External pressures on technology-assisted supplemental work: Motivation regulation styles and the impact on daily recovery

Abstract

Prior research has proposed detrimental but also beneficial effects of Technology-Assisted Supplemental Work (TASW) for employees’ recovery. The beneficial effects of TASW have been attributed to its autonomy-enhancing potential. Drawing on Self-Determination Theory (SDT), we distinguish two forms of TASW (reactive and proactive), depicting external and relative autonomous regulation of TASW, to explain its opposing associations with recovery. Based on SDT and the stressor-detachment model we predict detrimental effects for daily recovery experiences as TASW might comprise externally motivated behaviors that are internalized over time. We propose that external pressures (work stressors and normative pressures) serve as antecedents to TASW. Furthermore, we predict that more autonomous TASW has less detrimental effects on recovery experiences. Results from two consecutive diary studies with two and three daily measurements (N_Persons = 43, 65; n_observations = 514, 1211) largely support our hypotheses. Quantitative (overtaxing) and qualitative (hindering) work stressors and injunctive, but not descriptive norms, predicted TASW. Proactive TASW, but not reactive TASW, had a negative indirect effect on relaxation via lower levels of psychological detachment. Our findings provide insights into the temporal dynamics of external pressures, supplemental work, and subsequent impairment of recovery. Overall, our results did not support a beneficial effect of more autonomous TASW for recovery experiences.

Keywords

Recovery self-determination theory social norms technology-assisted supplemental work work stressors

Introduction

Modern work shows a clear trend toward an increasing usage of Information and Communication Technologies (ICT) in work tasks (Eurofound, 2022, 2023a). ICT make it easier to access work-related information and communicate with colleagues anytime. This trend has sparked considerable research interest in technology-assisted supplemental work (TASW)—employees’ utilization of ICT for work-related purposes outside working time without a formal contract or compensation agreement (Ďuranová and Ohly, 2016; Fenner and Renn, 2010). The consequences of TASW for recovery have been described as a “double-edged sword” (Kühner et al., 2023). While TASW is generally associated with impaired recovery (Gadeyne et al., 2023; Lanaj et al., 2014), recent studies propose potential benefits for short-term recovery (Heissler et al., 2022; Reinke and Ohly, 2021; Weigelt and Syrek, 2017). Scholars argued that TASW may extend employees’ autonomy in completing work tasks (Mazmanian et al., 2013), thus facilitating successful detachment from work. However, the degree of autonomy to engage in TASW-behavior has been neglected in empirical research. Taking employees’ autonomy in regulating TASW-behaviors into account might explain previous contradictory findings. Drawing on Self-Determination Theory (SDT, Deci and Ryan, 2000) and building on previous distinctions rooted in work-home conflict research (Höge et al., 2016) we distinguish two forms of TASW that reflect different levels of self-determination in TASW. Proactive TASW reflects employees’ self-initiated, more self-regulated, uncompensated extra or catch-up work during leisure time by means of ICT. Reactive TASW reflects employees’ other-initiated externally-regulated TASW.

We argue that TASW, as a “dark side” of work autonomy, stems from work characteristics that pressure employees to extend work into the leisure domain. We assume that work stressors and normative pressures serve as introjects that are subject to internalization processes (Deci and Ryan, 2000) and elicit both reactive and proactive TASW. Furthermore, prior research has focused on workload as a predictor of TASW (Kühner et al., 2023). However, other work stressors might be associated with TASW. Differential impacts of quantitative (overtaxing) and qualitative (hindering) work stressors (Greiner and Leitner, 1989) on TASW remain unexplored. Understanding the relative impact of both work stressors can help organizations tailor interventions more effectively. Regarding normative pressure, injunctive norms (i.e. organizational and supervisor expectations) and descriptive norms (i.e. colleagues’ behaviors; Fishbein and Ajzen, 2010) might be related to TASW-behaviors. Prior studies explored their separate influence on TASW (e.g. Derks et al., 2014; Park et al., 2011). By measuring only one source of normative pressure, previous studies may have overestimated their respective influence.

With few exceptions (e.g. Eichberger et al., 2022; Lanaj et al., 2014) research on TASW has focused on psychological detachment, but neglected other important recovery experiences proposed by the stressor-detachment model (Sonnentag and Fritz, 2015). In particular, relaxation experiences, defined as “state[s] of low activation and increased positive affect” (Sonnentag and Fritz, 2007: 206), have been proposed as important factor in the alleviation of short-term strain (Meijman and Mulder, 1998). We propose negative effects of TASW for employees’ recovery experiences because employees sacrifice leisure time that cannot be allocated to recovery activities. We propose a difference in the negative consequences of proactive and reactive TASW for recovery experiences due to differences in self-determination. More autonomous behaviors are related to more effective performance due to lower cognitive dissonance (Deci and Ryan, 1985; Lavergne and Pelletier, 2016). Therefore, employees might experience less detrimental effects of proactive TASW for detachment and relaxation due to lower cognitive dissonance. Overall, we propose that external pressures drive extrinsically motivated reactive and proactive TASW, which impair recovery experiences. Figure 1 outlines our conceptual model.

Figure 1.

Conceptual model.

Our research aims to contribute to the literature in three ways. First, by investigating the influence of external pressures on TASW, we provide a possible explanation for previous contradictory findings and answer research calls to ground the TASW-phenomenon in theory (e.g. Ďuranová and Ohly, 2016; Eichberger and Zacher, 2021; Kühner et al., 2023). Second, by examining differential effects of proactive and reactive TASW on recovery experiences, we clarify the consequences of TASW for recovery at different levels of self-determination. Third, by investigating temporal dynamics of external pressures, TASW, and recovery experiences, we contribute to the understanding of when and why work spills over into the leisure domain and impedes recovery from work. We conducted two diary studies to test our hypotheses. Study 1 utilized two daily measurements to establish the effects of work overload on proactive and reactive TASW and test effects on recovery experiences. In Study 2, we aimed to replicate the findings of Study 1 and to gain a more detailed picture of different external pressures for TASW and its consequences for recovery experiences. We employed an autoregressive cross-lagged daily diary design with three evening measurements to overcome limitations of predominant cross-sectional diary studies and examine temporal dynamics.

Proactive and reactive TASW-behavior

TASW encompasses various in-role work behaviors (Fenner and Renn, 2004) and has been broadly defined as any work activity during non-work time performed via ICT without a formal contract or compensation agreement (Arlinghaus and Nachreiner, 2014; Ďuranová and Ohly, 2016), which represents extra or catch-up work in order to fulfill prescribed tasks or duties (Fenner and Renn, 2004). During non-work time employees can be accessed by colleagues, clients or supervisors via calls or messages, or can self-initiate access to work materials via digital technologies.

Increased self-determination in TASW-behavior has served as a potential explanation for previous study outcomes, but has not yet undergone empirical examination. Scholars argued that TASW extends subjective autonomy due to more work flexibility (Mazmanian et al., 2013) with beneficial effects for recovery (Heissler et al., 2022; Weigelt and Syrek, 2017). Furthermore, previous studies showed that autonomy at work (Kühner et al., 2023) and via ICT extended flexibility (Diaz et al., 2012) are positively associated with TASW. However, TASW has also been associated with decreased control in terms of work spillover into the private domain (Derks et al., 2014; Dettmers et al., 2016) and reduced recovery (Kühner et al., 2023). SDT (Deci and Ryan, 2000) provides a theoretical lens to study autonomy in TASW-behavior.

SDT is a theory of human behavior, which emphasizes the importance of regulation processes for behavior (Deci and Ryan, 2000), defines autonomy as a “need to self-regulate one’s experiences and actions” (Ryan and Deci, 2017: 10) and has been frequently applied in the human resource management literature (Laguerre and Barnes-Farrell, 2025). SDT views motivations to engage in behaviors as a continuum of regulation styles. Differing types of regulation provide discriminant validity in predicting behavior (Howard et al., 2017), and findings support the assumption that regulation styles are arranged along a linear continuum from extrinsic to intrinsic regulation (Van Den Broeck et al., 2021). Fully autonomous behaviors are accompanied by a sense of voluntariness and choice and are intrinsically regulated (Gagné and Deci, 2005). In contrast, extrinsically regulated behaviors are a result of external pressures. Over time, people can accept the values of such external pressures and internalize them in varying degrees (Ryan and Deci, 2017). Whereas externally regulated behaviors are fully “initiated and maintained by contingencies external to the person” (Gagné and Deci, 2005: 334), introjected regulated behaviors are internal forces resulting from a partial assimilation of such external contingencies. Such assimilation processes can range from identifying with the values of the external contingency, to fully integrating them into the self through bringing these values into congruence with other aspects of one’s self (Deci and Ryan, 2000).

In the context of TASW employees might exhibit varying levels of autonomy. Getting accessed by colleagues reflects an external regulation which indicates less autonomous behavior. Self-initiating access to work indicates more autonomy, because no external event triggers the regulation of behavior (Ryan and Deci, 2017). Self-initiation of TASW can be subject to differing levels of autonomy due to internalization processes. Employees might experience an internally controlling force to engage in TASW as they have internalized, but not yet fully accepted external controls for TASW, thus undermining autonomous regulation (Ryan, 1982). However, if these external controls are integrated in an individuals’ value or belief system, TASW becomes more identified and thus more autonomous regulated (Deci and Ryan, 2000). Therefore, self-regulated proactive TASW represents a more autonomously regulated behavior than externally-regulated reactive TASW. However, TASW-behavior is unlikely to be fully intrinsically regulated, because technology-assisted supplemental work is a behavioral reaction due to external demands (Fenner and Renn, 2004). Rather, external pressures might be internalized, which shifts the regulation focus from fully dependent on external pressures to more autonomous regulation (Deci and Ryan, 2000).

Work stressors and TASW

Work stressors might be positively associated with TASW because they overtax or hinder goal achievement during work. The Job Demands-Resources (JDR) model (Bakker et al., 2023; Demerouti et al., 2001) defined job demands as work characteristics that trigger effort-driven processes, consume physical and psychological energy, and increase the risk of burnout and health impairment. Due to meta-analytic evidence of ambiguous (positive as well as negative) effects of job demands on motivation and health (Crawford et al., 2010), and drawing on transactional stress theory (Lazarus and Folkman, 1984), job demands have been further differentiated into challenge demands (facilitating learning and growth) and hindrance demands (threatening regulation capacities and health). Research drawing on the JDR model often uses the terms demands and stressors interchangeably (see e.g. Bennett et al., 2018: 268) and relies on subjective appraisals to classify demands (or stressors) as either challenging or hindering (Glaser et al., 2015). In contrast, Action Regulation Theory (Hacker, 2003) more clearly distinguishes between (learning) demands and work stressors. According to Action Regulation Theory, people actively regulate their work-behaviors toward goal achievement (Hacker, 2003; Hacker and Sachse, 2023). Work stressors are conceptualized as regulation problems that disturb the cognitive regulation of actions and subsequently overtax or hinder goal achievement (Leitner et al., 1987; Semmer, 1984). Additional effort for task completion (Leitner et al., 1987) causes time deficits and subsequent unfinished work tasks (Frese and Zapf, 1994). These unfinished (or unsatisfactorily finished) work tasks induce employees to finish tasks in their leisure time as a form of closure (Lewin, 1939; Syrek et al., 2017) or to reduce or prevent future workload. In support of this argument, Gadeyne et al. (2023) demonstrated positive associations between work overload, unfinished work tasks, and TASW in a daily diary study.

Work stressors might be associated with both proactive and reactive TASW. Prolonged exposure to work stressors may lead to internalizing external pressures (Deci and Ryan, 2000) and potentially trigger proactive TASW. Furthermore, work stressors are work characteristics that also affect coworkers in the organization (Frese and Zapf, 1994). Employees might also exhibit higher levels of reactive TASW because colleagues and supervisors also engage in TASW. Supporting this argument, Göllner and Rau (2021) showed that employees who are contacted outside of working hours report higher levels of work stressors compared to employees who are not contacted.

Hypothesis 1: Work stressors are positively related to (a) proactive TASW and (b) reactive TASW.

Normative pressure and TASW

Normative pressure can be described as “social pressures to perform (or not to perform) a given behavior” (Fishbein and Ajzen, 2010: 130). Individuals internalize behaviors from their social groups over time and transform them into self-regulations which allows them to be executed independently (Ryan and Deci, 2017). Individuals strive to internalize and enact these social norms in order to foster social cohesion or to fulfill own basic psychological needs (Chirkov et al., 2003; Deci and Ryan, 2000).

Individuals receive information about social norms in their environment from two sources of normative pressure (Fishbein and Ajzen, 2010). Injunctive norms refer to perceptions of what important referent individuals or groups expect individuals to do, such as supervisor or organizational expectations. Descriptive norms relate to perceived behaviors of relevant others. Scholars argued that both sources of normative pressure are conceptually different and separately influence individuals’ normative beliefs about expected behavior. Injunctive norms impact behavior through avoiding social sanctions, whereas descriptive norms offer heuristics about appropriate behaviors (Fishbein and Ajzen, 2010; Manning, 2009).

Both descriptive and injunctive norms might separately predict TASW. Measuring the relative impact of injunctive and descriptive norms helps to better predict the influence of normative pressure on behavior (Manning, 2009). Studies measuring either injunctive (e.g. Derks et al., 2014; Fenner and Renn, 2010; Gadeyne et al., 2018; Reinke and Gerlach, 2022) or descriptive norms (e.g. Park et al., 2011; Richardson and Benbunan-Fich, 2011) have linked normative pressure to TASW-behavior, but haven’t examined their relative influence on TASW. However, Palm et al. (2020) showed in a cross-sectional design that injunctive and descriptive norms independently predict behavior in the context of work-to-non-work integration, which demonstrates their incremental validity. Because previous studies did not control for covariance between social norms, they may have overestimated their effects. Furthermore, understanding the relative impact of both normative pressures can help organizations tailor interventions more effectively. Because social norms represent external normative pressures (Fishbein and Ajzen, 2010) that can be internalized over time (Deci and Ryan, 2000), we propose that both injunctive and descriptive norms serve as antecedents for both proactive and reactive TASW.

Hypothesis 2: Normative pressures (i.e. injunctive norms and descriptive norms) are positively related to (a) proactive TASW and (b) reactive TASW.

TASW, psychological detachment, and relaxation

Recovery from work is a crucial process for employees’ well-being, health, and work performance. Recovery is defined as “a process of psychophysiological unwinding after effort expenditure” (Geurts and Sonnentag, 2006: 482) that replenishes spent resources (Meijman and Mulder, 1998). Drawing on the stressor-detachment model (Sonnentag and Fritz, 2015), effective recovery from work requires both physical and mental detachment during leisure time. Psychological detachment serves as a critical mediating process that allows individuals to distance themselves cognitively from activation due to work-related stressors and facilitates subsequent recovery experiences (Sonnentag and Fritz, 2015). Especially the experience of relaxation has been proposed to be most beneficial to restore the pre-stressor state in the short-term (Meijman and Mulder, 1998; Sonnentag and Fritz, 2007). Failure to restore spent resources has been associated with detrimental outcomes such as lower worker safety and higher counterproductive work behaviors (Wendsche et al., 2021).

We expect negative effects of TASW for both recovery experiences. Due to TASW, individuals engage in work behaviors and experience work-related activation in their leisure time. Therefore, employees sacrifice time for work activities—leisure time that is usually allocated for recovery—and impose prolonged strain on their psychobiological systems (Meijman and Mulder, 1998). In support of this argument, a meta-analysis (Kühner et al., 2023) presented cross-sectional evidence that indicated adverse effects of TASW on psychological detachment. Several diary studies (e.g. Braukmann et al., 2018; Derks et al., 2014; Eichberger et al., 2021; Gadeyne et al., 2023; Ohly and Latour, 2014; Reinke and Ohly, 2021) offer additional support for the detrimental impact of TASW on psychological detachment. Surprisingly, empirical evidence for the proposed mediating effect of detachment on subsequent relaxation experiences (Sonnentag and Fritz, 2015) is still missing in the context of TASW.

Recently, potential benefits of TASW for short-term recovery were suggested, as TASW may extend employees’ autonomy in completing work tasks. Scholars argued that higher autonomy might facilitate successful detachment from work. Weigelt and Syrek (2017) suggested that completing work tasks during the weekend might buffer the negative effects of unfinished work tasks on psychological detachment and relaxation. In a daily diary study, Heissler et al. (2022) observed that employees engaged in TASW because they could not detach from work. However, employees didn’t report elevated (or diminished) levels of psychological detachment due to engaging in TASW or due to progress in task completion during leisure time. Reinke and Ohly (2021) examined the effects of autonomous and controlled motivation to use TASW for psychological detachment. While autonomously motivated employees appraised TASW more positively, they didn’t show higher levels of detachment. Additionally, the duration of TASW didn’t moderate the relationships. This implies that irrespective of the duration of TASW, it results in impaired psychological detachment.

Overall, we propose that both proactive and reactive TASW have negative effects on daily psychological detachment and relaxation. For proactive TASW, employees may autonomously allocate leisure time and sustained effort to work tasks, thereby impeding recovery experiences. For reactive TASW, being contacted for work-related reasons during leisure time interrupts leisure activities, which obstructs psychological detachment, and relaxation experiences (Sonnentag and Fritz, 2007).

Hypothesis 3: (a) Proactive TASW and (b) reactive TASW are negatively related to psychological detachment.

Hypothesis 4: (a) Proactive TASW and (b) reactive TASW are negatively related to relaxation via decreased psychological detachment.

We propose that more autonomous, proactive TASW-behaviors have less detrimental effects on recovery experiences than reactive TASW-behaviors. Employees might engage in TASW and complete work tasks more quickly by proactive compared to reactive TASW. First, behavior change is less dissonant to the self and less discomforting for more autonomous regulated behaviors (Lavergne and Pelletier, 2016). Employees might therefore turn from non-work behaviors to work behaviors more effortlessly in the case of more autonomous, proactive TASW. Furthermore, a state of cognitive dissonance is associated with higher activation (Lavergne and Pelletier, 2016). Employees might need more time to distance themselves from the higher work-related activation due to externally regulated reactive TASW. Second, autonomous regulated behavior is related to better functioning and more effective performance (Deci and Ryan, 1985). Therefore, employees might disengage from work more quickly for proactive TASW due to more effective task accomplishment. Overall, by proactive TASW employees may be able to complete their work more quickly. This provides them with more time to engage in activities that facilitate the experience of psychological detachment and relaxation.

Hypothesis 5: Proactive TASW has a less strong negative relation to (a) psychological detachment and (b) relaxation than reactive TASW.

We conducted two daily diary studies to test our hypotheses. Study 1 examined the predictive role of work overload on both forms of TASW and investigated the mediation effect of TASW on relaxation via psychological detachment. In Study 2, we replicated findings from Study 1 and expanded upon the results by including quantitative (overtaxing) and qualitative (hindering) work stressors (Greiner and Leitner, 1989) to predict proactive and reactive TASW. Additionally, we tested the proposed relationships of normative pressures for TASW. To gain a more detailed picture of the daily temporal dynamics of external pressures, TASW, and recovery experiences, we increased the number of measurement points during the evening, utilizing a longitudinal autoregressive cross-lagged daily diary design.

Study 1

Participants and procedures

Study 1 was conducted in German-speaking countries using a convenience sample recruited via the snowballing technique. Participation was incentivized with feedback on study results and entry into a raffle for four 50€ gift cards upon study completion. Eligible participants needed to (a) be at least 18 years old, (b) be fluent in German, (c) work at least 15 hours a week, (d) have access to ICT for TASW-behavior, and (e) not be on-call workers or have contractual availability requirements.

Initially, participants completed a baseline questionnaire (T0) including demographic questions and selected a typical workweek for daily surveys which was not situated directly before or after vacation to avoid its influences on recovery (De Bloom et al., 2013) or work stressors (Nawijn et al., 2013). Over 7 days, participants completed two daily surveys: one before going to bed (T1), assessing work overload and TASW-behavior, and another after waking up (T2), evaluating recovery experiences after work from the previous day.

Of the 117 baseline respondents, 45 completed at least one daily questionnaire. After excluding participants who didn’t complete two subsequent surveys (evening and following morning, n = 2), and screening for duplicate (n = 10) and delayed entries (n = 6), the final sample included N = 43 participants with n = 514 observations. On average, participants contributed 10.95 daily observations (SD = 4.19). 24 participants (56%) provided data for all 14 daily questionnaires. Participants (56% female) were on average 36.4 years old (SD = 11.43), worked in a broad range of sectors, such as health and social services (19%), research and development (14%) and technology and telecommunication (14%), with 69% holding a university degree and 30% being in a leadership position. Employees worked on average 39.2 hours (SD = 11.94) per week and held their positions in the company for on average 5.8 years (SD = 7.46). 23% reported having contractually fixed starting and ending times, 30% had a flexi-time model with fixed core working hours, 18% without core working hours. 27% had no fixed contractual working hours.

Measures

Consistent with recommendations by Gabriel et al. (2019) we adapted scale items for daily measurement. To estimate reliability, we calculated ω (MacDonald, 1999) by performing multilevel confirmatory factor analysis, to partition within- and between-level reliability (Geldhof et al., 2014). All scales were presented in German language.

Work overload was measured using three items from the respective subscale of the TAA (Glaser et al., 2020) on a scale ranging from 1 (no, not at all) to 5 (yes, exactly). A sample item is “Today at work, I often had to hurry and still couldn`t complete my work.”

Proactive TASW was examined with two items developed by Höge et al. (2016), assessing the frequency of use of ICT for TASW on a scale ranging from 1 (never) to 5 (very often). A sample item is “How often, out of your own accord, did you use the following technologies for work-related purposes outside of your actual working hours today? . . . - Laptop/Computer.”

Reactive TASW was examined with two items developed by Höge et al. (2016), assessing the frequency of use of ICT for TASW on a scale ranging from 1 (never) to 5 (very often). A sample item is “How often have you been contacted today by your employer, customers, or clients outside of actual working hours using the following technologies? . . . - Smartphone.”

Psychological detachment was measured using four items of the psychological detachment scale of the Recovery Experience Questionnaire (Sonnentag and Fritz, 2007) on a scale ranging from 1 (strongly disagree) to 5 (strongly agree). A sample item is “Yesterday during my free time, I forgot about work.”

Relaxation was measured using four items of the relaxation scale of the Recovery Experience Questionnaire (Sonnentag and Fritz, 2007) on a scale ranging from 1 (strongly disagree) to 5 (strongly agree). A sample item is “Yesterday during my free time, I did relaxing things.”

Analytical strategy

Our data have a multilevel structure, as days are nested within persons. In order to take this into account, we used multilevel structural equation modeling (MSEM) in Mplus 8.10 (Muthén and Muthén, 2023) to analyze our data. For testing our hypothesis, we used manifest variables by calculating mean scores of the respective items to simplify data complexity. We allowed proactive and reactive TASW to correlate in all models. Because MSEM takes the multilevel structure into account by latent decomposition of the covariance structure, separating within- from between-person effects when the variables are measured on both levels (Preacher et al., 2010), no centering of predictor variables was necessary. We used robust maximum likelihood estimation for our models. We used multiple imputation (Asparouhov and Muthén, 2025), including daily working hours as an auxiliary variable to handle missing data. We report mean estimates of 50 imputations. We controlled for the direct effect of daily work overload on daily psychological detachment in our models. Model 1 (M1) specified direct effects, Model 2 (M2) tested mediation effects by introducing paths for reactive and proactive TASW on relaxation (Preacher et al., 2010). For additional robustness checks, we employed Bayesian estimation (Muthén, 2010) suitable for small sample sizes and multilevel modeling (Yuan and MacKinnon, 2009). This allowed us to obtain credible intervals and standardized estimates. Fit indices for all models can be found in the Supplemental Material (Supplemental Table S1).

Results

Results from an attrition analysis are provided in the Supplemental Material. Table 1 depicts means, standard deviations, reliabilities, and correlations of the study variables. We tested whether a multilevel approach is justified by examining the between-person variance with null-models. Results revealed between-person variance for all variables (see Table 1).

Table 1.

Descriptive statistics and correlations of Study 1 variables.

Variables	M	SD	ICC	1	2	3	4	5	6
1. Working hours	5.30	2.48	–	–	0.54***	0.09	0.09	−0.40**	0.06
2. Work overload	2.17	0.79	0.26	0.60***	(0.90)	0.31*	0.28	−0.61***	−0.08
3. Proactive TASW	2.07	0.95	0.51	0.19**	0.22***	(0.77)	0.61***	−0.55***	−0.17
4. Reactive TASW	1.49	0.57	0.41	0.27***	0.23***	0.34***	(0.69)	−0.28	−0.16
5. Psychological detachment	3.50	0.87	0.51	−0.30***	−0.35***	−0.38***	−0.15*	(0.90)	0.41**
6. Relaxation	3.23	0.87	0.49	−0.33***	−0.28***	−0.26***	−0.10	0.47***	(0.84)

Coefficients below the diagonal represent group-mean centered within-person correlations (N_within = 246). Coefficients above the diagonal represent grand-mean centered between-person correlations (N_between = 43). Within-person reliabilities are represented in parenthesis on the diagonal. Coefficients represent non-imputed data.

M: between-person mean; SD: between-person standard deviation; ICC: intra class correlation coefficient.

p < 0.05. ^**p < 0.01. ^***p < 0.001.

We conducted multilevel confirmatory factor analysis (MCFA) to confirm our hypothesized factor structure. A model specifying one factor (C1) was compared to a three-factor model (C2) with psychological detachment and relaxation items loading on one overall recovery experiences-factor and proactive and reactive TASW loading on one overall TASW-factor. We also compared two 4-factor models, specifying the aforementioned overall factors separately in each model (C3: one factor recovery experiences; C4: one factor TASW). We omitted one item from the detachment scale and one item from the relaxation scale, as they didn’t fit the data. A five-factor-solution (C5) fitted the data substantially better than all other models and indicated good model fit (see Supplemental Material, Supplemental Table S1, Hu and Bentler, 1999).

Figure 2 illustrates our final MSEM-model with standardized within-level estimates. Supplemental Tables S2 and S3 in the Supplemental Material depict within-person and between-person parameter estimates and credible intervals for our final model (M2). In line with Hypothesis 1, the work stressor daily work overload predicted daily proactive (H1a, b = 0.16, p = 0.01) and daily reactive TASW (H1b, b = 0.12, p = 0.001). Beyond daily work overload (b = −0.22, p ⩽ 0.001), daily proactive (H3a, b = −0.34, p = 0.001) but not daily reactive TASW (H3b, b = 0.69, p = 0.42) predicted daily psychological detachment, thus partially supporting Hypothesis 3. The mediating effect of daily psychological detachment on the relation between daily TASW and daily relaxation, was significant for daily proactive TASW (H4a, b = −0.13, 95% CI (−0.21; −0.06)), but not for daily reactive TASW (H4b, b = −0.003, 95% CI (−0.03; 0.02)). Contrary to Hypothesis 5, daily proactive TASW (H5a, b = −0.34, p = 0.001) didn’t impair daily psychological detachment less strongly than daily reactive TASW (H5a, b = 0.07, p = 0.42). Both forms of daily TASW showed no significant direct effect for daily relaxation experiences (H5b, b_{proactive TASW} = −0.03, p_{proactive TASW} = 0.64; H5b, b_{reactive TASW} = −0.04, p_{reactive TASW} = 0.66).

Figure 2.

Results of Study 1.

Post hoc analysis

Due to the significant within-level paths of work overload to proactive TASW and proactive TASW to psychological detachment, we further examined the mediation effect of work overload on psychological detachment via proactive TASW. Results from Bayesian estimation indicates a significant within-level mediating effect (b = −0.05, 95% CI (−0.10; −0.02)). We additionally tested the within-level serial mediating effect of work overload via proactive TASW and psychological detachment on relaxation via Bayesian estimation. Results (b = −0.02, 95% CI (−0.04; −0.01)) indicate a significant serial mediation effect.

Brief discussion

Our results suggest that when employees experience more external pressures in terms of daily work overload, they engage in more proactive and reactive TASW-behavior in the evening. Furthermore, beyond the effects of daily work overload, higher levels of daily self-initiated, but not higher levels of daily other-initiated TASW, led to less psychological detachment during the evening. In line with propositions of the stressor-detachment model, we found evidence for a serial mediation effect between work overload and relaxation via proactive TASW and psychological detachment, suggesting that increases in external pressure interfere with employees’ recovery experiences by eliciting proactive TASW. Contrary to our hypothesis, proactive TASW was more detrimental for recovery experiences in the evening than reactive TASW. Correlations (see Table 1) provide additional evidence for this relationship as psychological detachment and relaxation experiences correlate more strongly with proactive TASW (r_detachment = −0.38, r_relaxation = −0.26), than with reactive TASW (r_detachment = −0.15, r_relaxation = −0.10). A possible explanation for the nonsignificant associations of reactive TASW may be overlapping variances with work overload and proactive TASW, which suppress reactive TASW’s consequences for recovery experiences. Although we assumed intraday temporal dynamics of work overload, proactive and reactive TASW, and recovery, we cannot infer the chronological order of all variables. Even though measurement points of TASW and its consequences were temporally separated, they each referred to behaviors during the whole evening, thus representing daily cross-sectional effects and limiting conclusions about chronological sequences.

Study 2

Our aims for Study 2 were to replicate the findings of Study 1 and to gain a more detailed picture of different external pressures, proactive and reactive TASW, and recovery experiences. First, in addition to work overload, we included information deficits and work interruptions in Study 2. From a practical perspective, it is important to distinguish between types of work stressors because they require different approaches to reduce them. Greiner and Leitner (1989) distinguish between two types of work stressors. Regulation obstacles are qualitative work stressors that refer to recurring events and working conditions that hinder goal achievement, such as information deficits or work interruptions. Overtaxing regulations are quantitative work stressors that refer to the speed and intensity in which work tasks must be completed, such as work overload. Interventions for quantitative stressors might focus on workload adjustments or task prioritization to help employees better handle the amount of work. Interventions for qualitative stressors might involve role clarification or improvement of information systems to mitigate regulation problems. Both quantitative (overtaxing) and qualitative (hindering) work stressors serve as external pressures that elicit proactive and reactive TASW because both classes of work stressors require additional effort and cause time deficits (Leitner et al., 1987). Therefore, we included unfinished work tasks at the end of the workday as additional measure. To reduce participant burden, we only included unfinished work tasks at the day-level because work stressors hinder task completion (Leitner et al., 1987; Semmer, 1984). Second, we included descriptive and injunctive norms—colleagues’ segmentation norms and employers’ flexibility requirements—as predictors of proactive and reactive TASW. Finally, we employed an autoregressive cross-lagged daily diary design to capture the temporal dynamics of work stressors, TASW, and recovery experiences in the evening. This allows us to establish temporal precedence for predictors, TASW, and its consequences.

Participants and procedures

Study 2 was conducted in German-speaking countries. We recruited a convenience sample by distributing flyers that contained study information and a link to an online questionnaire via social media and emails to HR departments and department heads in a broad range of companies, asking them to distribute the information among their employees. As an incentive, we offered feedback on the study results. Inclusion criteria of Study 2 remained the same as in Study 1.

Initially, participants completed a baseline questionnaire (T0) with measures for work stressors, normative pressures, and demographics. They selected a typical workweek for the daily surveys which was not situated directly before or after vacation to avoid its influences on recovery (De Bloom et al., 2013) or work stressors (Nawijn et al., 2013). Participants filled in three daily surveys over 7 days, shortly after work (T1), in the early evening (T2), and in the late evening (T3) via the Ethica smartphone application (Ethica Data, 2023). We further included daily psychological detachment at T1 to test the proposed beneficial effects of TASW on recovery experiences.

In the baseline survey, participants indicated the time when they wanted to receive their daily questionnaires. They received the first daily survey 30 minutes after finishing work (T1). To allow for paid overtime, commuting, and giving participants more flexibility, we allowed delaying answering each survey by 60 minutes. T2 was prompted 2 hours, T3 4 hours after T1 to gain a comprehensive picture of employees’ leisure time after work.

Our sample consisted of N = 65 participants who completed at least one daily questionnaire, resulting in 382 day within-level cases and n = 898 observations with on average 13.82 daily observations per participant (SD = 5.2). four participants (6.2%) provided data for all 21 daily questionnaires. Participants (59% female) were on average 38.2 years old (SD = 13.1), worked in a broad range of sectors, such as education (32%), research and development (19%), manufacturing industry (17%), health and social services (9%), with 75% holding a university degree and 35% holding a leadership position. Employees worked on average 38.3 hours (SD = 12.15) per week and held their positions in the company for on average 7.0 years (SD = 7.56). 11% reported having contractually fixed starting and ending work times, 23% had a flexi-time model with fixed core working hours, 20% had no core working hours. 43% percent had no fixed contractual working hours.

Measures

In Study 2, we used the same scales and procedures for assessing reliability as in Study 1. Table 2 shows reliabilities, means, standard deviations, and correlations of the study variables. Psychological detachment at T1 referred to participants’ psychological detachment after work, at T2 and T3 items referred to time frames since the last questionnaire. Proactive TASW, reactive TASW, and relaxation were measured at T2 and T3 and referred to participants’ behaviors and experiences since the last questionnaire. All measures were answered on a scale ranging from 1 (no, not at all) to 5 (yes, exactly). Listed below are the scales we used in Study 2 in addition to scales from Study 1.

Table 2.

Descriptive statistics and correlations of Study 2 variables.

Variables	M	SD	ICC	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17	18
1. Working hours	5.08	2.09	-	-	0.19	−0.26*	−0.05	−0.17	−0.17	−0.08	0.07	0.14	−0.08	0.08	0.03	0.07	0.05	0.13	0.04	−0.05	−0.07
2. T1 unfinished work tasks	2.43	0.78	0.27	0.24***	(0.88)	−0.55***	−0.52***	−0.48***	−0.40**	−0.42***	0.42***	0.36**	0.38**	0.33**	−0.04	0.20	0.28*	0.19	−0.02	−0.01	0.12
3. T1 psychological detachment	3.20	0.92	0.28	−0.45***	−0.22***	(0.96)	0.73***	0.67***	0.62***	0.56***	−0.50***	−0.44***	−0.40**	−0.32*	0.26*	−0.35**	−0.31*	−0.28*	−0.14	−0.14	−0.11
4. T2 psychological detachment	3.73	0.98	0.37	−0.28***	−0.23***	0.43***	(0.95)	0.81***	0.64***	0.61***	−0.62***	−0.52***	−0.49***	−0.35**	0.31*	−0.38**	−0.37**	−0.28*	−0.16	−0.08	0.01
5. T3 psychological detachment	4.02	0.86	0.43	−0.14*	−0.17*	0.16*	0.42***	(0.93)	0.58***	0.73***	−0.55***	−0.66***	−0.35**	−0.36**	0.40**	−0.48***	−0.47***	−0.19	−0.28*	−0.10	−0.09
6. T2 relaxation	3.26	0.91	0.32	−0.29***	−0.13	0.32***	0.66***	0.37***	(0.94)	0.75***	−0.47***	−0.39**	−0.24	−0.26*	0.26*	−0.33**	−0.26*	−0.17	−0.17	−0.07	−0.11
7. T3 relaxation	3.69	0.97	0.46	−0.19**	−0.09	0.19**	0.50***	0.75***	0.50***	(0.92)	−0.51***	−0.47***	−0.29*	−0.28*	0.35**	−0.50***	−0.44***	−0.20	−0.34**	−0.14	−0.14
8. T2 proactive TASWa	1.63	0.62	0.41	0.18**	0.23***	−0.27***	−0.53***	−0.42***	−0.43***	−0.38***	(0.51)	0.70***	0.70***	0.54***	−0.23	0.47***	0.45***	0.15	0.33**	0.07	0.09
9. T3 proactive TASWa	1.59	0.69	0.44	0.04	0.14*	−0.17**	−0.27***	−0.59***	−0.26***	−0.45***	0.30***	(0.62)	0.48***	0.59***	−0.23	0.48***	0.40**	0.08	0.30*	0.03	0.04
10. T2 reactive TASWa	1.28	0.45	0.37	0.14*	0.08	−0.02	−0.22***	−0.24***	−0.21***	−0.20**	0.50***	0.15*	(0.61)	0.72***	−0.14	0.30*	0.29*	0.22	0.37**	0.08	0.10
11. T3 reactive TASWa	1.23	0.42	0.31	0.08	0.09	−0.10	−0.09	−0.39***	−0.08	−0.26***	0.26***	0.51***	0.38***	(0.60)	−0.38**	0.40**	0.29*	0.15	0.21	0.07	0.13
12. T0 descriptive norms	2.61	1.07	-	-	-	-	-	-	-	-	-	-	-	-	(0.93)	−0.46***	−0.43***	−0.29*	−0.27*	0.17	0.08
13. T0 injunctive norms	3.86	1.40	-	-	-	-	-	-	-	-	-	-	-	-	-	(0.83)	0.59***	0.28*	0.44***	0.18	0.05
14. T0 work overload	3.44	1.04	-	-	-	-	-	-	-	-	-	-	-	-	-	-	(0.79)	0.25*	0.45***	0.11	0.21
15. T0 information problems	2.66	0.81	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	(0.77)	0.46***	−0.20	0.00
16. T0 work interruptions	2.62	0.87	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	(0.70)	0.16	0.00
17. Age	38.2	13.11	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	0.10
18. Gender	0.59	0.50	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-

Coefficients below the diagonal represent group-mean centered within-person correlations (N_within = 231–307). Coefficients above the diagonal represent grand-mean centered between-person correlations (N_between = 65). Within-person reliabilities are represented in parenthesis on the diagonal for within-person variables. Between-person reliabilities are represented in parenthesis on the diagonal for T0 variables.

M: between-person mean; SD: between-person standard deviation; ICC: intra class correlation coefficient.

M and SD before log-transformation.

p < 0.05. ^**p < 0.01. ^***p < 0.001.

Following Palm et al. (2020), descriptive norms were operationalized at T0 using three of the four items of the perceived segmentation norms scale by Park et al. (2011), originally developed by Kreiner (2006). Participants rated items on their colleagues’ global attitude regarding the segmentation of work and private matters. A sample item is “The people I work with keep work matters at work.”

Injunctive norms were operationalized at T0 using four items of the requirements for temporal flexibility at work scale developed by Höge (2011). Participants rated items regarding perceived employer expectations for flexible working time schedules. A sample item is “In my work, my employer expects me to be flexible as far as my working hours are concerned.”

Work stressors were operationalized at T0 via the quantitative work stressor work overload and the qualitative work stressors information problems and work interruptions using the corresponding scales from the TAA by Glaser et al. (2020). Information problems were measured using four items regarding problems of acquisition, processing, and transmission of information at work. A sample item is “Information needed for my work is often not available.” Work interruptions were measured using four items regarding disruptions of work processes due to colleagues, tools, or ICT. A sample item is “I am often interrupted in my tasks due to requests by other persons.” Unfinished work tasks were operationalized at T1 using four of six items of the scale by Syrek et al. (2017), which we adapted to the day level. A sample item is “I have not finished important tasks that I had planned to do today.” We excluded two less-suitable items for the day level (e.g. “I have not started working on urgent tasks that were due this week”).

Analytical strategy

Our data have a multilevel structure, as observations are nested within persons. Because we measured our study variables on multiple points during the day, we had an autoregressive daily diary design. We used MSEM with full information maximum likelihood estimation (N_Persons = 65 with n_observations = 1146) in Mplus 8.10 (Muthén and Muthén, 2023) to test our hypotheses. We used manifest variables by calculating mean scores of their respective items. We allowed variables that were measured at the same time point to correlate. To handle the complexity of the data, we tested our hypotheses by estimating two models. First, we modeled the effects of norms and work stressors on daily unfinished work tasks (M1). Second, we estimated within-person effects by specifying the hypothesized autoregressive cross-lagged model (M2). We then added paths for testing the mediation effects (M3) and finally estimated our model again on both levels (M4). Fit indices of all models are presented in the Supplemental Material, Supplemental Table S4. To check the robustness of effects we additionally estimated our final model using Bayesian estimation (Muthén, 2010; Yuan and MacKinnon, 2009), thus obtaining credible intervals and standardized estimates.

Results

Results from an attrition analysis can be found in the Supplemental Material. Table 2 depicts intraclass correlations, means, standard deviations, reliabilities, and correlations of our study variables. Results indicate between-person variance for all variables that were measured daily. We log-transformed proactive and reactive TASW with the logarithmus naturalis at all time points as they showed a floor effect. MCFA confirmed our hypothesized factor structure. A model specifying one factor at each level (C1), compared to our hypothesized 10 factors at the within- and 5 factors at the between-level (C3), fitted the data substantially worse. We omitted the same items of psychological detachment and relaxation as in Study 1 (C4). Furthermore, we tested a 10–15 factor solution (C5), in which we additionally estimated variables measured at the day level on the person level. This solution also indicated good model fit (see Supplemental Material, Supplemental Table S4, Hu and Bentler, 1999), thus confirming our hypothesized factor structure. Since we had a cross-lagged daily diary design, we tested measurement invariance via nested MCFAs (Putnick and Bornstein, 2016). Results confirmed strong factorial invariance (see Supplemental Material, Supplemental Table S5).

Table 3 depicts results from M1. Figure 3 depicts a visual representation of within-person results from M2 with standardized estimates. Supplemental Tables S6–S8 in the Supplemental Material depict parameter estimates for all models. Quantitative (b_{work overload} = 0.23, p_{work overload} = 0.01) and one qualitative work stressor (b_{information problems} = 0.23, p_{information problems} = 0.04; b_{work interruptions} = −0.22, p_{work interruptions} = 0.07) predicted T1 unfinished work tasks. In partial support of Hypothesis 1, T1 unfinished work tasks were positively related to T2 proactive (H1a, b = 0.06, p = 0.003) but not related to T2 reactive TASW (H1b, b = 0.03, p = 0.10). Injunctive norms were positively related to T2 proactive TASW (H2a, b = 0.08, p = 0.001) and marginally positively related to T2 reactive TASW (H2b, b = 0.04, p = 0.058). However, descriptive norms were not related to T2 proactive TASW (H2b, b = −0.04, p = 0.26) or T2 reactive TASW (H2b, b = −0.02, p = 0.49), thus partially supporting Hypothesis 2. In partial support of Hypothesis 3, T2 proactive TASW was negatively related to T3 psychological detachment (H3a, b = −0.49, p = 0.01), whereas T2 reactive TASW was not (H3b, b = −0.29, p = 0.14). The within-mediation-effect of T2 proactive TASW between T1 and T3 psychological detachment was significant (T1_{psychological detachment} → T2_{proactive TASW} → T3_{psychological detachment}, b = 0.02, 95% CI (0.003; 0.05)). Because our research design didn’t allow us to directly test the proposed mediating effect of Hypothesis 4 with temporally separated variables, we tested mediating models via Bayesian estimation to test proactive TASW’s within-person effect on relaxation via psychological detachment. Results revealed significant mediating effects in support of Hypothesis 4a (T1_{unfinished work tasks} → T2_{proactive TASW} → T3_{psychological detachment}, b = −0.03, 95% CI (−0.07; −0.003); T1_{unfinished work tasks} → T2_{psychological detachment} → T3_relaxation, b = −0.04, 95% CI (−0.08; −0.002)). Because reactive TASW was not significantly related to the mediators, Hypothesis 4b was not confirmed. Contrary to Hypothesis 5, T2 proactive TASW showed stronger effects for T3 psychological detachment (H5a, b = −0.49, p = 0.01) than T2 reactive TASW, which showed no significant effect (H5a, b = −0.29, p = 0.14). Both forms of T2 TASW showed no significant direct effect for T3 relaxation experiences (H5b, b_{proactive TASW} = −0.23, p_{proactive TASW} = 0.23; H5b, b_{reactive TASW} = −0.15, p_{reactive TASW} = 0.47). Furthermore, psychological detachment was negatively related to proactive TASW in later timepoints. (T1_{psychological detachment} → T2_{proactive TASW}, b = −0.05, p = 0.002; T2_{psychological detachment} → T3_{proactive TASW}, b = −0.05, p = 0.04). Finally, psychological detachment was positively related to relaxation experiences in later timepoints (T1_{psychological detachment} → T2_relaxation, b = 0.25, p = < 0.001; T2_{psychological detachment} → T3_relaxation, b = 0.28, p = < 0.001).

Table 3.

Between-person results for M1.

Outcome and predictor	b	SE	p	β*
T2 proactive TASW
T1 unfinished work tasks	0.07	0.02	0.00	0.21
T0 injunctive norms	0.08	0.03	0.001	0.45
T0 descriptive norms	−0.04	0.03	0.26	−0.14
T2 reactive TASW
T1 unfinished work tasks	0.03	0.02	0.16	0.11
T0 injunctive norms	0.04	0.02	0.06	0.32
T0 descriptive norms	−0.02	0.03	0.49	−0.11
T1 unfinished work tasks
T0 work overload	0.23	0.09	0.01	0.41
T0 information problems	0.23	0.11	0.04	0.29
T0 work interruptions	−0.22	0.12	0.07	−0.32

b: unstandardized estimate; SE: standard error; β: standardized estimate.

Results obtained via Bayesian estimation.

Figure 3.

Results of Study 2.

Brief discussion

In Study 2, we replicated and expanded upon our results from Study 1. Our results show that both quantitative (overtaxing) and qualitative (hindering) work stressors are positively associated with unfinished work tasks. When employees experienced higher external pressures in terms of unfinished work tasks, they engaged in more proactive, but not in more reactive TASW-behavior in the early evening. Furthermore, injunctive norms but not descriptive norms were positively associated with proactive and reactive TASW-behavior, suggesting that normative pressure from the employer had a stronger impact on employees TASW-behavior than colleagues’ segmentation norms. In line with results from Study 1 proactive TASW but not reactive TASW during the early evening was negatively related to daily psychological detachment during the later evening. This indicates that especially proactive TASW impedes employees’ ability to detach from work in the evening. Furthermore, we found that a lack of initial detachment after work leads to more proactive TASW and subsequent detachment difficulties. This contradicts the proposed beneficial effect of more autonomous TASW (Heissler et al., 2022; Weigelt and Syrek, 2017) and indicates that TASW impedes, rather than facilitates, recovery. Our findings further suggest a detrimental effect of TASW on relaxation via psychological detachment: (a) daily work stressors promote proactive TASW, which impedes employees’ ability to detach from work later that evening and (b) if employees’ ability to detach from work in the evening is impeded as a result of continued activation from work, relaxation experiences in the evening are negatively affected. However, the results of the mediation analyses must be interpreted with caution, as model fit decreased slightly from M2 to M3 (see Supplemental Material, Supplemental Table S4) and credible intervals didn’t include, but were close to zero. Consistent with the results of Study 1, proactive TASW was more detrimental for recovery experiences in the evening than reactive TASW.

General discussion

Drawing on Self-Determination Theory (Deci and Ryan, 2000), Action Regulation Theory (Hacker, 2003; Leitner et al., 1987), and the stressor-detachment model (Sonnentag and Fritz, 2015) we conducted two consecutive daily diary studies to investigate relationships of external pressures, autonomy in TASW and recovery experiences. We distinguished between two forms of TASW, reactive and proactive TASW, describing external, and relative internal regulation styles for TASW. We investigated two dimensions of external pressures, work stressors and normative pressures, respectively, as predictors of proactive, and reactive TASW-behaviors. Study 1 provided initial evidence that quantitative (overtaxing) work stressors promote proactive and reactive TASW-behaviors in the evening. We replicated this finding in Study 2 and found that qualitative (hindering) work stressors also predict TASW in the evening. Regarding normative pressures for TASW, only injunctive but not descriptive norms positively predicted TASW. Furthermore, our studies provided evidence for detrimental effects of proactive TASW, but not of reactive TASW, on psychological detachment and relaxation in the evening. We found a serial indirect effect of work overload during the workday on lower levels of relaxation in the evening via higher levels of proactive TASW and subsequent lower levels of psychological detachment. Overall, our results neither support the argument of a beneficial effect of more autonomous TASW, nor our hypothesis of a less detrimental effect. Rather, they highlight its even stronger detrimental consequences for recovery experiences.

Theoretical implications

Our research contributes to the knowledge about the origin of TASW-behaviors. Drawing on SDT (Deci and Ryan, 2000), our findings provide evidence for the argument that TASW-behaviors constitute extrinsically motivated behaviors that emerge because of the internalization of external pressures. Drawing on Action Regulation Theory (Hacker, 2003; Leitner et al., 1987), we distinguished quantitative (overtaxing) and qualitative (hindering) work stressors. Both types of work stressors might be perceived as introjected motivators that initially exert external pressure, which fosters reactive TASW. Over time, these external pressures become more internalized, thereby facilitating more autonomous, proactive forms of TASW. Moreover, by considering Action Regulation Theory’s (Hacker, 2003) differentiation of work stressors within the stressor-detachment model (Sonnentag and Fritz, 2015), our arguments open avenues for future research on work stressors’ differential impact on recovery experiences. Future research might additionally examine the impact of (learning) demands (Glaser et al., 2015) on TASW, and recovery experiences. Furthermore, by considering qualitative (overtaxing) and quantitative (hindering) work stressors, we broaden the theoretical framework from which practical interventions that aim to reduce TASW might be developed. Ultimately, this may enhance employees’ well-being, health, and work performance by improving recovery processes (Wendsche et al., 2021).

Regarding normative pressures, our research considered the relative impact of injunctive and descriptive norms for TASW. Our results show that especially employer-imposed norms promote TASW-emergence. Whereas descriptive norms only convey what behaviors colleagues perform, injunctive norms may exert stronger motivation to comply, since they confer explicit information of what is expected of employees (Fishbein and Ajzen, 2010). Considering that organizations can leverage more bases of social power (Fishbein and Ajzen, 2010; French and Raven, 1959), injunctive norms may exert greater influence on shaping the intentions and performance of employee behavior compared to descriptive norms.

We contribute to the recovery literature by examining the consequences of TASW for recovery experiences depending on autonomy. Previous literature (e.g. Heissler et al., 2022; Mazmanian et al., 2013; Reinke and Ohly, 2021) has discussed detrimental but also beneficial effects of TASW based on its autonomy-enhancing potential. Our research is the first to empirically examine self-determination in TASW. Our results indicate an overall detrimental effect of both forms of TASW on psychological detachment. However, contrary to our hypothesis, the findings revealed even more detrimental effects of more autonomous-regulated TASW-behavior in both studies. One possible explanation may be differing working time arrangements. The high proportion of employees with temporal flexibility in organizing their workday in our sample may have led to overestimating the effect of proactive TASW for the population. Workers with higher time flexibility might extend their workday via proactive TASW in the evening while working less during “normal” working hours. Thus, they may impair recovery experiences in the evening more severely. However, a post-hoc analysis that compares employees with and without core working hours revealed no substantial differences in effects between these subgroups in our models. Methodologically, overlapping variances in proactive and reactive TASW (see Tables 1 and 2) may have suppressed effects in our samples as proactive TASW was more prevalent in both studies. A further explanation may relate to persistence in work behaviors. Behaviors that are regulated with greater autonomy are pursued with greater persistence (Deci and Ryan, 1985). Therefore, employees might engage in subsequent supplemental work tasks when performing more autonomous regulated TASW, rather than merely meeting the minimum requirements imposed by external pressures. Therefore, employees may spend more instead of less time on supplemental work if the work activity is proactive compared to reactive, which impedes the recovery process more severely.

Another explanation may relate to the situational regulation style that underlies proactive TASW. We assumed proactive TASW to reflect more autonomous-determined situational behaviors. However, proactive TASW may include several internalized regulation styles, some of which are located on the lower end of the internalization continuum. These may exhibit different effect strengths on recovery experiences. For example, situational regulation for proactive TASW might “get stuck” at introjected regulation or compartmentalized identification, thus representing a self-initiated, however, less self-determined behavior. Introjected proactive TASW might be responsible for more detrimental consequences for recovery experiences, whereas less detrimental consequences are expected for more internalized self-determined TASW-behaviors (Ryan and Deci, 2017). Furthermore, compartmentalized identified-regulation of proactive TASW might also exhibit more detrimental consequences. Compartmentalized identification can lead to self-deception about the autonomy over actions (Ryan and Deci, 2017). They refer to strong introjections that are perceived by individuals as identifications that “conflict with either intrinsic motives and needs or with other deeply internalized values” (Ryan and Deci, 2017: 201). Therefore, employees might also experience higher levels of activation due to cognitive dissonance that hinders recovery experiences. For example, participants in a qualitative study of academics reported that external pressures in the form of completing their high workload from their job drove internalization processes (Weiss and Ortlieb, 2025). Consequently, they reported being “haunted” by work problems and engaged in proactive supplemental work, which in turn reduced their recovery experiences.

Our research contributes to the recovery literature by providing empirical evidence for the proposed mediator effect of psychological detachment in the stressor-strain process in the context of TASW (Sonnentag and Fritz, 2015). Our results offer initial evidence for detrimental effects of work stressors on relaxation via enhanced TASW and subsequently impaired psychological detachment. The impairment of the strain-alleviating process after work (Meijman and Mulder, 1998; Sonnentag and Fritz, 2007) constitutes a risk factor for burnout (Schaufeli and Taris, 2005). Exhaustion, one of four core dimensions of the burnout syndrome, may intensify when employees are unable to unwind from work due to TASW and the continued impairment of recovery. Furthermore, the negative impact of TASW on relaxation may offer an explanation for earlier research findings regarding TASW and affect responses (e.g. Eichberger et al., 2021; Reinke and Ohly, 2021). TASW may increase negative affect and lower positive affect in the evening as relaxation experiences, defined as states of low activation with increased positive affect (Sonnentag and Fritz, 2007), are thwarted.

Practical implications

This research has implications for employees and human resource management (HRM) in organizations regarding the processes of work spill-over into the leisure domain and recovery impediments. It is crucial for HRM to mitigate TASW, as it adversely affects employees’ recovery and potentially renders them ineffective. In the short term, resources such as energy are not restored. In the long term, TASW may lead to psychophysiological health impairments such as burnout. Our research implicates that organizations have several levers in preventing TASW, as it stems from external pressures from the work environment. First, we recommend that HRM should structurally monitor and eliminate different types of work stressors by performing regular risk assessments for psychosocial hazards in the workplace (International Organization for Standardization, 2021). Our research identified quantitative work stressors (i.e. too much work) as well as qualitative work stressors (i.e. obstacles in goal achievement) that may lead to TASW. HRM and supervisors should establish interventions to reduce both kinds of work stressors to mitigate TASW. For example, HRM could offer health-oriented leadership training (Wegge et al., 2014) to help supervisors become healthy role models and to improve the distribution and coordination of work, thereby preventing employee overtaxation. Moreover, line managers and employees may identify and eliminate redundant or inefficient work processes to decrease work overload. Furthermore, increasing decision latitude and working time flexibility for employees fosters well-being and health while mitigating TASW (Shifrin and Michel, 2022).

Second, organizations should lower normative pressures to engage in work activities beyond compensated working hours. Our results suggest that especially employer-based normative pressure contributes to TASW-behaviors. Therefore, employee relations departments should implement strategies to reduce availability expectations beyond contractually obligated on-call duties by promoting taking time off work. In addition, compliance and legal departments should ensure compliance with labor laws regarding overtime. While colleagues’ segmentation norms had no significant influence on TASW-behaviors in our study, they may nevertheless exert influence over individuals’ behavior (Fishbein and Ajzen, 2010; Palm et al., 2020). Therefore, supervisors should lead by example (Yaffe and Kark, 2011) by respecting employees’ time off and demonstrating a healthy life-domain balance themselves. For example, HRM may implement programs to increase self-awareness and self-regulation of supervisors and employees (Day, 2000).

Third, our results suggest that TASW does not serve as a coping mechanism to better detach from work. TASW should not be interpreted as an autonomy-enhancing phenomenon whereby employees have the flexibility to complete work tasks in their leisure time. Rather, it should be understood as a detrimental obligation to keep on working due to external pressures. By acknowledging TASW as a form of obligation rather than autonomy, organizations might more effectively support employees’ well-being. HRM could implement easily accessible software tools to record overtime and should offer compensation for additional work in the form of extra leave to compensate loss of recovery time.

HRM, supervisors, and employees should be particularly mindful of the detrimental consequences of proactive TASW. Legislation (e.g. right to disconnect, Eurofound, 2023b) and concrete actions against supplemental work (e.g. implementing restrictions in receiving messages or calls via work channels) predominantly aim to reduce reactive TASW-behavior. Interventions to restrict or discourage proactive TASW could pose a more complicated challenge. Some organizations restricted access to work-related content. However, concerns were expressed that this may diminish employees’ working time flexibility, which helps to manage work demands (Shifrin and Michel, 2022). Furthermore, HRM and supervisors should be aware of implicit reward systems that promote TASW-behaviors. These may range from more overt practices of managerial or peer recognition for additional effort that has been invested through supplemental work, to more subtle forms such as expedited career advancements or permanent contracts. These may be especially relevant in fields with high job insecurity (e.g. academia, see Weiss and Ortlieb, 2025), as employees compete for limited resources or positions (Swab and Johnson, 2019).

Limitations and future research directions

Our research has limitations that should be considered. First, despite replicating our results in a cross-lagged daily diary study and employing Bayesian estimation (Yuan and MacKinnon, 2009), generalizability may be limited due to relatively small sample sizes in both studies. Although our samples encompassed a broad range of sectors, flextime work arrangements, and age groups, they were predominantly highly educated. Future research could aim to replicate our findings in a larger, more diverse sample. Moreover, participants in both studies reported rather few reactive and proactive TASW-behaviors, which might have suppressed effects. Future studies could target populations which exhibit TASW-behaviors more frequently. Furthermore, both samples exhibited small attrition effects (see Supplemental Material). We accounted for effects of working hours by including it as an auxiliary variable. However, employees with higher levels of proactive and reactive TASW were more likely to miss subsequent time points, thus impairments of psychological detachment or relaxation were potentially underreported.

Second, concerning our research designs, in Study 1 we found a mediating effect of proactive TASW on relaxation via psychological detachment, but could not establish chronological order of these variables. In Study 2, temporal precedence was given, but we were unable to test certain mediation effects due to too few measurement points during the evening. Future studies could implement a study design in which TASW, psychological detachment and relaxation are measured at three time points during the evening. Furthermore, our study design captured intraday dynamics. Mid-term effects of TASW on recovery may differ from short-term results. Since TASW stems from work stressors which vary from week to week or month to month, employees might tolerate short-term recovery impairments if they believe to sufficiently recover in the future.

Third, regarding the measurement of proactive and reactive TASW, both scales exhibited low reliability coefficients in both studies. TASW was assessed via employees’ use of different devices. The use of varying ICT at varying timepoints might not necessarily correlate and imply an index rather than a formative construct (Bollen and Bauldry, 2011) and therefore explain low internal consistency. Furthermore, our measures of proactive and reactive TASW didn’t directly assess more autonomous or more external regulation of TASW-behaviors, but rather self- or other-initiation of TASW-behaviors. As outlined above, proactive TASW could be predominantly motivated by regulation styles that might be located on the lower end of the internalization continuum. Future research may develop new measurement instruments for TASW that distinguish between forms of TASW-behaviors based on their specific regulation styles (Deci and Ryan, 2000). Furthermore, objective data collection methods should be explored in the context of TASW. For instance, smartphone sensor data could help to mitigate limitations of subjectively perceived ICT-use, which tends to be underreported by study participants (Andrews et al., 2015; Ellis et al., 2019). Future research might also investigate when and under which conditions TASW becomes more internalized. For example, longitudinal studies could examine how external pressures promote the transformation of extrinsic motivated TASW-behaviors to introjected or identified TASW-behaviors over time. Especially employees with low job tenure may be suitable to study such internalization processes.

Finally, research on supplemental work should expand beyond the mere technology-assisted component. Future research should investigate differential effects of TASW-behaviors that are attributable solely to its ICT-component, and other important and perhaps widespread forms of “offline” supplemental work. Additionally, researchers should more precisely distinguish between supplemental work which encompasses observable behaviors and forms of “invisible” supplemental work. For example, work activities of knowledge workers contain functional (e.g. problem-solving pondering or positive work reflection) as well as dysfunctional cognitive components (e.g. work related rumination; Weigelt et al., 2019), which are typically considered to be part of the work-role when performed during working hours. They may also be classified as supplemental work when they are not compensated.

Conclusion

Our research provides insights into intraday dynamics of external pressures, regulation of TASW-behavior, and consequences for recovery experiences. Through the theoretical lens of Self-Determination Theory (Deci and Ryan, 2000), Action Regulation Theory (Hacker, 2003), and the stressor-detachment model (Sonnentag and Fritz, 2015), we examined the role of external pressures, self-determination in TASW-behaviors and their consequences for recovery experiences in two diary studies. Our findings suggest that TASW-behaviors are extrinsically motivated behaviors due to external pressures, that is, quantitative (overtaxing) and qualitative (hindering) work stressors, and social norms (especially employer expectations). Employees engage in TASW to cope with work stressors or in response to normative pressures. Contrary to the proposed beneficial effect of more autonomous TASW-behavior, self-initiated proactive TASW showed more detrimental effects for recovery experiences than other-initiated reactive TASW. Overall, our findings emphasize the importance of considering the multifaceted nature of TASW and its detrimental impact on recovery.

Supplemental Material

sj-docx-1-gjh-10.1177_23970022251356383 – Supplemental material for External pressures on technology-assisted supplemental work: Motivation regulation styles and the impact on daily recovery

Supplemental material, sj-docx-1-gjh-10.1177_23970022251356383 for External pressures on technology-assisted supplemental work: Motivation regulation styles and the impact on daily recovery by Ferdinand Baierer and Jürgen Glaser in German Journal of Human Resource Management

Footnotes

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Ethical considerations

Ethical approval was granted by the Advisory Board for Ethical Issues in Scientific Research at the University of Innsbruck (26/2021).

Consent to participate

Respondents gave written consent.

Consent for publication

Not applicable.

ORCID iD

Ferdinand Baierer

Supplemental material

Supplemental material for this article is available online.

References

Andrews

Ellis

Shaw

, et al. (2015) Beyond self-report: Tools to compare estimated and real-world smartphone use. PLoS One 10(10): e0139004.

Arlinghaus

Nachreiner

(2014) Health effects of supplemental work from home in the European Union. Chronobiology International 31(10): 1100–1107.

Asparouhov

Muthén

(2025) Multiple Imputation with Mplus. Available at: https://www.statmodel.com/download/Imputations7.pdf

Bakker

Demerouti

Sanz-Vergel

(2023) Job demands–resources theory: Ten years later. Annual Review of Organizational Psychology and Organizational Behavior 10(1): 25–53.

Bennett

Bakker

Field

(2018) Recovery from work-related effort: A meta-analysis. Journal of Organizational Behavior 39(3): 262–275.

Bollen

Bauldry

(2011) Three Cs in measurement models: Causal indicators, composite indicators, and covariates. Psychological Methods 16(3): 265–284.

Braukmann

Schmitt

Ďuranová

, et al. (2018) Identifying ICT-Related affective events across life domains and examining their unique relationships with employee recovery. Journal of Business and Psychology 33(4): 529–544.

Chirkov

Ryan

Kim

, et al. (2003) Differentiating autonomy from individualism and independence: A self-determination theory perspective on internalization of cultural orientations and well-being. Journal of Personality and Social Psychology 84(1): 97–110.

Crawford

Lepine

Rich

(2010) Linking job demands and resources to employee engagement and burnout: A theoretical extension and meta-analytic test. Journal of Applied Psychology 95(5): 834–848.

10.

Day

(2000) Leadership development: A review in context. Leadership Quarterly 11(4): 581–613.

11.

De Bloom

Geurts

SAE

Kompier

MAJ

(2013) Vacation (after-) effects on employee health and well-being, and the role of vacation activities, experiences and sleep. Journal of Happiness Studies 14(2): 613–633.

12.

Deci

Ryan

(1985) The general causality orientations scale: Self-determination in personality. Journal of Research in Personality 19(2): 109–134.

13.

Deci

Ryan

(2000) The ‘what’ and ‘why’ of goal pursuits: Human needs and the self-determination of behavior. Psychological Inquiry 11(4): 227–268.

14.

Demerouti

Bakker

Nachreiner

, et al. (2001) The job demands-resources model of burnout. Journal of Applied Psychology 86(3): 499–512.

15.

Derks

van Mierlo

Schmitz

(2014) A diary study on work-related smartphone use, psychological detachment and exhaustion: Examining the role of the perceived segmentation norm. Journal of Occupational Health Psychology 19(1): 74–84.

16.

Dettmers

Vahle-Hinz

Bamberg

, et al. (2016) Extended work availability and its relation with start-of-day mood and cortisol. Journal of Occupational Health Psychology 21(1): 105–118.

17.

Diaz

Chiaburu

Zimmerman

, et al. (2012) Communication technology: Pros and cons of constant connection to work. Journal of Vocational Behavior 80(2): 500–508.

18.

Ďuranová

Ohly

(2016) Persistent Work-Related Technology Use, Recovery and Well-Being Processes. SpringerBriefs in Psychology. Cham: Springer International Publishing.

19.

Eichberger

Derks

Zacher

(2021) Technology-assisted supplemental work, psychological detachment, and employee well-being: A daily diary study. Journal of Human Resource Management 35(2): 199–223.

20.

Eichberger

Derks

Zacher

(2022) A daily diary study on technology-assisted supplemental work, unfinished tasks, and sleep: The role of problem-solving pondering. International Journal of Stress Management 29(1): 61–74.

21.

Eichberger

Zacher

(2021) Toward definitional clarity of technology-assisted supplemental work: A bridge over muddied waters. Industrial and Organizational Psychology 14(3): 428–431.

22.

Ellis

Davidson

Shaw

, et al. (2019) Do smartphone usage scales predict behavior? International Journal of Human-Computer Studies 130: 86–92.

23.

Ethica Data (2023) Ethica. Available at: https://ethicadata.com; https://avicennaresearch.com/.

24.

Eurofound (2022) European Working Conditions Survey Integrated Data File, 1991-2015. Available at: https://beta.ukdataservice.ac.uk/datacatalogue/doi/?id=7363#!#9.

25.

Eurofound (2023a) European Working Conditions Telephone Survey, 2021. Available at: https://beta.ukdataservice.ac.uk/datacatalogue/doi/?id=9026#!#2.

26.

Eurofound (2023b) Right to disconnect: Implementation and impact at company level. Working conditions and sustainable work. Luxembourg: Publications Office of the European Union. DOI: 10.2806/430915.

27.

Fenner

Renn

(2004) Technology-assisted supplemental work: Construct definition and a research framework. Human Resource Management 43(2-3): 179–200.

28.

Fenner

Renn

(2010) Technology-assisted supplemental work and work-to-family conflict: The role of instrumentality beliefs, organizational expectations and time management. Human Relations 63(1): 63–82.

29.

Fishbein

Ajzen

(2010) Predicting and Changing Behavior: The Reasoned Action Approach. New York, NY: Psychology Press.

30.

French

JRP

Jr. Raven

(1959) The bases of social power. In: Carthwright D (ed.) Studies in Social Power. Oxford, England: Univer. Michigan, pp.150–167.

31.

Frese

Zapf

(1994) Action as the core of work psychology: A German approach. In: Triandis HC, Dunnette MD and Hough LM (eds) Handbook of Industrial and Organizational Psychology, Vol. 4, 2nd edn. Palo Alto, CA: Consulting Psychologists Press, pp.271–340.

32.

Gabriel

Podsakoff

Beal

, et al. (2019) Experience sampling methods: A discussion of critical trends and considerations for scholarly advancement. Organizational Research Methods 22(4): 969–1006.

33.

Gadeyne

Darouei

Verbruggen

, et al. (2023) Do behavioural intentions matter? A diary study on work-related ICT-use after work hours. European Journal of Work and Organizational Psychology 32(5): 746–759.

34.

Gadeyne

Verbruggen

Delanoeije

, et al. (2018) All wired, all tired? Work-related ICT-use outside work hours and work-to-home conflict: The role of integration preference, integration norms and work demands. Journal of Vocational Behavior 107: 86–99.

35.

Gagné

Deci

(2005) Self-determination theory and work motivation. Journal of Organizational Behavior 26(4): 331–362.

36.

Geldhof

Preacher

Zyphur

(2014) Reliability estimation in a multilevel confirmatory factor analysis framework. Psychological Methods 19(1): 72–91.

37.

Geurts

Sonnentag

(2006) Recovery as an explanatory mechanism in the relation between acute stress reactions and chronic health impairment. Scandinavian Journal of Work Environment & Health 32(6): 482–492.

38.

Glaser

Hornung

Höge

, et al. (2020) Das Tätigkeits- und Arbeitsanalyseverfahren (TAA). Innsbruck: innsbruck university press.

39.

Glaser

Seubert

Hornung

, et al. (2015) The impact of learning demands, work-related resources, and job stressors on creative performance and health. Journal of Personnel Psychology 14(1): 37–48.

40.

Göllner

Rau

(2021) Erreichbarkeit: Symptom oder Merkmal der Arbeit? Was bedeutet das Ergebnis für die Arbeitsgestaltung? Zeitschrift fur Arbeitswissenschaft 75(1): 1–14.

41.

Greiner

Leitner

(1989) Assessment of job stress: The RHIA instrument. In: Landau K and Rohmert W (eds) Recent Developments in Job Analysis. London: Routledge, pp.53–66.

42.

Hacker

(2003) Action regulation theory: A practical tool for the design of modern work processes? European Journal of Work and Organizational Psychology 12(2): 105–130.

43.

Hacker

Sachse

(2023) Allgemeine Arbeitspsychologie: Psychische Regulation von Arbeitstätigkeiten, 4th edn. Zürich: vdf Hochschulverlag.

44.

Heissler

Kern

Ohly

(2022) When thinking about work makes employees reach for their devices: A longitudinal autoregressive diary study. Journal of Business and Psychology 37(5): 999–1016.

45.

Höge

(2011) Perceived flexibility requirements at work and the entreployee-work-orientation. Concept and Measurement 4(1): 3–21.

46.

Höge

Palm

Strecker

(2016) Anforderungen an selbstorganisierte Arbeit und das Verhältnis von Arbeit und Privatleben. Zur Rolle von passiver und aktiver IuK-Technologie-Nutzung in der Freizeit. Wirtschaftspsychologie 2: 35–43.

47.

Howard

Gagné

Bureau

(2017) Testing a continuum structure of self-determined motivation: A meta-analysis. Psychological Bulletin 143(12): 1346–1377.

48.

Bentler

(1999) Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal 6(1): 1–55.

49.

International Organization for Standardization (2021) ISO 45003:2021 Occupational Health and Safety Management — Psychological Health and Safety at Work — Guidelines for Managing Psychosocial Risks. Geneva: International Organization for Standardization. Available at: https://www.iso.org/standard/64283.html (accessed 13 May 2025).

50.

Kreiner

(2006) Consequences of work-home segmentation or integration: A person-environment fit perspective. Journal of Organizational Behavior 27(4): 485–507.

51.

Kühner

Rudolph

Derks

, et al. (2023) Technology-assisted supplemental work: A meta-analysis. Journal of Vocational Behavior 142: 103861.

52.

Laguerre

Barnes-Farrell

(2025) Bringing self-determination theory to the forefront: Examining how human resource practices motivate employees of all ages to succeed. Journal of Business and Psychology 40(1): 1–37.

53.

Lanaj

Johnson

Barnes

(2014) Beginning the workday yet already depleted? Consequences of late-night smartphone use and sleep. Organizational Behavior and Human Decision Processes 124(1): 11–23.

54.

Lavergne

Pelletier

(2016) Why are attitude–behaviour inconsistencies uncomfortable? Using motivational theories to explore individual differences in dissonance arousal and motivation to compensate. Motivation and Emotion 40(6): 842–861.

55.

Lazarus

Folkman

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

56.

Leitner

Volpert

Greiner

, et al. (1987) Analyse psychischer Belastung in der Arbeit. Das RHIA-Verfahren. Manual und Antwortblätter. Köln: Verlag TÜV Rheinland.

57.

Lewin

(1939) Field theory and experiment in social psychology: Concepts and methods. American Journal of Sociology 44: 868–896.

58.

MacDonald

(1999) Test Theory: A Unified Treatment. New York, NY: Lawrence Erlbaum Associates Publishers.

59.

Manning

(2009) The effects of subjective norms on behaviour in the theory of planned behaviour: A meta-analysis. British Journal of Social Psychology 48(Pt 4): 649–705.

60.

Mazmanian

Orlikowski

Yates

(2013) The autonomy paradox: The implications of mobile email devices for knowledge professionals. Organization Science 24(5): 1337–1357.

61.

Meijman

Mulder

(eds) (1998) Psychological aspects of workload. In: Drenth PJD, Thierry H, and de Wolff CJ (eds) Handbook of Work and Organizational: Work Psychology, vol. 2, 2nd edn. Handbook of work and organizational psychology (2nd ed.). Hove, England: Psychology Press/Erlbaum (UK) Taylor & Francis, pp.5–33.

62.

Muthén

(2010) Bayesian Analysis In Mplus: A Brief Introduction. Available at: https://www.statmodel.com/download/IntroBayesVersion%203.pdf

63.

Muthén

(2023) Mplus 8.10. Available at: https://www.statmodel.com/.

64.

Nawijn

De Bloom

Geurts

(2013) Pre-Vacation time: Blessing or burden? Leisure Sciences 35(1): 33–44.

65.

Ohly

Latour

(2014) Work-related smartphone use and well-being in the evening. Journal of Personnel Psychology 13(4): 174–183.

66.

Palm

Seubert

Glaser

(2020) Understanding employee motivation for work-to-nonwork integration behavior: A reasoned action approach. Journal of Business and Psychology 35(5): 683–696.

67.

Park

Fritz

Jex

(2011) Relationships between work-home segmentation and psychological detachment from work: The role of communication technology use at home. Journal of Occupational Health Psychology 16(4): 457–467.

68.

Preacher

Zyphur

Zhang

(2010) A general multilevel SEM framework for assessing multilevel mediation. Psychological Methods 15(3): 209–233.

69.

Putnick

Bornstein

(2016) Measurement invariance conventions and reporting: The state of the art and future directions for psychological research. Developmental Review 41: 71–90.

70.

Reinke

Gerlach

(2022) Linking availability expectations, bidirectional boundary management behavior and preferences, and employee well-being: An integrative study approach. Journal of Business and Psychology 37(4): 695–715.

71.

Reinke

Ohly

(2021) Double-edged effects of work-related technology use after hours on employee well-being and recovery: The role of appraisal and its determinants. Journal of Human Resource Management 35(2): 224–248.

72.

Richardson

Benbunan-Fich

(2011) Examining the antecedents of work connectivity behavior during non-work time. Information and Organization 21(3): 142–160.

73.

Ryan

(1982) Control and information in the intrapersonal sphere: An extension of cognitive evaluation theory. Journal of Personality and Social Psychology 43(3): 450–461.

74.

Ryan

Deci

(2017) Self-Determination Theory: Basic Psychological Needs in Motivation, Development, and Wellness. New York, NY, US: The Guilford Press.

75.

Schaufeli

Taris

(2005) The conceptualization and measurement of burnout: Common ground and worlds apart. Work and Stress 19(3): 256–262.

76.

Semmer

(1984) Streßbezogene Tätigkeitsanalyse: Psychologische Untersuchungen zur Analyse von Streß am Arbeitsplatz. Beltz-Forschungsberichte: Beltz.

77.

Shifrin

Michel

(2022) Flexible work arrangements and employee health: A meta-analytic review. Work and Stress 36(1): 60–85.

78.

Sonnentag

Fritz

(2007) The recovery experience questionnaire: Development and validation of a measure for assessing recuperation and unwinding from work. Journal of Occupational Health Psychology 12(3): 204–221.

79.

Sonnentag

Fritz

(2015) Recovery from job stress: The stressor-detachment model as an integrative framework. Journal of Organizational Behavior 36(S1): S72–S103.

80.

Swab

Johnson

(2019) Steel sharpens steel: A review of multilevel competition and competitiveness in organizations. Journal of Organizational Behavior 40(2): 147–165.

81.

Syrek

Weigelt

Peifer

, et al. (2017) Zeigarnik’s sleepless nights: How unfinished tasks at the end of the week impair employee sleep on the weekend through rumination. Journal of Occupational Health Psychology 22(2): 225–238.

82.

Van Den Broeck

Howard

Van Vaerenbergh

, et al. (2021) Beyond intrinsic and extrinsic motivation: A meta-analysis on self-determination theory’s multidimensional conceptualization of work motivation. Organizational Psychology Review 11(3): 240–273.

83.

Wegge

Shemla

Haslam

(2014) Leader behavior as a determinant of health at Work: Specification and evidence of five key pathways. Journal of Human Resource Management 28(1-2): 6–23.

84.

Weigelt

Gierer

Syrek

(2019) My mind is working overtime-towards an integrative perspective of psychological detachment, work-related rumination, and work reflection. International Journal of Environmental Research and Public Health 16(16): 2987.

85.

Weigelt

Syrek

(2017) Ovsiankina’s great relief: How supplemental work during the weekend may contribute to recovery in the face of unfinished tasks. International Journal of Environmental Research and Public Health 14(12): 12. MDPI AG: 1606.

86.

Weiss

Ortlieb

(2025) Professional–personal boundary work: Individuals torn between integration and segmentation. Journal of Human Resource Management 39(2): 121–147.

87.

Wendsche

De Bloom

Syrek

, et al. (2021) Always on, never done? How the mind recovers after a stressful workday? Journal of Human Resource Management 35(2): 117–151.

88.

Yaffe

Kark

(2011) Leading by example: The case of leader OCB. Journal of Applied Psychology 96(4): 806–826.

89.

Yuan

MacKinnon

(2009) Bayesian mediation analysis. Psychological Methods 14(4): 301–322.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.06 MB