ABSTRACT
The popularity of remote work continues to rise, but uncertainty remains about how it influences employee well-being. We extend the Demand-Control-Person (DCP) model to test both person and job factors as important considerations in remote work, suggesting that emotional stability influences the utility of autonomy as a job resource in protecting employees from strain. Additionally, we test self-determination theory (SDT), positioning need satisfaction for autonomy, relatedness, and competence as mechanisms explaining the relationship between remote work and strain. In two field studies, high–emotional stability employees reporting high levels of autonomy experienced the lowest levels of strain, with negative relationships between extent of remote work and strain. In contrast, low–emotional stability employees who also have high autonomy appear more susceptible to strain, and this may increase when they work remotely more often. Our multilevel structural equation modelling revealed that high–emotional stability employees with high autonomy appear best positioned to meet their needs for autonomy and relatedness, even when remote work is more frequent; these in turn reduced the likelihood of strain. Thus, our results support the DCP and SDT models, revealing theoretical and practical implications for designing and managing remote work arrangements.
Acknowledgements
The authors would like to thank L. A. Witt, Robert T. Keller, Steven C. Currall, and Corina Rice for their valuable input during the early stages of this research.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes
1. Although other personality traits may also be important in this context, we focused on emotional stability in order to directly test and extend the DCP model in conjunction with the extent of remote work (Rubino et al., Citation2012).
2. We also tested this three-way interaction using conscientiousness and extraversion in place of emotional stability, but neither was significant for any strain outcome. A few significant main effects emerged, as follows: for exhaustion: conscientiousness (b = −.34, p < .01); for disengagement: conscientiousness (b = −.22, p < .01) and extraversion (b = −.21, p < .01); and for dissatisfaction: conscientiousness (b = −.29, p = .01). Thus, conscientiousness and extraversion both appear to be helpful in reducing some forms of strain, but these did not alter the effects of the extent of remote work on strain in Study 1.
3. For Study 2, we also collected conscientiousness. In running the same analyses, the three-way interaction was not significant in directly predicting any form of strain (Hypothesis 2). In testing Hypothesis 3, the three-way interaction including conscientiousness was not significant in predicting competence or relatedness, but it was significant in predicting autonomy need satisfaction (b = 2.29, p < .01), and this mediator was significant in predicting each of the three strain outcomes. Only the simple slopes for Conditions “1” (high–high, positive slope) and “3” (high autonomy–low conscientiousness, negative slope) were significant at p < .05 and significantly different from each other (t = 2.98, p < .01). Autonomy need satisfaction was a significant mediator of the effect of this three-way interaction on all three strain outcomes (exhaustion: μ = −0.572 (SE = 0.291; 95% confidence interval (CI) [−1.228, −0.103]; disengagement: μ = −0.939 (SE = 0.403; 95% CI [−1.807, −0.229]; dissatisfaction: μ = −0.847 (SE = 0.361; 95% CI [−1.622, −0.208]). These results mirror the results for emotional stability and provide further support for the DCP model.
4. We thank an anonymous reviewer for these suggestions.