Are core self-evaluations a suitable moderator in stressor-detachment relationships? A study among managers’ perceived job demands,detachment and strain reactions

Abstract

BACKGROUND:

The relationships among job demands, personality factors, recovery and psychological health receive increasing attention but are not well understood

OBJECTIVE:

Therefore, the present study tests moderating effects among a sample of managers as proposed by the stressor-detachment model. We aimed to determine whether core self-evaluations (CSE) had an influence on the correlations between detachment and strain reactions (depressive symptoms, irritation, exhaustion) and between job demands and detachment. Further, we tested whether detachment attenuates the positive relation between job demands and strain reactions.

METHODS:

A convenience sample of managers in three German settings (N = 282) participated in the cross-sectional study.

RESULTS:

Results based on hierarchical regression analysis showed that high CSE significantly weakened the negative relationship between detachment and depressive symptoms in this sample. However, CSE did not moderate the negative relationship between job demands and detachment. Moreover, results revealed that detachment moderated the positive relation between job demands and exhaustion.

CONCLUSIONS:

The authors tested whether CSE was able to moderate the relationship between job demands, psychological detachment and different stress reactions. Although we found a significant interaction effect, CSE may be too distal to moderate all respective associations.

Keywords

Psychological detachment recovery stress psychosocial factors depression

1 Introduction

Today, work in organizations can be very demanding for managers as well as for employees [2]. This is especially the case for managers, since managerial positions are often associated with a high workload and long working hours [2]. In line with the stressor-detachment model [3], the high job demands which managers face may be associated with various negative health impacts. These can range from stressful short-term experiences and a poor sense of well-being to the development of severe strain reactions such as depressive symptoms. In this context, the stressor-detachment model emphasizes that rather the sustained activation of an individual than the acute stress reaction is crucial for manager’s well-being. Lack of psychological detachment from work during nonwork time is related to enhanced strain reactions and poor well-being since the negative impact of job demands on wellbeing cannot successfully be attenuated by psychological detachment [4, 5]. In the present study, we focused on three different strain reactions: Irritation, emotional exhaustion and depressive symptoms. Irritation is a sensitive indicator for well-being, and ideally suited to describe slight deviations from normal well-being [6]. Emotional exhaustion is a negative emotional reaction resulting from prolonged exposure to high job demands [7]. Depressive symptoms are mainly characterized by negative affective states and lack of interest persisting over the course of two or more weeks. Moreover, depression is viewed as a leading cause of psychiatric morbidity, and increasing prevalence rates confirm its importance for society [8]. Consequently, all of these outcomes are associated with increased health care costs for both society and organizations [9].

It is crucial that managers be in optimal physical and psychological states of health to meet their requirements in an increasingly demanding environment. Recovery from work, a process in which physical and psychological strain is reduced by unwinding and recuperating from work, is one way to achieve this. Recovery experiences such as psychological detachment from work that is, not thinking of one’s work during nonwork time, or relaxation have been increasingly studied in this regard. Results showed that those experiences during leisure time predicted strain reactions. Specifically, psychological detachment from work has been viewed as a powerful recovery experience and shows closer relationships with wellbeing than other recovery experiences [10].

However, individuals may differ in the perceived need for recovery, both due to their coping behavior with certain stressors at work, as well as in the way they appraise work-related experiences at work or at home. For example, if one is only able to recover poorly due to high job demands, this could have a greater negative impact on an individual who tends to interpret things in a negative way than on an individual who appraises experiences in a consistently positive manner [11]. Thus, one could argue that the relationship between detachment and strain reactions is dependent on several individual factors. For instance, the higher order construct of core self-evaluation (CSE) has been shown to be associated with stress-related health outcomes, coping behavior and performance [12]. All of these aspects are essential to successful management in highly demanding working environments. CSE is constituted by the shared variance of four traits: self-esteem, generalized self-efficacy, locus of control, and neuroticism [13].

In the present study, we focused on CSE since it captures much of the overlap of its trait measures. Aggregated measures such as CSE substantially increase predictive validity of personality variables. As a result, CSE may be able to predict outcomes better or at least more consistently than each trait measure alone [14].

The arguments above are in line with a call for systematic research into possible moderator variables. Thus, research should also address the influence of individual-level factors in the relationship between job demands, detachment, strain and well-being. Therefore, in the present study we examined interaction effects proposed by the stressor-detachment model [3].

Specifically, we tested whether (a) CSE moderates the negative correlation between perceived job demands and psychological detachment, (b) CSE moderates the negative correlation between detachment and strain reactions and (c) detachment moderates the positive correlation between perceived job demands and strain reactions. The latter might influence the results of (a) and (b) since individuals high in CSE are thought to perceive or create situations that are less stressful, and managers with lower levels of perceived job demands might have a lower need for detachment [15].

2 Theoretical background

2.1 Psychological detachment

Psychological detachment is one way of restoring low mood induced by stressful job demands; it was found to be negatively associated with strain reactions such as depressive symptoms or exhaustion [3, 16]. Psychological detachment can be seen as a state in which individuals are not cognitively, emotionally, or behaviorally engaged in work-related activities. In this regard, psychological detachment occurs by being mentally involved in any other content area at home, for example by pursuing a hobby [3].

2.2 The stressor-detachment model

The stressor-detachment model suggests that job demands impair psychological detachment during leisure time and that the sustained activation that results from high job demands could lead to an impaired health state. Negative activation initialized by high job demands makes it difficult to mentally disengage from work. Specifically, managers may stay occupied with work-related tasks trying to solve work-related problems, or continue to worry about job demands. For example, there is strong evidence that work intensity and emotional demands are negatively associated with detachment [3]. Therefore, our study focused on perceived work intensity and emotional demands. Moreover, we aimed to broaden the scope of perceived job demands and included unpredictability of tasks [17] and work-home-family conflicts [18].

Additionally, the stressor-detachment model proposes that psychological detachment may buffer the positive correlation between job demands and strain reactions. If there is no detachment from work during leisure time, job demands remain mentally present. This may enhance strain reactions, such as depressive symptoms, exhaustion or irritation. However, if one can detach during leisure time, psychological resources can be restored leading to reduced strain [3].

H1: Detachment negatively moderates the positive correlation between job demands and strain reactions, such that individuals with higher levels of detachment after work experience less strain and enjoy better well-being when job demands are high than individuals low in detachment.

However, job demands might not always impair psychological detachment. For example, when job demands are not appraised as threatening or when they are overcome immediately due to successful coping processes [19], they could lose their health-related relevance for an individual. The stressor-detachment model proposes that an individual’s attention and job or personal resources may moderate the association between job demands and detachment, whereas the content of job-related related thoughts should moderate the correlation of detachment and strain reactions. CSE is a higher order-factor constituted by several personal resources [13]. Therefore, CSE may moderate the associations hypothesized in the stressor-detachment model.

Thus, there are at least two explanations why individuals high in CSE may have less need to recover. First, it is possible that individuals high in CSE may appraise their experiences at work differently. It could be that individuals high in CSE cope more effectively with high job demands or use more problem-oriented coping behaviors which reduces strain [15, 20]. Therefore, managers high in CSE are better able to cope with the demands. As a result, these individuals can better detach during non-work time since they are able to stop thinking about the demands at work. Additionally, individuals high in CSE may pay more attention to positive aspects of their jobs [12]. Thus, managers high in CSE may appraise their job demands as less stressful since the demands are tied to the nature of a managerial position [21]. On the contrary, individuals low in CSE may feel overwhelmed by the demands, which could lead to poorer detachment during non-work time.

H2: CSE negatively moderates the correlation between job demands and psychological detachment such that individuals high in CSE experience higher levels of detachment from work when job demands are high than individuals low in CSE.

Individuals high in CSE may appraise their job demands and work-related experiences differently when thinking about them while at home. Research findings suggest that thinking about one’s job during non-work time can also be beneficial, if the content is positive [3]. For example, Meier, Cho [22] showed that positive work reflection is associated with an increase in well-being and, moreover, that work reflection has incremental validity above psychological detachment and the absence of negative work reflection. Chang, Ferris [12] pointed out that employees high in CSE appraise situations positively, which has an impact on both primary and secondary appraisals with regards to the transactional stress model [19]. Given that individuals high in CSE (re)appraise situations in a consistently positive manner, it seems likely that those individuals may think and talk positively about their jobs, which could reduce strain even when they do not detach.

H3: CSE positively moderates the negative relation between psychological detachment and strain reactions such that individuals high in CSE experience less strain and enjoy better well-being when detachment is low than individuals low in CSE.

3 Methods

3.1 Participants and procedure

Study participants were recruited from training facilities of the social and health services (sample A), a leaders association (sample B) and a pharmaceutical and law firm (sample C). In all samples, we distributed questionnaires including a declaration of consent and assured the participant’s anonymity. The participants in sample A sent their completed questionnaires directly to the research group in a pre-stamped envelope. In return for the participation in the study, we offered a summary report for each participant involved. A total of 408 questionnaires were distributed and 161 returned, yielding a response rate of 39.5%. In sample B, 70 managers participated who had subscribed to an internal newsletter of a leaders association representing managers on the federal political scene in Germany and Europe. Managers received an invitation via e-mail to fill out an online questionnaire. However, the calculation of a meaningful response rate was not possible in this case. Participants in sample C sent their completed questionnaires directly to the research group. A total of 51 managers of a pharmaceutical and a law firm participated, yielding a response rate of 65%.

The final study sample consisted of 282 managers. Half of the participants were women (54.4%). Managers were between 25 and 70 years old (M = 47.2). The sample was fairly well educated since 51% held a university degree. 95% had managerial responsibility whereas 5% had also project, process or product responsibility. A total of 15.80% of the participants worked in a top management position (e.g., managing director or CEO). The majority of the sample held mid-level management positions such as divisional managers (32.39%) or as a department manager (31.97). A total of 19.80% of the participants were group managers.

Some 43% had responsibility for between 1 and 19 employees, 21.4% between 21 and 30 employees, 8.2% between 31 and 40 employees and 20.6% for more than 40 employees, 61% of the participants worked full-time, and 39% part-time. The total sample had fairly long weekly working hours (M = 47.07; SD = 8.0). The majority of the sample (60.1%) worked in the personal services sector. The remaining managers worked in the commercial sector (15.7%), IT- and sciences services sectors and other services (5.2%); 62% worked for private companies, 38% in the public sector. Thus, it is fair to say that the sample included a broad cross-section of the managers in the participating institutions. It should be noted that the study sample was also used in another publication, with different variables, research questions and hypotheses 1 [23]. Ethical approval for the study was obtained by the German Psychological Society.

3.2 Measures

Depressive symptoms were measured by the nine-item German version of the Patient Health Questionnaire (PHQ-9) [24]. Responses to such items as “Little interest or pleasure in doing things” were rated on a 4-point scale where 0 = not at all to 3 = every day. Total scores ranged from 0–27.

Irritation was assessed by the eight-item scale developed by [6]. Responses to such items as “I get grumpy when others approach me” ranged from 1 = strongly disagree to 7 = strongly agree.

Exhaustion was assessed with five items from the German version [25] of the Maslach Burnout Inventory– General Survey [26]. For example “I feel used up at the end of a work day”. Total scores can range between 1 = never to 6 = always.

Core Self Evaluation (CSE) was measured with the 12 item German version [13] which included items such as “I am confident to get the success I deserve in life”. Items were scored on a five-point Likert scale where 1 = strongly disagree to 5 = strongly agree.

Psychological detachment was measured by using the Recovery Experience Questionnaire [10]. The scale included four items (e.g. “I distance myself from my work”). Participants were asked to respond with respect to their leisure time after work. Items were scored on a 5-point scale where 1 = totally disagree and 5 = totally agree.

Job demands: The German “Fragebogen zum Erleben von Intensität und Tätigkeitsspielraum in der Arbeit” questionnaire [27] was used to assess work intensity. There are six items such as: “At this work, I have to do many things at the same time.” Items were scored on a 4-point scale where 1 = no to 4 = yes. The German version [28] of the Copenhagen Psychosocial Questionnaire [29] was used to assess emotional demands (three items, e.g., “Is your work emotionally

demanding?”), unpredictability (two items, e.g., “Do you receive all the information you need in order to do your work well?” and work-home-family conflicts (five items, e.g., “The demands of my work interfere with my home and family life.” Scores can range from 0 to 100).

Demographic variables. Gender, age, educational status, type of contract (full or part time), weekly working hours and branches of industry as demographic variables were recorded. Moreover, we asked for management position, level and the number of assigned employees to determine the degree of responsibility, which was deemed appropriate given the possible relationship to strain reactions [2, 30].

3.3 Statistical analysis

T-tests were used to compare the depression-score of the present sample with a representative German sample [31, 32].

To test the hypotheses, we used hierarchical regression analyses. To conduct a concise but valid analysis, the four specific job demands were incorporated into one factor “job demands”. To justify this operationalization, we performed a confirmatory factor analysis (CFA) and monitored the fit indices (Tucker-Lewer-Index [TLI], Comparative Fit Index [CFI], Root Mean Square Error of Approximation [RMSEA] with the 90% Confidence Interval around the RMSEA and Standardized Root Mean Square Residual [SRMR] of this factor. Values for these analyses are displayed in the results section.

First, we centered all independent variables [33]. The predicted two-way interaction was then tested with hierarchical regression analyses. The independent variable and the moderator variable were included in the first step of the regression equation. The interaction term between both variables was included in the second step of the regression. Through this approach, we examined the extent to which the interaction term contributed significantly to the explanation of the variance in strain reactions, after controlling for the main effects. In the last step, control variables were entered [34]. We used R for statistical analysis [35].

3.4 Control variables

Knowing that gender differences are associated with depressive symptoms, we decided to control for the sex of the participants [36]. Moreover, we tested for age, branch, type of contract, weekly working hours, management level, the number of assigned employees [30] and the organization (sample A, B, C) since those factors may have an influence on the respective relationships. However, from the control variables tested, only weekly working hours significantly predicted irritation and detachment. Accordingly, this variable was controlled throughout the study. No control variable significantly predicted exhaustion.

4 Results

Descriptive statistics, zero-order correlations and Cronbach’s α are provided in Table 1.

Table 1
Descriptive statistics, zero-order correlations and Cronbach’s α

Skala M SD 1 2 3 4 5 6 7 8 9

1. Depressive symptoms 5.81 4.05 (0.84)

2. Exhaustion 3.45 0.97 0.63^*** (0.89)

3. Irritation 3.43 1.25 0.56^* 0.63^* (0.92)

4. Psychological detachment 3.05 0.86 –0.36^* –0.39^* –0.66^*** (0.91)

5. CSE 3.80 0.52 –0.69^* –0.52^* –0.52^* 0.38^* (0.85)

6. Work intensity 3.04 0.50 0.33^* 0.49^* 0.36^* –0.24^* –0.21^*** (0.77)

7. Emotional demands 63.21 18.55 0.34^* 0.40^* 0.29^* –22^* –0.24^* 0.32^* (0.78)

8. Work-home-family conflicts 52.81 24.14 0.27^* 0.42^* 0.39^* –0.33^* –0.22^* 0.43^* 0.33^*** (0.91)

9. Unpredictability 41.00 21.88 0.05 0.08 0.15^ –0.10 –0.09 0.05 –0.28^* 0.10 (0.80)

Skala	M	SD	1	2	3	4	5	6	7	8	9
1. Depressive symptoms	5.81	4.05	(0.84)
2. Exhaustion	3.45	0.97	0.63^***	(0.89)
3. Irritation	3.43	1.25	0.56^***	0.63^***	(0.92)
4. Psychological detachment	3.05	0.86	–0.36^***	–0.39^***	–0.66^***	(0.91)
5. CSE	3.80	0.52	–0.69^***	–0.52^***	–0.52^***	0.38^***	(0.85)
6. Work intensity	3.04	0.50	0.33^***	0.49^***	0.36^***	–0.24^***	–0.21^***	(0.77)
7. Emotional demands	63.21	18.55	0.34^***	0.40^***	0.29^***	–22^***	–0.24^***	0.32^***	(0.78)
8. Work-home-family conflicts	52.81	24.14	0.27^***	0.42^***	0.39^***	–0.33^***	–0.22^***	0.43^***	0.33^***	(0.91)
9. Unpredictability	41.00	21.88	0.05	0.08	0.15^**	–0.10	–0.09	0.05	–0.28^***	0.10	(0.80)

Notes: N = 274–282; **p < 0.01; ***p < 0.001; Cronbach’s α are shown in brackets on the diagonal.

All constructs had appropriate reliability. Correlations were as expected and low to medium in size. The depression score (M = 5.81, SD = 4.05) was significantly (t(2712) = 14.25, p < 0.001) higher when compared to a representative sample of the German general population.

To test whether detachment buffers the positive relationship between job demands and strain reactions (H1), we incorporated the four job demands into one general factor “job demands”. The CFA revealed satisfactory fit to the data for the factor (CFI = 0.944, TLI = 0.923, RMSEA = 0.070 [90% CI_Lo –90% CI_Up = 0.06–0.07], SRMR = 0.060) and hence supported further analyses. Furthermore, we tested in an additional analysis a one-factor model (meaning that all variables under study load on the same single latent factor) as a diagnostic tool for the presence of a common method bias. However, the one-factor model compared to the eight-factor model showed significantly poorer fit, indicating that the measures do not load on a single factor (Δχ ² (20) = 109.01, p < 0.001). Results of hierarchical multiple regression analyses can be found in Table 2. Psychological detachment buffered the positive relation between job demands and exhaustion since the incorporation of an interaction effect between detachment and job demands revealed a significant increase in R² (ΔR² = 0.02, F(1, 276) = 3.96, p < 0.05) (see Fig. 1). No significant moderation was found with regard to depressive symptoms or irritation. Thus, H1 is partially supported.

Table 2

Strain reactions regressed on psychological detachment and job demands

Dependent variable:		Depressive symptoms (N = 274)				Exhaustion (N = 280)				Irritation (N = 279)
Step	Independent variables	ß	R²	Δ R²	F	ß	R²	Δ R²	F	ß	R²	Δ R²	F
1	Psychological detachment	–0.29^***				–0.28^***				–0.61^***
	Job demands	0.24^***	0.17			0.37^***	0.26			0.17^***	0.45
2	Psychological detachment	–0.29^***				–0.29^***				–0.61^***
	Job demands	0.23^***				0.39^***				0.19^***
	Psychological detachment *	0.00	0.17	0.00	0.11	0.01^*	0.28	0.02	3.96^*	0.00	0.45	0.00	0.06
	job demands
3	Control variable	0.16^**				–				–0.02
	Psychological detachment	–0.30^***				–				–0.60^***
	Job demands	0.20^***				–				0.19^***
	Psychological detachment *	0.00	0.18	0.01	8.22^**	–	–	–	–	0.00	0.45	0.00	0.18
	job demands

Fig.1

CSE buffers the relation between detachment and depressive symptoms and detachment buffers the relation between job demands and exhaustion.

Next, we tested hypothesis H2 and examined whether CSE buffered the relationship between job demands and detachment (see Table 3). In the second step of the regression analysis, the interaction term between CSE and job demands was entered. However, it did not explain significantly more variance in detachment indicating that CSE was not a significant moderator in the relation between job demands and detachment. Thus, H2 is rejected.

Table 3

Psychological detachment regressed on job demands and core self-evaluations

Dependent variable:		Psychological detachment (N = 279)
Step	Independent variables	ß	R²	Δ R²	F
1	Job demands	–0.22^***
	CSE	0.33^***	0.19
2	Job demands	–0.22^***
	CSE	0.32^***
	Job demands * CSE	–0.08	0.19	0.00	0.55
3	Weekly working hours	–0.22^***
	Job demands	–0.21^***
	CSE	0.36^***
	Job demands * CSE	–0.05	0.24	0.05	18.90^***

Notes: Predictors have been centered; ß= standardized ß-coefficients; R² = explanation rate; Δ R² = change in explanation rate; F = F test is based on the difference between the residual sums of squares for the two models, on the difference in degrees of freedom for error and on the estimate of the error variance from the larger model.

To test the hypothesis H3 that CSE moderates the negative relation between detachment and strain reactions, we followed the same procedure (see Table 4). The incorporation of an interaction effect between psychological detachment and CSE on depressive symptoms revealed a significant increase in R² (ΔR² = 0.02, F(1, 270) = 5.68, p < 0.05).

Table 4

Strain reactions regressed on psychological detachment and core self-evaluations

Dependent variable:		Depressive symptoms (N = 274)				Exhaustion (N = 278)				Irritation (N = 279)
Step	Independent variables	ß	R²	Δ R²	F	ß	R²	Δ R²	F	ß	R²	Δ R²	F
1	Psychological detachment	–0.09^*				–0.21^***				–0.53^***
	CSE	–0.65^***	0.47			–0.42^***	0.31			–0.30^***	0.52
2	Psychological detachment	–0.11^*				–0.21^***				–0.53^***
	CSE	–0.62^***				–0.42^***				–0.31^***
	Psychological detachment * CSE	0.18^*	0.49	0.02	5.70^*	–0.00	0.31	0.00	0.00	0.06	0.52	0	0.58
3	Control variable	0.11^**				–				0.03
	Psychological detachment	–0.10^*				–				–0.52^***
	CSE	–0.62^***				–				–0.32^***
	Psychological detachment * CSE	0.11^*	0.50	0.01	8.42^**	–	–	–	–	0.09	0.52	0	1.05

Notes: Predictors have been centered; ß= standardized ß-coefficients; R² = explanation rate; Δ R² = change in explanation rate; Control variables: sex for depressive symptoms and weekly working hours for irritation; F = F test is based on the difference between the residual sums of squares for the two models, on the difference in degrees of freedom for error and on the estimate of the error variance from the larger model.

Subsequent simple slope analyses revealed significant slopes for the regression of depressive symptoms on psychological detachment at low CSE (25%) for (β= –0.85, t(269) = 3.24, p < 0.001), and at medium CSE (50%) (β= –0.50, t(269) = 2.25, p < 0.05). However, the slope for the regression of depressive symptoms on psychological detachment at high CSE (75%) was not significant (β= –0.20, t(269) = 0.80, p > 0.05) (see Fig. 2). No significant moderation was found with respect to exhaustion and irritation. Thus, H3 received partial support.

Fig.2

CSE moderates the negative relationship between psychological detachment and depressive symptoms.

5 Discussion

The purpose of this study was to examine the moderating role of CSE in (a) the negative correlation between job demands and detachment and (b) the negative correlation between psychological detachment and strain reactions. Moreover, we tested whether (c) detachment moderated the positive relationship between job demands and strain reactions. Based on the stressor-detachment model, hierarchical regression analyses were performed.

With respect to research question (a), analyses showed no significant moderation for CSE, hence H2 was rejected. Regarding research question (b), results revealed that CSE is a buffer for poor psychological detachment on depressive symptoms, such that managers high in CSE had fewer depressive symptoms when detachment was low than individuals low in CSE. No significant interaction was found for either exhaustion or irritation. Therefore, H3 received partial support. With regard to research question (c), results revealed that detachment moderated the positive correlation between job demands and exhaustion. No significant interaction was found for either depressive symptoms or irritation. Thus, H1 was partially supported.

Regarding the association of job demands and detachment, it is reasonable to assume that CSE moderates the negative relationship between job demands and detachment [3, 15]. Surprisingly, CSE did not moderate this relationship in the present study. With respect to the buffering role of CSE in the relationship between job demands and strain reactions, research has revealed mixed findings. Whereas some studies found that CSE buffers the relation between job demands and strain [20], other studies found no moderating effect [15]. In terms of psychological detachment as an outcome, however, no other studies regarding the moderating role of CSE have been published as of writing. Interaction effects are often hard to detect [3]. Therefore, one might speculate that this finding may be due to methodological issues since regression analysis is a rather conservative approach for testing interaction effects [37]. Moreover, CSE might be a construct that is too distal to moderate the respective associations. For example, Naragon-Gainey, Watson, and Markon (2009) found that specific facets of extraversion are differently associated with depressive symptoms and social anxiety. In their study, social anxiety was correlated with all four analyzed facets of the trait (sociability, ascendance, positive emotionality, fun-seeking), whereas depressive symptoms were linked strongly only with low positive emotionality. Other constructs such as lower order personal resources (e.g., occupational self-efficacy) might be more important in this context. Thus, other lower level personal resources should be taken into account in future studies [38, 39].

Regarding the association between detachment and depressive symptoms, we found due to a simple slope analysis that the effect of detachment depends on the individual’s CSE and was only significantly associated with depressive symptoms among managers with relatively low CSE scores (in the 25% and 50% percentile range). This finding may be due to differences in coping with work-related thoughts since not detaching from work in non-work time may be characterized by positive thinking and talking about work [22]. Since individuals high in CSE evaluate things in a consistently positive manner, they may think and speak more positively about work at home than individuals low in CSE, which reduces depressive symptoms even when they not detach. Moreover, managers high in CSE may feel more confident to be able to cope with high demands. Therefore, it is possible that individuals high in CSE can better detach since they do not have to worry about the demands in non-work time. Conversely, individuals low in CSE may feel frightened by the demands, which could lead to poorer detachment during nonwork time and more depressive symptoms [3].

While in terms of depressive symptoms our results show that CSE is able to compensate for a lack of psychological detachment, recovery from work is also important for individuals high in CSE. The odds are that individuals high in CSE may underestimate their perceived need for recovery since they feel in control regardless of how high the job demands they are facing [14]. Since it has been convincingly shown that poor psychological detachment is associated with poor well-being and health, it seems important to raise individuals’ awareness for health-relevant aspects of recovery.

Furthermore, our results showed that detachment moderated the positive correlation between demands and exhaustion. This is in line with other studies and provides further evidence that detachment may reduce the mental presence of job demands as well as exhaustion [3]. However, the correlation between perceived job demands and exhaustion are strongest for participants high in detachment, and highly detached managers show less exhaustion only under conditions of low, but not of high, job demands. This finding could be attributed to the relatively high job demands managers have reported since high job demands may influence the possibility to detach at home. To illustrate this line of thinking, Safstrom and Hartig [40] point out that bullying at work or a lack of job control might not trouble a person at home. These authors further argued that high job demands at work, may result in a perceived need to take work home, which would make detachment at home even more difficult, regardless of the individual’s detachment capability. Thus, Safstrom and Hartig [40] concluded that when the context of high job demands undermines psychological detachment; detachment may also work as a mediator. Therefore, future studies should also examine moderated mediation of detachment.

Furthermore, no moderating effect of detachment was found in the relationship between job demands, depressive symptoms and irritation. In terms of depressive symptoms, this finding could be attributed to the divergent role of self-related versus other well-being measures [41]. For instance, managers may perceive working conditions as less stressful since they are evaluated as normal for a managerial position [21]. As a result, individuals would not react with depressive symptoms, although they are exhausted. For example, results of Jacobshagen, Amstad [18] showed that the number of working hours predicted irritation and somatic complaints among managers but not depression. In terms of irritation, however, this argument does not apply. Findings from other studies regarding irritation are inconsistent since a lack of detachment is often accompanied by rumination [3, 42]. Therefore, one might speculate that this finding is attributable to the managerial sample. Generally, individuals differ in the way they switch off from work [43]. Whereas high ruminators habitually find it difficult to switch off work, low ruminators find it easy to switch off. High ruminators express a ‘live to work’ philosophy. While low ruminators express rather a ‘work to live’ mentality. For instance, high ruminators use their non- paid work time as an extension to work, thus they will work in their leisure time, respond to emails and make business phone calls [44]. Given that these attitudes and behaviors may be common among managers, this may explain why detachment was not a significant buffer in the relationship between job demands and irritation.

Results of the present study add to the existing literature. We have expanded previous research [45] that examined personality in the relationship between work stressors and psychological detachment by analyzing more stress reactions. Due to the high workloads, many managers face, psychological detachment seems to be particularly important for this target group [3].

5.1 Limitations and further research

Although our results clearly add to the existing literature, the study has certain limitations that should be acknowledged. First, our results do not allow us to determine cause-effect relationships.

For instance, high levels of depressive symptoms might also predict poor detachment. Accordingly, Sonnentag, Fritz [46] hypothesized and found in a short-term prospective study that exhaustion predicted a later decrease in psychological detachment from work. Furthermore, Nolen-Hoeksema, Stice [47] found a vicious cycle between depressive symptoms and rumination, that impaired detachment from work as well [43]. In line with this, results of Meier, Cho [22] indicated reciprocal relationships between mood, work-related thoughts and psychological detachment in three diary studies. Moreover, one might additionally argue that CSE influences the effect of depressive symptoms on psychological detachment since individuals high in CSE have a greater reservoir of psychological resources available to overcome low mood. Thus, reversed effects cannot be ruled out. Second, our results could be biased due to common method bias. However, we hypothesized and analyzed a quite specific interaction with self-reported data. Research has shown that this is already a conservative approach in moderated regression analysis since interaction effects could be deflated [37, 48]. In an additional analysis, we tested also for the presence of a common method bias by using statistical means [49]. Results suggest that the measures do not load on a single factor. Therefore, we concluded that our results were not easily attributable to third factors or common method variance. Nevertheless, further studies should use a longitudinal designs and random sampling to shed light on causal inferences. Another important short coming of the study is that we sampled only three organizations and the results are not representative for all managers in Germany. Moreover, the sample was more had higher depression scores compared to the general population, which may have an effect on the perception of job demands. Finally, epidemiological studies should analyze differences between managers and employees in more detail, also with respect to recovery behavior [2].

5.2 Implications

Our results have theoretical and practical value. In theoretical terms, our results show that a high reservoir of CSE helps to compensate for a lack of psychological detachment among managers. Individual level factors such as CSE are relatively stable constructs and only slightly malleable by intervention [50]. In practical terms, it could therefore be beneficial to assess CSE in personnel selection or development measures for managers since CSE could buffer depressive symptoms associated with poor psychological detachment [51]. As a result, managers high in CSE can better compensate for short-term periods of low psychological detachment.

Research has revealed that workload, time pressure and overtime could impair psychological detachment from work during non-work time [3]. However, high workload and overtime are common for managers and the whole situation in modern organizations cannot be changed easily [3, 52]. In line with this, managers are often seen as role models who should encourage their assigned employees to develop and implement individual strategies for detachment [53]. Therefore, it is fair to say that managers should be trained to encourage their employees’ detachment by avoiding the assignment of nonessential work tasks during leisure time [3]. However, managers are also an important target group and they may face even higher job demands such as workload, time pressure and overtime than employees [2]. In this regard, prioritizing job tasks and goal setting strategies are viewed as essential to enhance detachment even in situations of high workload [3]. However, targeted interventions based on the needs of managers should be designed to achieve better outcomes and be more cost efficient.

Conflict of interest

The authors are not aware of any affiliations, memberships, funding, or financial interests that might be perceived as affecting the objectivity of this study.

Ethical approval

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

Informed consent

Informed consent was obtained from all individual participants included in the study.

Footnotes

1

The aim of the study in Hentrich et al. 2016 was to examine the moderating role of CSE in the relation between job demands, resources, depressive symptoms and work engagement by using demands-resources theory. Thus, there is neither a big overlap of data, nor in hypotheses or examined constructs with the study presented here.

Acknowledgments

We would like to thank the BGW, the ULA and all colleagues involved.

References

Skakon

, Kristensen

, Christensen

, Lund

, Labriola

. Do managers experience more stress than employees? Results from the Intervention Project on Absence and Wellbeing (IPAW) study among Danish managers and their employees. Work. 2011;38:103–9.

Zimber

, Hentrich

, Bockhoff

, Wissing

, Petermann

. Wie stark sind Führungskräfte psychisch gefahrdet? Eine Literaturübersicht zu Gesundheitsrisiken und arbeitsbezogenen Risiko- und Schutzfaktoren. Zeitschrift für Gesundheitspsychologie. 2015;23(3):123–40.

Sonnentag

, Fritz

. Recovery from job stress: The stressor-detachment model as an integrative framework. Journal of Organizational Behavior. 2015;36:72–103.

Theorell

, Hammarström

, Aronsson

, Bendz

, Grape

, Hogstedt

, et al. A systematic review including meta-analysis of work environment and depressive symptoms. BMC Public Health. 2015;15(738):1–14.

Shirom

. The effect of work stress on health. In: Schabracq

, Winnubst

JAM

, Cooper

, editors. The Handbook of Work and Health Psychology. 2. San Francisco: John Wiley & Sons; 2003. pp. 63–82.

Mohr

, Müller

, Rigotti

, Aycan

, Tschan

. The assessment of psychological strain in work contexts. European Journal of Psychological Assessment. 2006;22(3):198–206.

Maslach

, Schaufeli

, Leiter

. Job burnout. Annu Rev Psychol. 2001;52:397–422.

Haslam

, Atkinson

, Brown

, Haslam

. Anxiety and depression in the workplace: Effects on the individual and organisation (a focus group investigation). Journal of Affective Disorders. 2005;88:209–15.

Ganster

, Fox

, Dwyer

. Explaining employees’ health care costs: A prospective examination of stressful job demands, personal control, and physiological reactivity. Journal of Applied Psychology. 2001;86:954–64.

10.

Sonnentag

, Fritz

. The recovery experience questionnaire: Development and validation of a measure for assessing recuperation and unwinding from work. Journal of Occupational Health Medicine. 2007;12(3):204–21.

11.

Semmer

. Individual Differences, Work Stress and Health. In: Schabracq

, Winnubst

JAM

, Cooper

, editors. The Handbook of Work & Health Psychology. 2. San Francisco: John Wiley & Sons; 2003. pp. 73–93

12.

Chang

C-H

, Ferris

, Johnson

, Rosen

, Tan

. Core self-evaluations: A review and evaluation of the literature. Journal of Management. 2012;38:81–128.

13.

Stumpp

, Muck

, Hülsheger

, Judge

, Maier

. Core self-evaluations in Germany: Validation of a German measure and its relationships with career success. Applied Psychology: An International Review. 2010;59(4):674–700.

14.

Judge

. Core self-evaluations and work success. Current Directions In Psychological Science. 2009;18(1):58–63.

15.

Kammeyer-Mueller

, Judge

, Scott

. The role of core self-evaluations in the coping process. Journal of Applied Psychology. 2009;94(1):177–95.

16.

Kinnunen

, Feldt

. Job Characteristics, Recovery Experiences and Occupational Well-being: Testing Cross-lagged Relationships across 1 Year. Stress & Health. 2013;29:369–82.

17.

Mohr

, Wolfram

H-J

. Stress among managers: The importance of dynamic tasks, predictability, and social support in unpredictable times. Journal of Occupational Health Psychology. 2010;15(2):167–79.

18.

Jacobshagen

, Amstad

, Semmer

, Kuster

. Work-family-balance im top-management. Zeitschrift für Arbeits- und Organisationspsychologie. 2005;49(4):208–19.

19.

Lazarus

, Folkman

. Cognitive theories of stress and the issue of circularity. In: Trumbull

MAR

, editor. Dynamics of stress. New York: Plenum; 1986. pp. 63–80.

20.

Doorn

RRv

, Hülsheger

. What makes employees resilient to job demands? The role of core self-evaluations in the relationship between job demands and strain reactions. European Journal of Work and Organizational Psychology. 2015;24(1):76–87.

21.

Peeters

MCW

, Schaufeli

, Buunk

. The role of attributions in the cognitive appraisal of work-related stressful events: An event-recording approach. 9, 463-474. Work & Stress. 1995;9(4):463–74.

22.

Meier

, Cho

, Dumani

. The effect of positive work reflection during leisure time on affective well-being: Results from three diary studies. Journal of Organizational Behavior. 2016;37:255–78.

23.

Hentrich

, Zimber

, Sosnowsky-Waschek

, Gregersen

, Petermann

. The role of core self-evaluations in explaining depression and work engagement among managers. Current Psychology 2016.

24.

Löwe

, Spitzer

, Zipfel

, Herzog

. PHQ-D Manual: Pfizer; 2002.

25.

Büssing

, Glaser

. Managerial Stress und Burnout A Collaborative International Study (CISMS) München: TUM, 1998.

26.

Maslach

, Jackson

, Leiter

. Maslach Burnout Inventory Manual. Palo Alto, CA: Consulting Psychologists Press; 1996.

27.

Richter

, Hemmann

, Merboth

, Fritz

, Hansgen

, Rudolf

. Das Erleben von Arbeitsintensität und Tätigkeitsspielraum – Entwicklung und Validierung eines Fragebogens zur orientierenden Analyse (FIT). Zeitschrift für Arbeits- und Organisationspsychologie. 2000;44(3):129–39.

28.

Nübling

, Stößel

, Hasselhorn

H-M

, Michaelis

, Hofmann

. Methoden zur Erfassung psychischer Belastungen: Erprobung eines Messinstruments (COPSOQ). Dortmund/Berlin/Dresden: Bundesanstalt für Arbeitsschutz und Arbeitsmedizin; 2005.

29.

Kristensen

, Borg

. AMI’s sporgeskema om psykisk arbejdsmiljo. Copenhagen: National Institute of Occupational Health; 2000.

30.

Limm

, Angerer

, Heinmueller

, Marten-Mittag

, Nater

, Guendel

. Self perceived stress reactivity is an indicator of psychosocial impairment at the workplace. BMC Public Health. 2010;10(252):1–10.

31.

Lakens

. Calculating and reporting effect sizes to facilitate cumulative science: A practical primer for t-tests and ANOVAs. Frontiers in Psychology. 2013;4(863):1–12.

32.

Glaesmer

, Wittig

, Brähler

, Martin

, Mewes

, Rief

. Sind Migranten häufiger von psychischen Störungen betroffen? Eine Untersuchung an einer repräsentativen Stichprobe der deutschen Allgemeinbevölkerung? [Are Migrants More Susceptible to Mental Disorders? An Evaluation of a Representative Sample of the German General Population]. Psychiat Prax. 2009;36:16–22.

33.

Aiken

, West

. Multiple regression: Testing and interpreting interactions. Newbury Park, CA: Sage; 1991.

34.

Becker

, Atinc

, Breaugh

, Carlson

, Edwards

, Spector

. Statistical control in correlational studies: 10 essential recommendations for organizational researchers. Journal of Organizational Behavior. 2016;37:157–67.

35.

R-Core-Team. R: A language and environment for statistical computing. R Foundation for Statistical Computing. Vienna, Austria: Bezogen über: http://www.r-project.org/ (08.08.2013);2015.

36.

Herrero

, Saldaña

MÁM

, Rodriguez

, Ritzel

. Influence of task demands on occupational stress: Gender differences. Journal of Safety Research. 2012;43:365–74.

37.

Siemsen

, Roth

, Oliveira

. Common method bias in regression models with linear, quadratic, and interaction effects. Organizational Research Methods. 2010;13:456–476.

38.

Kotov

, Gamez

, Schmidt

, Watson

. Linking “Big” personality traits to anxiety, depressive, and substance use disorders: A meta-analysis. Psychological Bulletin. 2010;136(5):768–821.

39.

Rigotti

, Schyns

, Mohr

. A short version of the occupational self-efficacy scale: Structural and construct validity across five countries. Journal of Career Assessment. 2008;16(2):238–55.

40.

Safstrom

, Hartig

. Psychological detachment in the relationship between job stressors and strain. Behavioral sciences. 2013;3:418–33.

41.

Widmer

, Semmer

, Kälin

, Jacobshagen

, Meier

. The ambivalence of challenge stressors: Time pressure associated with both negative and positive well-being. Journal of Vocational Behavior. 2012;80:422–33.

42.

Müller

, Mohr

, Rigotti

. Differenzielle Aspekte psychischer Beanspruchung aus Sicht der Zielorientierung. Die Faktorstruktur der Irritationsskala. Zeitschrift für Differentielle und Diagnostische Psychologie. 2004;25(4):213–25.

43.

Cropley

, Zijlstra

FRH

. Work and rumination. In: Langan-Fox

CLCJ

, editor Handbook of stress in the occupations Northampton, MA: Edward Elgar Publishing Ltd.; 2011. pp. 487–502.

44.

Cropley

, Millward

Purvis L

. How do individuals ‘switch-off’ from work during leisure? A qualitative description of the unwinding process in high and low ruminators. Leisure Studies. 2009;28(3):333–47.

45.

Potok

, Littmann-Ovadia

. Does personality regulate the work stressor–psychological detachment relationship? Journal of Career Assessment. 2014;22(1):43–58.

46.

Sonnentag

, Fritz

, Arbeus

, Mahn

. Exhaustion and lack of psychological detachment from work during off-job time: Moderator effects of time pressure and leisure experiences. Journal of Occupational Health Psychology. 2014;19(2):206–16.

47.

Nolen-Hoeksema

, Stice

, Wade

, Bohon

. Reciprocal relations between rumination and bulimic, substance abuse, and depressive symptoms in female adolescents. Journal of Abnormal Psychology. 2007;116(1):198–207.

48.

Podsakoff

, MacKenzie

, Podsakoff

. Sources of method bias in social science research and recommendations on how to control it. Annual Review of Psychology. 2012;63:539–69.

49.

Podsakoff

, MacKenzie

, Lee

, Podsakoff

. Common method biases in behavioural research: A critical review of the literature and recommended remedies. Journal of Applied Psychology. 2003;88:879–903.

50.

Chen

, Gully

, Eden

. General self-efficacy and self-esteem: Toward theoretical and empirical distinction between correlated self-evaluations. Journal of Organizational Behavior. 2004;25:375–95.

51.

Lyons

. Core self-evaluation can help in making better recruitment and selection choices: It is economical and easy to use and gives a useful new dimension. Human Resource Management International Digest. 2015;23(3):17–9.

52.

Smith

, Folkard

, Tucker

, Evans

. Work schedules, health and safety. In: Quick

Tetrick

, editors. Handbook of occupational health psychology. 2nd ed. Washington, DC: American Psychological Association; 2011. pp. 185–204.

53.

Hauck

, Chard

. How do employees and managers perceive depression: A worksite case study. Work. 2009;33:13–22.