Working from home, work–family conflict, and the role of gender and gender role attitudes

ABSTRACT

Previous studies have shown that societal norms around gender roles can shape gender-based outcomes of working from home. This paper extends these findings to see how individuals’ gender role attitudes can moderate the relationship between working from home and work–family conflict, but again with varying outcomes for men and women. We use data from around 3150 employees who participated in wave 10 (2017–2018) of the German Family Panel Survey (pairfam). Results suggest that compared to employees with fixed work locations, those who work from home report higher levels of family-to-work conflict, but not higher work-to-family conflict. Positive associations between working from home and both types of work – family conflict are found only for women, not for men. Specifically, the positive association between working from home and family-to-work conflict is mainly present among women with traditional gender role attitudes, while the positive association between working from home and work-to-family conflict is mainly present among women with egalitarian gender role attitudes. No such variation, however, was found for men. This study highlights the importance of taking gender and gender role attitudes into account when examining the consequences of working from home.

KEYWORDS:

• Flexible working
• Working from home
• gender
• gender role attitudes
• work-family conflict

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 According to the wave 10 pairfam anchor codebook, changing work locations covers people with multiple work venues, such as ‘truck drivers, traveling salespersons, or people away on a job for installation/repairs’ (Pairfam Group Wave 10 Anchor Codebook, 2019, p. 71). These occupational sectors are mostly known for their poor working conditions such as long work hours, shift work, and excess work demands, combined with lack of supervisor or colleague support, which typically lead to higher levels of job-related stress, sleep and health problems, and higher levels of work – family conflict (Hege et al., 2019; Peasley et al., 2020; Shin & Jeong, 2020; Wu et al., 2018). Due to the specific characteristics of this group, we decided to exclude this sample from our analyses. Comparing the remaining two categories (i.e. working from home and working in fixed work locations) is consistent with the approach in prior research (Anderson et al., 2015; Golden et al., 2006; Madsen, 2006; Fonner & Roloff, 2010).

2 The coding remained the same when those who were one standard deviation above the mean (i.e. scores 4 or 5) were coded as egalitarian, those who were one standard deviation below the mean (i.e. scores 1 or 2) were coded as traditional, and those who were in between (i.e. score of 3) were coded as moderate GRAs.

3 We tested the convergent validity of this single item by showing the correlations between this single-item measure and the scale with four items. The high correlation (r=0.76) indicates the convergent validity. Moreover, we tested the predictive validity of this single item (compared to the scale with four items) by comparing the coefficient sizes when regressed on our outcome measures separately (i.e. WTFC and FTWC). First, using WTFC and FTWC as the dependent variables separately, the results showed that both full-scale and single-item measures for GRA are both significant predictors of each outcome. Specifically, using WTFC as the outcome measure, the results of the full-scale measure, F(2, 3136) = 4.67, p < .001, R² = 0.022, were almost identical for the single-item measure for GRA, F(2, 3136) = 5.14, p < .001, R² = 0.021. In addition, using FTWC as the outcome measure, the results of the full-scale measure, F(2, 3130) = 15.23, p < .001, R² = 0.035, were almost identical for the single-item measure for GRA, F(2, 3130) = 14.68, p < .001, R² = 0.032.

4 We included age squared and work hours squared in the analyses to test the nonlinear effects, but they were not significant, so we decided to drop them from the control variables.

5 While it is expected that the effect of working from home on work – family conflict varies between men and women, and based on the presence of a preschool child in the household, we constructed a combined variable between gender and preschool child living in the household. We ended up having four groups: women with preschool children (reference category), men with preschool children, women with no preschool children, and men with no preschool children. Next, we estimated additional analyses to test whether the effect of working from home on WTFC and FTWC varies by this combined variable and GRA(s). The results show that the positive effect of working from home on both WTFC and FTWC is significantly weaker among men with no preschool children compared to women with preschool children (b = =−0.235, SE = 0.101, p <.05; b = −0.358, SE = 0.101, p <.001 respectively). In addition, the positive effect of working from home on FTWC is significantly weaker among women with no preschool children compared to women with preschool children (b = −0.225, SE = 0.099, p <.05). Finally, the effect of working from home on WTFC (and FTWC) does not vary jointly by the combined variable and individuals’ GRAs. Full tables showing these results are available in the Appendix (Tables C and D). The marginal effects plot (available upon request) further demonstrates that there is no evidence of any significant effect of working from home on WTFC and FTWC by GRA among women with preschool children, men with preschool children, women with no preschool children, and men with no preschool children.

6 Non-fixed daily working hours includes those who answer ‘company defined, partially varying daily working hours,’ ‘no formal working hours, I choose my own schedule,’ or ‘flexible working hours with a time account and a certain autonomy within this framework regarding daily working hours.’ The reference category is having ‘fixed daily working hours (start and end time).’

7 Prior research shows evidence of some differences in terms of gender equality in paid and unpaid work between East and West Germany before reunification, but these two societies experienced some convergence in terms of gender equality after reunitication, although some differences still prevail (see Rosenfeld et al., 2004 for more details). There is also evidence from prior research that immigrants are less advantaged and have difficulty integrating into the labor market (Beyer, 2016). Considering how all these factors might impact work – family balance, we estimated multi-group analyses to see whether the main effects and moderating effects vary between East and West Germany, as well as between natives and immigrants. The results showed no difference between East and West Germany or between natives and immigrants (results available upon request). In addition, the control variables for region and ethnicity were not significant in predicting WTFC or FTWC, as shown in Tables 2 and 3.

Additional information

Notes on contributors

Deniz Yucel

Deniz Yucel is a Professor of Sociology at the Department of Sociology and Criminal Justice at William Paterson University in New Jersey, USA. Using different analytical techniques, her research focuses on how and why family and work environments, work-family (im)balance affect relationship, job, health-related, and child developmental outcomes across different unions and social contexts. Her work has been published in journals such as Applied Research in Quality of Life, European Sociological Review, Journal of Family Issues, Social Science Research, Society and Mental Health, and Journal of Health and Social Behavior.

Heejung Chung

Heejung Chung is a labour market researcher and Professor of Sociology and Social Policy at the School of Social Policy, Sociology and Social Research, University of Kent, UK. Her work focuses on flexible working, gender equality, and cross-national variations therein. Her work has been published in journals such as the European Sociological Review, Journal of European Social Policy, Human Relations, and Social Policy & Administration.