insecurity perceptions that are similar to the effects of individual job insecurity. The first study that conceptualized job insecurity climate as an aggregation of individual job insecurity found these aggregated perceptions to be associated with lower levels of job satisfaction and organizational commitment (Sora et al., 2009). A second study (Sora et al., 2013) showed that aggregated perceptions of individual job insecurity were negatively associated with job satisfaction, organizational commitment, work involvement, and organizational trust, and that these relations were more pronounced in units where the employees had similar perceptions (operationalized as climate strength). Further, in a multilevel study on aggregated job insecurity (where, notably, aggregated individual job insecurity was not defined as job insecurity climate), Mauno et al. (2013) found that the relationship between job insecurity and turnover intention was partly mediated by occupational wellbeing at both the department and person levels. They also found support for a cross-level interaction, indicating that poor wellbeing at the department level strengthened the negative association between job insecurity and wellbeing at the person level. Taken together, these studies indicate that the relationships between aggregated perceptions of individual job insecurity and work- and health-related outcomes are similar to those found in previous studies on person-level individual job insecurity, but there is a lack of empirical evidence regarding the role of job insecurity climate in predicting these outcomes.

The present study

The first aim of the present study is to contribute to a better understanding of the role of the workgroup by investigating its role as a social context for individual job insecurity and job insecurity climate perceptions. We expect that job insecurity climate perceptions have a stronger social component than perceptions of individual job insecurity. The second aim is to investigate possible outcomes of job insecurity climate. In order to do this we decided to focus on outcomes that in previous research have been identified as outcomes of job insecurity (for meta-analysis results, see Cheng and Chan, 2008; Sverke et al., 2002). More specifically, and in accordance with previous research, we focused on both work-related and health-related outcomes. The present study adds to previous research by including a direct measure of job insecurity as a psychological climate perception and investigating its association with work- and health-related outcomes at the person level. Regarding the relationships between job insecurity (climate and individual) and the specific outcome variables included in this study, it would be expected for the direction of effects to be in line with findings from previous studies. As research by Sora et al. (2009, 2013) has already indicated, the total variance explained in the outcomes tends to increase when individual job insecurity aggregated to the group level is included in the analyses. We propose that a direct measure of job insecurity climate (i.e., asking respondents to report on the climate of job insecurity at their workplace), in contrast to the indirect measure (aggregation of individuals’ perceptions of their own job insecurity), will further increase the variance explained in these outcomes, which would contribute to an increased understanding of the negative associations between the experiencing of job insecurity of any type and attitudinal and wellbeing outcomes. We expect not only to replicate previous findings regarding the relationships between individual insecurity and outcomes, but to also find similar patterns of relationships between job insecurity climate and outcomes.

Method

Data collection and participants

The data were collected as part of a larger research project in a Swedish branch of a private sector IT-consultancy company. The data collection was conducted via an online questionnaire. Of the 209 employees invited to take part in the study, 150 filled in the questionnaire, producing a response rate of 72%. The respondents were white-collar workers who were working either in IT-consultancy or in-house services such as administration, HR, or marketing. The employees were informed that their answers would be treated confidentially, that their participation was voluntary, and that they could discontinue their participation in the study at any time. After missing values analysis, 11 respondents were removed from the sample since they had not responded to any of the items measuring our study variables. Also, since the nested structure of the dataset was to be analyzed, an additional 13 respondents were removed after their unique responses to who their closest manager was would have placed them alone in their groups. A second missing values analysis of the effective sample showed that for the remaining respondents (N = 126), 6% of the values in the study variables were missing (missing at random; MAR). Considering the small sample size, missing values were imputed using the EM algorithm (see e.g., Allison, 2001), including only the study variables. The effective sample consisted of 18 groups, made up of 126 respondents in total (with a range of between 3 and 17 members in each workgroup). The respondents’ age ranged from 24 to 65 years; 55.6% had a university degree; the average organizational tenure was 4.4 years (SD = 1.5); they all had permanent contracts; and 39% were women.

Measures

Table 1 shows the means, standard deviations, Cronbach’s alphas, and correlations for the study variables. Regarding Cronbach’s alphas, with the exception of .67 (just below .70) for the scale measuring burnout, the reliability of all the other scales was satisfactory, ranging from .82 to .96. As expected, there was a positive correlation between job insecurity and job insecurity climate. Unless otherwise stated, the response scales for the measures had a Likert scale format ranging from 1 (strongly disagree) to 5 (strongly agree).

Table 1.

Correlations and descriptive statistics of all scales.

Scale	M	SD	α	1.	2.	3.	4.	5.	6.	7.	8.
1. Gender (woman)	0.39	–	–
2. Age	47.49	10.25	–	.18*
3. Education (university)	0.56	–	–	−.11	.02
4. Individual job insecurity	1.95	1.09	.96	.22*	.13	−.10
5. Job insecurity climate	2.46	1.04	.91	.04	.16	.06	.37**
6. Job satisfaction	3.72	.83	.92	−.10	−.21*	−.03	−.06	−.12
7. Productivity	3.27	.49	.82	−.05	−.16	.02	−.13	−.03	.01
8. Self-rated health	3.98	.79	–	−.12	.01	.03	.00	−.10	.21	−.04
9. Burnout	3.88	.60	.67	.09	.02	.01	−.04	.12	−.47**	.11	−.42**

N = 126. *p < .05, **p < .01. – not applicable.

Note: Scale range for burnout 1–7, for all other latent variables 1–5.

Job insecurity

Job insecurity was measured with a three-item scale (Hellgren et al., 1999). One sample item was: ‘I am worried about being let go.’

Job insecurity climate

Job insecurity climate was measured by a four-item scale (Låstad et al., 2015), including items such as ‘At my workplace there is a general feeling of anxiety over being let go.’

Outcomes

Job satisfaction was measured by three items from Hellgren et al. (1997), based on Brayfield and Rothe (1951). One sample item was ‘I am satisfied with my job.’ Productivity was measured by four items developed for this data collection: ‘How would you personally evaluate…? a) the quality of your work; b) the amount of work you output; c) how much you contribute, in total, to your department’s productivity; and d) how much you contribute to a positive work climate at your workplace.’ The response scale ranged from 1 (‘far below expectations’) to 5 (‘far exceeding expectations’). Self-rated health (SRH) was measured with a single item: ‘How would you rate your health at the present time?’ (Idler and Benyamini, 1997), with a response scale ranging from 1 (very poor) to 5 (excellent). Burnout was measured by 16 items from the Swedish version (Schutte et al., 2000) of the Maslach Burnout Inventory-General Survey (MBI-GS) (Schaufeli et al., 1996), comprising the subscales cynicism (five items; e.g., ‘I’ve become less engaged in my work’), reduced professional efficacy (six items, e.g., ‘I can solve problems that occur in my work in an efficient way’; reverse scored), and emotional exhaustion (five items, e.g., ‘I feel emotionally drained from my work’). Response alternatives ranged from 1 (never) to 7 (every day). The three subscales were combined into an overall burnout index.

Control variables

Previous research has consistently shown demographic variables to be related to differences in work-related attitudes and health (Ahola et al., 2006; Idler, 2003; Kooij et al., 2011). For job insecurity specifically, previous studies have indicated that older workers (Cheng and Chan, 2008) and workers with lower education (e.g., De Bustillo and De Pedraza, 2010; Näswall and De Witte, 2003) perceive higher levels of job insecurity than others. Consequently, age (in years) and education (1 = university degree, 0 = lower education) were included as control variables in the analyses. Regarding gender, previous research shows somewhat mixed results, with some studies reporting higher levels of individual job insecurity among men (e.g., Rosenblatt et al., 1999) and some reporting more job insecurity among female respondents (Kinnunen et al., 2000). In order to control for possible gender effects, gender (0 = man, 1 = woman) was also included.

Data analysis

All analyses were performed in SPSS 21. Since the respondents in this study were clustered within workgroups, and thus not independent of each other, a multilevel approach was appropriate for the analysis of the data. In order to confirm this, we estimated both within-group inter-rater agreement (r_WG(J)) and the intraclass correlation (ICC; described in more detail below) for the measures of both job insecurity and job insecurity climate. As a first step, the r_WG(J) for job insecurity and job insecurity climate in each group was estimated using the procedure described by LeBreton and Senter (2008), after which the mean r_WG(J) for all groups for the two variables were calculated. For job insecurity climate, the mean r_WG(J) for all groups was .71, and for job insecurity the mean r_WG(J) was .60. This is below the traditional lower limit cut-off value of .70, but there has been considerable debate on whether this is an appropriate cut-off point (LeBreton and Senter, 2008). Some researchers have for instance suggested a lower limit of .60 (Wagner et al., 2010). It should also be noted that with small groups (i.e., < 10 judges), the r_WG(J) values will be attenuated (Kozlowski and Hattrup, 1992; LeBreton and Senter, 2008; Lindell et al., 1999). Further, it has been suggested to look at the pattern of r_WG(J) scores for the groups (LeBreton and Senter, 2008). In the present study, the r_WG(J) scores for job insecurity climate ranged from .10 to .98, and 78% of the groups had an r_WG(J) score above the lower recommended value of .60. For job insecurity, r_WG(J) ranged from .02 to 1 and 61% of the groups’ r_WG(J) scores were above the cut-off (.60). This pattern implies that the vast majority of the groups were in agreement on their perceptions of job insecurity climate, and, to some extent, this is also true for individual job insecurity.

In order to examine the amount of variance in job insecurity and job insecurity climate that could be attributed to group membership, the ICCs were estimated as the first step of the multilevel analysis in SPSS 21. This was done by fitting an empty two-level model for the latent variables, and estimating only random intercepts, without any predictors. Based on this analysis, the ICCs for each variable were calculated by dividing the random intercept variance by the total variance (residual and intercept variance; see e.g., Hox, 2002). The results showed that workgroup membership accounted for 19% of the variance in job insecurity climate (ICC = .21/(.21+.88) = .19) and none of the variance in individual job insecurity (ICC = .00/(.00+1.19) = 0). With the exception of self-rated health which had an ICC of 3% (ICC = 0.02/(0.02+0.64) = .03), the same was true for the dependent variables in the study, which were around zero. Multilevel analysis is deemed meaningful and appropriate when part of the total variance can be attributed to group membership (ICC > 0), and when the design effect, defined as 1+(n–1)ICC, is greater than 2.0 (Peugh, 2010). For job insecurity climate, the design effect was 2.14 (1+(7–1).19 = 2.14). This indicates, together with the non-zero ICC value for job insecurity climate, and the nested structure of the data with predictors on both the individual and group level, that multilevel modeling is appropriate.

Two variables were created to represent job insecurity climate at the two levels: person-level job insecurity climate, which was centered around the group mean, and workgroup job insecurity climate (the person-level ratings aggregated to the group level) centered around the grand mean. The same procedure was followed for individual job insecurity. The relationships between individual job insecurity and job insecurity climate at the person and workgroup levels and the outcome variables job satisfaction, productivity, self-rated health, and burnout were tested next. Job insecurity climate and individual job insecurity were entered both as a person-level and a group-level predictor. The four predictors were added to the model in three steps: In the first step, person-level and workgroup-level job insecurity climate and control variables (age, gender, education) were added (Model 1). In the second step, the same analysis was run with person-level and workgroup-level individual job insecurity and control variables (Model 2). The final model included both person-level and workgroup-level job insecurity climate and person-level and workgroup-level individual job insecurity as predictors and the same control variables as in the previous analyses (Model 3).

Results

The results of the multilevel analyses predicting job satisfaction, productivity, and health outcomes are presented in Table 2. For job satisfaction and productivity, neither of the predictor variables was a significant predictor (Models 1–3). For self-rated health and burnout, person-level job insecurity climate was a significant predictor in that perceiving a stronger job insecurity climate in the workgroup than other group members was related to poorer self-rated health (Estimate = –.19, p < .05 [Model 1]) and higher levels of burnout (Estimate = .20, p < .05 [Model 1]). This was true even when including person-level and workgroup-level individual job insecurity in the analysis (for self-rated health: Estimate = –.22, p < .05, and for burnout: Estimate = .25, p < .001 [Model 3]). Being in a group with a higher mean level of job insecurity climate than other groups did not affect any of the outcomes (Models 2 and 3). For job satisfaction, self-rated health, and burnout, the results showed a slight increase in explained variance when job insecurity climate was included as a predictor (Models 1 and 3), compared to when individual job insecurity was the sole predictor (Model 2). The lack of associations found between individual job insecurity and the outcomes were in line with the non-significant bivariate correlations between person-level individual job insecurity and outcomes. Further, the correlation between individual job insecurity and job insecurity climate was lower than what one might have expected considering the nature of the two constructs. This may suggest that individuals are able to perceive different levels of individual job insecurity and job insecurity climate.

Table 2.

Multilevel models for outcome variables with individual job insecurity and job insecurity climate as main predictors.

Parameter	Job satisfaction						Productivity
	Model 1		Model 2		Model 3		Model 1		Model 2		Model 3
	Coefficient	SE	Coefficient	SE	Coefficient	SE	Coefficient	SE	Coefficient	SE	Coefficient	SE
Intercept	4.39**	.39	4.38**		4.40**	.39	3.54**	.25	3.52**	.24	3.52**	.25
Fixed effects
Workgroup job insecurity climate	.00	.13			.06	.13	−.02	.09			−.01	.09
Person-level job insecurity climate	−.10	.08			−.12	.09	−.01	.05			.04	.06
Workgroup individual job insecurity			−.10	.18	−.13	.20			−.06	.60	−.07	.12
Person-level individual job insecurity			−.00	.07	.04	.08			−.08	.05	−.10	.05
Age	−.02*	.01	−.02*	.01	−.02*	.01	−.01	.00	−.01	.01	−.01	.00
Gender (woman)	.12	.15	.10	.15	−.12	.15	.06	.10	.04	.10	.04	.10
Education (university)	.04	.15	.06	.15	.04	.15	.01	.09	.02	.09	.03	.09
Variance components (random effects)
Individual (Residual, level 1)	.69**	.09	.66**	.08	.65**	.08	.24**	.03	.23**	.03	.23**	.03
Workgroup (Intercept, level 2)	.00	.00	.00	.00	.00	.00	.00	.00	.00	.00	.00	.00
Parameter	Self-rated health						Burnout
	Model 1		Model 2		Model 3		Model 1		Model 2		Model 3
	Coefficient	SE	Coefficient	SE	Coefficient	SE	Coefficient	SE	Coefficient	SE	Coefficient	SE
Intercept	3.72**	.39	3.70**	.39	3.73**	.39	3.87**	.38	3.83**	.38	3.85**	.37
Fixed effects
Workgroup job insecurity climate	.18	.13			.15	.14	−.13	.13			−.08	.13
Person-level job insecurity climate	−.19*	.08			−.22*	.09	.20*	.08			.25**	.09
Workgroup job insecurity			.22	.18	.15	.19			−.27	.18	−.23	.19
Person-level job insecurity			−.03	.08	.06	.08			−.01	.08	−.11	.08
Age	.00	.01	.00	.01	.00	.01	.00	.01	.00	.01	.00	.01
Gender (woman)	.26	.15	.25	.16	.29	.15	−.25	.15	−.25	.15	−.30*	.15
Education (university)	−.04	.15	−.04	.15	−.05	.15	−.07	.15	−.05	.15	−.03	.14
Variance components (random effects)
Individual (Residual, level 1)	.61**	.08	.64**	.08	.61**	.08	.57**	.08	.61**	.08	.56**	.08
Workgroup (Intercept, level 2)	.00	.00	.00	.00	.00	.00	.01	.03	.00	.03	.01	.03

N = 126; *p < .05, **p < .01.

Discussion

This study used a multilevel approach for studying the possible relations between both individual job insecurity and job insecurity climate and work- and health-related outcomes using a sample of white-collar workers in a private sector company in Sweden. The first aim of the study was to investigate the role of the workgroup for individual job insecurity and job insecurity climate perceptions. The finding that workgroup membership accounted for 19% of the variance in job insecurity climate perceptions and none of the variance in individual job insecurity perceptions gives an indication that the individual job insecurity and job insecurity climate constructs are both conceptually and empirically different. This finding, together with the relatively low correlation between individual job insecurity and job insecurity climate, also suggests that the respondents differentiate between insecurity regarding their own job, and the job insecurity climate at work. Previous multilevel studies have also reported relatively low ICC values for individual job insecurity: In a sample of Finnish employees, the ICC was 6% (Mauno et al., 2013), and likewise a 6% ICC was reported in two other samples, consisting of Spanish employees (Sora et al., 2009, 2013). An exception to these is the 27% ICC found in a Belgian sample (Sora et al., 2013). These generally low ICC levels suggest that group membership is not a strong explanation for why individual job insecurity varies. Our measure of job insecurity climate seems to capture some aspect of job insecurity climate which aggregated individual job insecurity does not. This finding, along with recommendations by Mauno et al. (2013), also suggests that future research on job insecurity climate should take the referent-shift approach (cf. Chan, 1998) into consideration by using climate measures with an organizational referent. The measure of job insecurity climate used in the present study is one way of doing this, and our results regarding intra-group agreement and the proportion of variance attributed to group membership suggest that job insecurity climate has a relatively large social component, further supporting the differentiation between individual job insecurity and job insecurity climate.

The second aim of the study was to investigate the roles of shared perceptions of individual job insecurity and job insecurity climate for work- and health-related outcomes. Based on previous research on aggregated individual job insecurity, along with transactional stress theory, we expected both individual job insecurity and job insecurity climate to be negatively related to job satisfaction, productivity, self-rated health, and positively related to burnout, but the results only partially supported this. More specifically, person-level job insecurity climate perceptions were found to have a small but significant effect on both self-rated health and burnout, indicating that those who perceive the greatest job insecurity climate within their workgroup also tend to report lower wellbeing. The fact that the individuals’ job insecurity climate perceptions relative to their workgroup (i.e., person-level job insecurity climate) were the only salient predictor was therefore a surprising finding, and the lack of associations between the remaining predictors and outcomes was contrary to our expectations. In a study on the relative importance of person- and organizational-level perceptions in predicting individual work-related attitudes, Schulte et al. (2006) showed that aggregated climate perceptions contributed to a small but significant increase in explained variance in job satisfaction. We expected to find a similar pattern in the present study. One possible explanation for the lack of associations between individual job insecurity and the outcomes could be that the mean levels and variance of individual job insecurity, and perhaps also job insecurity climate, were too low to have an impact on work-related attitudes and health, potentially indicating that job insecurity was not a great concern for these employees. As Sora et al. (2010b) point out, other studies have reported non-significant relations between individual job insecurity and typical outcomes (see e.g., Näswall et al., 2012). The lack of associations between individual job insecurity and outcomes in this study could also be related to the relatively low number of groups in the sample, which may have lowered the power to detect significant relationships in this particular sample.

The type of organization or sector may matter for the associations between job insecurity and outcomes. It could be that employees in the IT sector generally perceive a high level of employability (see e.g., Pio et al., 2013), in which case we would expect employability to buffer the effects of job insecurity on outcomes (see e.g., Green, 2011). However, as employability was not included in this study, this would be a question for future research. It could also be that the relatively short organizational tenure for this sample (mean = 4.4 years, SD = 1.5) is an indicator of a high job mobility culture in the IT sector, which might make job insecurity less of a concern to the respondents. Future replication studies using samples from other areas would assist in determining the generalizability of the results in the present study.

Methodological considerations and implications for future research

Since the present study had a cross-sectional design, we cannot draw any conclusions about cause and effect. In fact, it cannot be ruled out that the directionality of the effects found in this study could be the reverse of what is theorized in this article – one could for instance argue that employees with health problems will experience more job insecurity than their healthy peers. However, given the fairly established evidence regarding the temporal precedence of job insecurity over outcomes (e.g., Hellgren and Sverke, 2003), we consider this to be unlikely. Another consideration that needs to be taken into account when interpreting these results is the small sample size, which has been split into few and relatively small workgroups. While larger samples are always desirable (recommendations for Level 2 sample size range between 20 [Meuleman and Billiet, 2009] and 50 [Maas and Hox, 2005]), the practical constraints placed on the data collection for the present study (size of organization, number of naturally occurring groups, etc.) were such that the effective sample size was quite small. Also, ignoring the nested structure of the data would involve a violation of the requirement of independence in traditional statistical analyses (Bell et al., 2014; Peugh, 2010). And further, in organizations with very large workgroups, there is the risk that informal subgroups could have formed within those groups, making it difficult to identify social units or nested structures that have relevance for employee perceptions.

A potential limitation of the present study is that the item wording of the job insecurity climate scale asks the respondents to report on their ‘workplace.’ As a consequence, the respondents could have reported on a different social entity than what we used as a grouping variable for this study: employees with the same manager. This could have influenced the results through the effect sizes potentially being lower than they really are. On the other hand, one could also argue that employees usually belong to multiple social entities at their workplace (e.g., based on vocation, hierarchical position, working proximity, tenure), and that no matter which of the possible grouping variables is chosen, some potential structures which may account for agreement will nevertheless be overlooked. By asking respondents to report on their workplace, one could reason that they will think of the more salient co-worker segment of their work environment.

Conclusion

Despite the abovementioned limitations, this study found that person-level job insecurity climate perceptions were associated with lower self-rated health and higher levels of burnout. Thus, the results imply that job insecurity climate perceptions matter for the individual; working in a social environment characterized by a job insecurity climate may have a negative impact on individuals’ health and wellbeing. The finding that individual job insecurity was not associated with any of the outcomes was unexpected. In addition, the results of this study imply that unlike individual job insecurity, job insecurity climate has a social component. Further, since employees’ ratings of the climate are in part dependent on workgroup membership, how job insecurity climate is measured matters, suggesting that it could be worthwhile to include the referent-shift approach in future studies on job insecurity.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by grants from the Swedish Research Council for Health, Working Life and Welfare (FORTE). The first author was supported by Stockholm Stress Center (SSC). All authors are affiliated with SSC and this research was carried out as a joint collaboration within the center.

Author biographies

Lena Låstad is researcher and lecturer at the Division of Work and Organizational Psychology at the Department of Psychology, Stockholm University, Sweden. Her current research interests include job insecurity, work-related stress, social climates, and organizational culture.

Katharina Näswall is Associate Professor at the Department of Psychology, University of Canterbury, New Zealand. Her research focuses on work-related stress and wellbeing, and factors which aid coping with work-related stress. She also investigates organizational and employee resilience, with a special focus on measuring employee resilience and organizational factors which enhance employee wellbeing and resilience during turbulent times.

Erik Berntson is researcher and senior lecturer in the Department of Psychology, Stockholm University, Sweden. His research focuses on working conditions for managers, specifically for public sector managers. Other research interests include labor market mobility with a special focus on employability, organizational changes, management, and work environment issues.

Aram Seddigh has a PhD in psychology, and is a licensed psychologist and economist. He is affiliated to both the Stress Research Institute at Stockholm University and Stockholm School of Economics. His work concerns how new ways of working affect employees’ health and performance.

Magnus Sverke is Professor and Chair of the Division of Work and Organizational Psychology at the Department of Psychology, Stockholm University, Sweden. He is also Extraordinary Professor at the Faculty of Economic and Management Sciences, North-West University, South Africa. His current research interests include organizational change and its effects on employees, job insecurity, workplace wellbeing, and individualized pay.

References

Ahola

Honkonen

Isometsä

. (2006) Burnout in the general population. Social Psychiatry and Psychiatric Epidemiology 41(1): 11–17.

Allison

(2001) Missing Data (Quantitative Applications in the Social Sciences, Vol. 136). Thousand Oaks, CA: Sage.

Baltes

Zhdanova

Parker

(2009) Psychological climate: A comparison of organizational and individual level referents. Human Relations 62(5): 669–700.

Bell

Morgan

Schoeneberger

. (2014) How low can you go? An investigation of the influence of sample size and model complexity on point and interval estimates in two-level linear models. Methodology 10(1): 1–11.

Brayfield

Rothe

(1951) An index of job satisfaction. Journal of Applied Psychology 35(5): 307–311.

Burchell

(2009) Flexicurity as a moderator of the relationship between job insecurity and psychological well-being. Cambridge Journal of Regions, Economy and Society 2(3): 365–378.

Burgard

Brand

House

(2009) Perceived job insecurity and worker health in the United States. Social Science and Medicine 69(5): 777–785.

Chan

(1998) Functional relations among constructs in the same content domain at different levels of analysis: A typology of composition models. Journal of Applied Psychology 83(2): 234–246.

Cheng

GHL

Chan

DKS

(2008) Who suffers more from job insecurity? A meta-analytic review. Applied Psychology 57(2): 272–303.

10.

De Bustillo

De Pedraza

(2010) Determinants of job insecurity in five European countries. European Journal of Industrial Relations 16(1): 5–20.

11.

De Cuyper

Sora

De Witte

. (2009) Organizations’ use of temporary employment and a climate of job insecurity among Belgian and Spanish permanent workers. Economic and Industrial Democracy 30(4): 564–591.

12.

De Witte

(1999) Job insecurity and psychological well-being: Review of the literature and exploration of some unresolved issues. European Journal of Work and Organizational Psychology 8(2): 155–177.

13.

Green

(2011) Unpacking the misery multiplier: How employability modifies the impacts of unemployment and job insecurity on life satisfaction and mental health. Journal of Health Economics 30(2): 265–276.

14.

Greenhalgh

Rosenblatt

(1984) Job insecurity: Toward conceptual clarity. Academy of Management Review 9(3): 438–448.

15.

Heery

ESJ

(2000) The Insecure Workforce. London: Routledge.

16.

Hellgren

Sverke

(2003) Does job insecurity lead to impaired well-being or vice versa? Estimation of cross-lagged effects using latent variable modelling. Journal of Organizational Behavior 24(2): 215–236.

17.

Hellgren

Sjöberg

Sverke

(1997) Intention to quit: Effects of job satisfaction and job perceptions. In: Avalone

Arnold

De Witte

(eds) Feelings Work in Europe. Milan: Guerini, pp. 415–423.

18.

Hellgren

Sverke

Isaksson

(1999) A two-dimensional approach to job insecurity: Consequences for employee attitudes and well-being. European Journal of Work and Organizational Psychology 8(2): 179–195.

19.

Hox

(2002) Multilevel Analysis: Techniques and Applications. New York: Routledge.

20.

Schaufeli

(2011) Job insecurity and remuneration in Chinese family-owned business workers. Career Development International 16(1): 6–19.

21.

Huang

Niu

Lee

. (2012) Differentiating cognitive and affective job insecurity: Antecedents and outcomes. Journal of Organizational Behavior 33(6): 752–769.

22.

Idler

(2003) Discussion: Gender differences in self-rated health, in mortality, and in the relationship between the two. The Gerontologist 43(3): 372–375.

23.

Idler

Benyamini

(1997) Self-rated health and mortality: A review of twenty-seven community studies. Journal of Health and Social Behavior 38(1): 21–37.

24.

ILO (International Labour Office) (2015) World Employment Social Outlook: The Changing Nature of Jobs. Geneva: ILO. Available at: www.ilo.org/wcmsp5/groups/public/—dgreports/—dcomm/—publ/documents/publication/wcms_368626.pdf.

25.

Jahoda

(1982) Employment and Unemployment: A Social-psychological Analysis. Cambridge: Cambridge University Press.

26.

Kalil

Ziol-Guest

Hawkley

. (2009) Job insecurity and change over time in health among older men and women. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences 65B(1): 81–90.

27.

Kinnunen

Mauno

Nätti

. (1999) Perceived job insecurity: A longitudinal study among Finnish employees. European Journal of Work and Organizational Psychology 8(2): 243–260.

28.

Kinnunen

Mauno

Nätti

. (2000) Organizational antecedents and outcomes of job insecurity: A longitudinal study in three organizations in Finland. Journal of Organizational Behavior 21(4): 443–459.

29.

König

Probst

Staffen

. (2011) A Swiss-US comparison of the correlates of job insecurity. Applied Psychology 60(1): 141–159.

30.

Kooij

De Lange

Jansen

. (2011) Age and work-related motives: Results of a meta-analysis. Journal of Organizational Behavior 32(2): 197–225.

31.

Kozlowski

Hattrup

(1992) A disagreement about within-group agreement: Disentangling issues of consistency versus consensus. Journal of Applied Psychology 77(2): 161–167.

32.

Länsisalmi

Peiró

Kivimäki

(2000) Collective stress and coping in the context of organizational culture. European Journal of Work and Organizational Psychology 9(4): 527–559.

33.

Låstad

Berntson

Näswall

. (2015) Measuring quantitative and qualitative aspects of the job insecurity climate: Scale validation. Career Development International 20(3): 202–217.

34.

László

Pikhart

Kopp

. (2010) Job insecurity and health: A study of 16 European countries. Social Science and Medicine 70(6): 867–874.

35.

Lazarus

Folkman

(1984) Stress, Appraisal, and Coping. New York: Springer.

36.

LeBreton

Senter

(2008) Answers to 20 questions about interrater reliability and interrater agreement. Organizational Research Methods 11(4): 815–852.

37.

Lindell

Brandt

Whitney

(1999) A revised index of interrater agreement for multi-item ratings of a single target. Applied Psychological Measurement 23(2): 127–135.

38.

Locke

(1969) What is job satisfaction? Organizational Behavior and Human Performance 4(4): 309–336.

39.

Maas

CJM

Hox

(2005) Sufficient sample sizes for multilevel modeling. Methodology 1(3): 86–92.

40.

Maslach

Jackson

(1981) The measurement of experienced burnout. Journal of Occupational Behavior 2(2): 99–113.

41.

Mauno

De Cuyper

Tolvanen

. (2013) Occupational well-being as a mediator between job insecurity and turnover intention: Findings at the individual and work department levels. European Journal of Work and Organizational Psychology 23(3): 381–393.

42.

Mauno

Kinnunen

Mäkikangas

. (2005) Psychological consequences of fixed-term employment and perceived job insecurity among health care staff. European Journal of Work and Organizational Psychology 14(3): 209–237.

43.

Meuleman

Billiet

(2009) A Monte Carlo sample size study: How many countries are needed for accurate multilevel SEM? Survey Research Methods 3(1): 45–58.

44.

Näswall

De Witte

(2003) Who feels insecure in Europe? Predicting job insecurity from background variables. Economic and Industrial Democracy 24(2): 189–215.

45.

Näswall

Lindfors

Sverke

(2012) Job insecurity as a predictor of physiological indicators of health in healthy working women: An extension of previous research. Stress and Health 28(3): 255–263.

46.

Näswall

Sverke

Hellgren

(2005) The moderating role of personality characteristics on the relationship between job insecurity and strain. Work and Stress 19(1): 37–49.

47.

Parker

Baltes

Young

. (2003) Relationships between psychological climate perceptions and work outcomes: A meta-analytic review. Journal of Organizational Behavior 24(4): 389–416.

48.

Peugh

(2010) A practical guide to multilevel modeling. Journal of School Psychology 48(1): 85–112.

49.

Pio

Syed

Srinivasan

. (2013) Career persistence of women software professionals in India. Gender in Management: An International Journal 28(4): 210–227.

50.

Probst

(2002) Layoffs and tradeoffs: Production, quality, and safety demands under the threat of job loss. Journal of Occupational Health Psychology 7(3): 211–220.

51.

Rosenblatt

Talmud

Ruvio

(1999) A gender-based framework of the experience of job insecurity and its effects on work attitudes. European Journal of Work and Organizational Psychology 8(2): 197–217.

52.

Schaufeli

Leiter

Maslach

. (1996) MBI-General Survey. In: Maslach

Jackson

Leiter

(eds) Maslach Burnout Inventory Manual, 3rd edn. Palo Alto, CA: Consulting Psychologists Press.

53.

Schreurs

Van Emmerik

Notelaers

. (2010) Job insecurity and employee health: The buffering potential of job control and job self-efficacy. Work and Stress 24(1): 56–72.

54.

Schulte

Ostroff

Kinicki

(2006) Organizational climate systems and psychological climate perceptions: A cross-level study of climate–satisfaction relationships. Journal of Occupational and Organizational Psychology 79(4): 645–671.

55.

Schutte

Toppinen

Kalimo

. (2000) The factorial validity of the Maslach Burnout Inventory-General Survey (MBI-GS) across occupational groups and nations. Journal of Occupational and Organizational Psychology 73(1): 53–66.

56.

Silla

De Cuyper

Gracia

. (2009) Job insecurity and well-being: Moderation by employability. Journal of Happiness Studies 10(6): 739–751.

57.

Sora

Caballer

Peiró

. (2009) Job insecurity climate’s influence on employees’ job attitudes: Evidence from two European countries. European Journal of Work and Organizational Psychology 18(2): 125–147.

58.

Sora

Caballer

Peiró

(2010a) The consequences of job insecurity for employees: The moderator role of job dependence. International Labour Review 149(1): 59–72.

59.

Sora

Caballer

Peiró

. (2010b) Moderating influence of organizational justice on the relationship between job insecurity and its outcomes: A multilevel analysis. Economic and Industrial Democracy 31(4): 613–637.

60.

Sora

De Cuyper

Caballer

. (2013) Outcomes of job insecurity climate: The role of climate strength. Applied Psychology: An International Review 62(3): 382–405.

61.

Staufenbiel

König

(2010) A model for the effects of job insecurity on performance, turnover intention, and absenteeism. Journal of Occupational and Organizational Psychology 83(1): 101–117.

62.

Størseth

(2006) Changes at work and employee reactions: Organizational elements, job insecurity, and short-term stress as predictors for employee health and safety. Scandinavian Journal of Psychology 47(6): 541–550.

63.

Sverke

Hellgren

Näswall

(2002) No security: A meta-analysis and review of job insecurity and its consequences. Journal of Occupational Health Psychology 7(3): 242–264.

64.

Tilakdharee

Ramidial

Parumasur

(2010) The relationship between job insecurity and burnout. South African Journal of Economic and Management Sciences 13(3): 254–271.

65.

Wagner

Rau

Lindemann

(2010) Multiple informant methodology: A critical review and recommendations. Sociological Methods and Research 38(4): 582–618.