When family interrupted work: The implications of gendered role perception in the face of COVID-19.

During the COVID-19 pandemic, many individuals are confronted with the work-from-home challenge, which often results in work-family interference. Although prior to COVID-19, the influence of traditional gender role expectations was shown to be reduced over time, it is unclear whether and how such traditional worldview might influence judgments towards men and women when family interrupted work under the threat of COVID-19. This study presented and tested competing predictions derived from the gender role theory. An experimental study with 971 adults showed that during (vs. before) COVID-19 pandemic, men were evaluated more negatively when they experienced family interruption to work compared with women. The negative evaluation further led to more punitive reactions and less support at work. The results suggested that gender role expectations reinforced the traditional status quo by punishing status-quo-breakers under the threat of COVID-19.

INTRODUCTION

The COVID‐19 pandemic has affected many lives globally. It did not only make the existential threat to humanity salient (Hu et al., 2020) but also brought about drastic changes that disrupted established normative practices. For instance, before the COVID‐19 pandemic, work and family tend to exist in different physical spaces, with a relatively clear physical boundary. However, with the pandemic, many people are forced to work from home due to lockdown and social‐distancing rules. Although remote work is not a new concept, COVID‐19 has brought that to a new level. Whereas remote work before COVID‐19 has focused on the flexibility and autonomy for the employees to choose where and when to work (Evans et al., 2004), COVID‐19 has left individuals with no choice but to work from home, dealing with both work and family simultaneously. With the temporary closure of schools and daycare facilities, working parents are expected to support their children's virtual learning at home (Sevilla & Smith, 2020), while managing their work from home. Family and work routines often interrupted each other, competing for attention.

There have been intense public discourses about how the COVID‐19 pandemic has resulted in a drastic shift in managing work and family responsibilities. In particular, more working parents report that balancing work and family has gotten more difficult than pre‐COVID‐19 with increased household responsibilities (Heggeness & Fields, 2020; Igielnik, 2021). Women, who are traditionally seen as the primary caregivers in a household (Kobrynowicz & Biernat, 1997), have disproportionately borne the extra household burden during the pandemic while also being expected to manage their careers remotely (Donner & Purtill, 2020). With the blurred work and family boundaries, men also faced unique challenges. Before the pandemic, men, compared to women, took on significantly fewer household responsibilities (OECD, 2021). However, the COVID lockdown measures have altered men's roles in the household, leading them to take on more domestic and childcare duties (Uhereczky, 2020; Valero & Tinonin, 2020). Men needed to quickly adapt to a new work environment at home, in which work and family are simultaneously fighting for attention in the same physical and mental space. Content in social media also reflected a similar trend (e.g., #equalinqurantine, #itakemyshare, Father Support Program; Uhereczky, 2020). Accordingly, the blurred work‐family boundary during COVID‐19 has also impacted men as they take on household responsibilities that deviate from traditional expectations.

As men's participation in the caretaking role surged with the pandemic outbreak, it is important to understand how individuals respond to the changing gender role of both women and men at time of a global crisis. With the social impact of traditional gender role expectations (i.e., women as homemakers, men as breadwinners) decreasing gradually and slowly over time before COVID‐19 (Donnelly et al., 2016; Joshi et al., 2015), men's increasing participation in the caretaking role should alleviate the caretaking pressures for women (Persson & Rossin‐Slater, 2019) which could encourage women's participation and progression in the workplace even more. However, with the uncertainty and threat posed by the COVID‐19 pandemic, it raises the question of whether the traditional gender worldview would gain prominence again during COVID‐19. Specifically, given the traditional gender role expectations, how would people react toward men and women when their family interrupts work in the face of COVID‐19? Our findings showed that compared with women, men were evaluated more negatively and were more likely to be punished when they experienced family interruption to work during the pandemic, suggesting that traditional gender role expectations prevailed.

This study makes two key contributions. First, it echoes recent research showing that stereotypes are not static (Haines et al., 2016), but can change over time (Eagly et al., 2020). However, contrary to the existing research that shows that the malleability of stereotypes can bring a positive change towards equality, this study reveals that a dramatic societal event such as the COVID‐19 pandemic might reverse the positive strides. This study demonstrates that despite the progress in gender equality in the past decades, traditional gender stereotypes prevail again in the face of the pandemic, holding both men and women back to their traditional social roles. With the blurring of work and family boundaries, the pandemic presented a situation in which women were expected to behave according to their gender role, whereas men deviated from their gender roles. Understanding people's perception towards gender role and responsibilities during the pandemic is important in addressing gender equality issues because women are often held back from advancing their careers due to their caregiving responsibilities (i.e., Correll et al., 2007; Rivera & Tilcsik, 2016). If men are being punished for taking up a share of caregiving responsibilities, equality cannot be achieved.

Second, the study also contributes to the work‐family literature. So far, the literature has assumed that work and family are independent domains, and their boundary is managed by individual preferences and choices (Kossek & Lautsch, 2012; Kossek et al., 2006). Given that women are perceived to be less agentic and more family‐oriented, and men are seen as more agentic and career‐focused (Eagly & Karau, 2002; Eagly & Wood, 1991), women are expected to experience more work‐family conflict (Hoobler et al., 2009), to forgo career opportunities for personal and family reasons (Rivera & Tilcsik, 2016), and to be less committed to work in general (Correll et al., 2007). However, our study is uniquely situated in a context in which individuals are evaluated based on the actual blurring of work‐family boundaries, not expected blurring of boundaries. The unprecedented work‐family situations resulting from the pandemic enable us to gain insight on the unexplored grounds by revealing that involuntary blurring of the work‐family boundary could also have consequences for employees.

The pandemic and stereotypical judgements

Research in the past decade before the COVID‐19 pandemic has shown that gender perceptions have changed over time due to the increased participation of women in the workforce (Eagly et al., 2020). The gender gap in performance evaluation (Joshi et al., 2015) and perceived competence (Eagly et al., 2020) has started to close, as more egalitarian gender role expectations prevail (Donnelly et al., 2016). In some cases, female executives are seen more positively and are paid more than men (Leslie et al., 2017). Although traditional gender stereotypes still exist in our society (Eagly et al., 2020), biased judgments and stereotypes have reduced before the pandemic. However, bias reduction would only be observed when individuals are motivated and possess the cognitive capacity to do so (Govorun & Payne, 2006). The COVID‐19 pandemic has given rise to much uncertainty and threat, leading to a feeling of exhaustion (Zacher & Rudolph, 2020). It has also presented people with constant reminders of their mortality (Hu et al., 2020). Individuals have an instinctual desire to strive for symbolic immortality (Greenberg et al., 1997) and to retain a sense of predictability in life (Lerner, 1980). When there are uncertainty and threat, people tend to adhere to their cultural worldviews to restore a sense of meaning and structure (Jonas et al., 2008). Therefore, the threat of COVID‐19 would likely lead people to revert to the dominant cultural worldview, such as traditional gender roles.

Moreover, the COVID‐19 pandemic presents an important context to understand the theoretical dynamic of gender role expectations. Work and family lives were relatively separate from each other before the pandemic. Traditional gender norms expect men to prioritize work and women to prioritize family (Ellemers, 2018). However, due to social distancing measures and work‐from‐home policies implemented during the pandemic, people's work and family lives no longer have clear separation. With a murky boundary between work and family, it is unclear whether and how the impact of traditional gender role expectations would unfold. The COVID‐19 pandemic has brought about uncertainty and existential threat, and consequently, people are more likely to adhere to traditional worldviews for psychological safety (Hu et al., 2020; Jonas et al., 2008; Zacher & Rudolph, 2020). However, it remains unclear how adherence to traditional gender stereotypes would influence individuals’ reactions toward male and female employees in the face of work‐family interruption. On the one hand, the existing stereotype towards men being good workers and women being the primary caregiver can provide a buffer for men to shield them from being seen negatively when family interrupted work (i.e., Brescoll et al., 2013). Men are not expected to manage domestic issues in the family and the assumption around men as good workers will help to alleviate concerns about family interruptions. At the same time, it can disadvantage women as family interruption to work might be seen as an indication of their inability to manage work and family, and as unfit for work (i.e., Correll et al., 2007). On the other hand, it is also possible that because of the gender expectation that men should not allow the family to interrupt work, if the interruption did occur, men would fall short of the gender expectation that they are good workers (Vandello et al., 2008). Consequently, they would be seen more negatively. However, for women who are the stereotypical caregivers that should prioritize family over work, they might be exempted from being blamed when family interrupted work (Kobrynowicz & Biernat, 1997). To this point, we propose two countervailing predictions about how stereotypes might influence individual judgments in the face of the pandemic outbreak: (a) Prescriptive stereotype hypothesis, and (b) Proscriptive stereotype hypothesis. These competing predictions were pre‐registered: https://osf.io/d48cv/?view_only=b677b9927eb04cb58cc9093bcf68fa6c.

Prescriptive stereotype perspective

Prescriptive stereotypes refer to beliefs on how men and women should behave (Eagly & Karau, 2002). Traditional gender roles expect men to be more career‐oriented and agentic and women to be more family‐oriented and communal (Prentice & Carranza, 2002). Given that there is a strong preference in the workplace towards employees who prioritize work over non‐work issues, particularly family (Dumas & Sanchez‐Burks, 2015), men are perceived as more “ideal” because they are expected to be able to devote themselves fully to their careers compared to women who are expected to take up family responsibilities (Blair‐Loy, 2003). In situations where work devotions can be under threat, such stereotypic expectations that favor working men over women provides a buffer for men. For example, research has shown that family responsibilities are more likely to be considered a threat to working mothers’ work commitment compared to working fathers (Correll et al., 2007; Hoobler et al., 2009). When COVID‐19 brought about threat and uncertainty, such stereotypic expectations that favor working men over women could represent the salient cultural worldview (Greenberg et al., 1997; Jonas et al., 2008). Hence, the prescriptive stereotype hypothesis suggests that during the pandemic traditional expectations of men being more agentic and competent than women would prevail. Such expectation could alleviate men from negative consequences when family interrupted work during COVID‐19 because positive beliefs about men's capabilities could provide a buffer for men, shielding them from being evaluated negatively. At the same time, the existing beliefs that working women are less competent and less committed might further be reinforced when family interrupted work, leading to negative perceptions toward women at work in the face of COVID‐19.

Proscriptive stereotype perspective

Proscriptive stereotypes refer to the beliefs about undesirable behaviors that individuals should not engage in. Traditional gender roles expect men to focus on work and not to be interfered with family caretaking (Eagly et al., 2020; Ellemers, 2018). For example, men face more negative evaluations for taking family leave compared with women (Butler & Skattebo, 2004). In the face of COVID‐19, violating gender roles can disrupt the sense of meaning and continuity as these normative expectations represent valued cultural worldviews (Greenberg et al., 1997; Jonas et al., 2008). Given the proscriptive stereotypes dictate men should not devote to family caretaking at the expense of work, family interruption to work would defy the proscriptive stereotype for men. On the contrary, the proscriptive stereotypes advocate that, women should not rid themselves of family and care‐taking responsibilities for work. Therefore, family interruptions to work may be considered counter‐stereotypical for men but not for women. Behaviors that are consistent with gender roles are likely to be preferred over inconsistent behaviors, and role violation often results in negative consequences (Rudman & Fairchild, 2004). In fact, individuals would be more forgiving to gender‐role congruent transgressions than gender‐role incongruent transgressions (Wei & Ran, 2019). This means that in the face of work‐family conflict during the pandemic outbreak, individuals might evaluate role‐congruent behaviors positively but role‐incongruent behaviors negatively as a means to reinforce a valued traditional worldview. Therefore, according to the proscriptive stereotype hypothesis, under the threat of the pandemic, men, compared with women, would be evaluated more negatively when their work is interrupted by family responsibility.

In sum, the uncertainty and existential threat resulted from the pandemic (Zacher & Rudolph, 2020) would increase adherence to traditional worldviews for psychological safety (Hu et al., 2020; Jonas et al., 2008). The gender role theory represents two alternative hypotheses about how traditional gender stereotypes might influence individuals’ reactions toward work‐family interruption in the face of threat. When family interrupts work during the pandemic, the prescriptive stereotype hypothesis predicts that, because men are seen as more competent and career‐oriented than women, such traditional stereotypes would shield men, but not women, from being evaluated negatively when family interrupted work. The proscriptive stereotype hypothesis, however, predicts that, because men are not expected to take up caregiving responsibilities, men would be more vulnerable to receiving negative evaluations than women when their work was interrupted by family responsibility. We test our hypotheses by examining how individuals respond to work‐family interruptions during (vs. before) the pandemic in order to understand how traditional gender expectations influence their judgements at a time of immense psychological threat.

METHOD

Participants and procedure

As noted, this study was preregistered. Data, preregistration, and Supplemental materials are available here https://osf.io/gjvqu/?view_only=39608f2dc2934ae085e5528308b5de0f. All data were collected on Prolific (www.prolific.co) and the survey was administered in English. Participants who have completed the study received £1.00. The recruiting criteria included anyone of 18 years or above, and who identified with male and female genders only.

This study involved a 2 (Pandemic: Pre‐Pandemic vs. Pandemic) × 2 (Employee gender: Male vs. Female) design. The Pandemic data collection was unplanned at the time when the Pre‐Pandemic data was collected, so we recruited participants during the pandemic with the same recruiting criteria as pre‐pandemic to ensure comparability between the two samples. All participants were randomly assigned to one of two experimental manipulations in which they read about either a male or a female employee experiencing family interruption to work. The Pre‐Pandemic data were collected in July 2018. WHO declared Coronavirus as a pandemic on March 11th, 2020. By May 2020, many regions in the world had declared lockdown and required employees to work from home (Parker et al., 2020; Salvador et al., 2020). The Pandemic data were collected in May 2020.

The sample size was pre‐determined before the data collection based on budgetary constraints. We collected data from 1014 participants. After excluding 37 participants who did not pass the two manipulation checks (see below for details), or who did not properly recall the manipulation scenarios, 977 participants remained. Of the 977 participants, six participants provided duplicate ID numbers indicating the potential of participating in the study more than once. To ensure the quality of the data, we excluded these duplicated participants. The final analysis consisted of 971 participants (pre‐pandemic N = 482, pandemic N = 489). A post hoc analysis using the g*power software (Faul et al., 2007; f = .10, α = .05) revealed that a sample size of 971 yields 88% power for a test with four groups. For the moderated mediation analyses, we used bootstrapping method to reduce the concern for underpowered results through resampling simulations (Preacher et al., 2007). Therefore, the study was sufficiently powered to detect the expected effects across different analyses. The GLM analyses were pre‐registered as main analyses, whereas the moderated mediation analyses were noted as exploratory analyses. For simplicity, additional analyses in the pre‐registration not reported in the main text are reported in the Supplemental materials.

During the recruitment process, our participants were not preselected based on specific demographic information. Thus, the demographic composition reflected that of a random sample from the online platform. Participants were on average 30.98 years old (SD = 10.89), 47.5% female. There were 83.7% White participants. The remaining participants self‐identified with other racial groups, such as Asian (6.6%) and multi‐racial (3.7%). Most participants were from the U.K. (37.9%), the U.S. (16.1%), and other regions such as Canada (3.0%), European countries such as Poland, Portugal, Italy, and Spain (34.7%), South & Central America (3.9%), and Asian countries such as China, Korea, and Singapore (1.4%). In terms of work characteristics, 72.2% had either full‐ or part‐time employment.

Experimental manipulations

Our manipulation scenario depicted a situation in which a male or female employee experienced family interruption to work (e.g., showed up late for group meetings, missed project deadlines due to emergency with childcare issues). The final product of the employee's project group fell short of expectations. Members of the group expressed that the outcomes were due to how they had to take on extra responsibilities for the employee who failed to do his/her part of the job. We manipulated employee gender using names and pronouns (Male: Mark, he, his, him; Female: Mary, she, hers, her). The scenario used in the pre‐pandemic and pandemic data collection was identical, except that in the pandemic scenario, the scenario was situated in the context of the pandemic. Full scenarios are shared in Supplemental materials.

Measures

Manipulation check

The following measures were used to ensure that the participants understood the manipulation and the scenario correctly. First, we had the participants recall the name and the gender of the target employee. Second, we asked the participants to recall the scenario. Those participants who incorrectly recalled or provided irrelevant information were excluded. As noted above, 37 participants were excluded for failing one or more of the manipulation checks.

Culpability judgment

After reading the scenario, the participants responded to four items that assessed the extent to which the employee should be held responsible (0 = no responsibility at all to 10 = all the responsibility) and should be blamed (0 = no blame at all to 10 = all the blame) for the failure of (a) the group, and (b) the project (Chao et al., 2008). The mean composite score of all four items was used as the measure for culpability. The scale reliability is α = .92.

Punishment

We measured the extent to which the target would be punished by having resources withheld and being put on probation. In terms of resources, the participants indicated how much bonus (in percentage) they would allocate to the target employee if they were the manager of the group who could allocate up to an extra 1‐month pay as a year‐end bonus. Participants also rated the extent to which they would support putting the target on probation on an 11‐point scale (1 = none to 11 = a great deal).

Future support

We also measured the extent to which participants would support the target by providing future work opportunities and showing empathetic understanding. Future work opportunity was measured by asking participants to rate the extent to which they would invite the target to join their next project on an 11‐point scale (1 = none to 11 = a great deal). The empathetic understanding was measured by having participants indicate the likelihood of trying to talk with the target and to understand the issue that the target is struggling with on an 11‐point scale (1 = none to 11 = a great deal).

RESULTS

Culpability judgment

Using a 2 (Pandemic: Pre‐Pandemic vs. Pandemic) × 2 (Employee Gender: Male vs. Female) GLM to examine culpability judgment. There was no significant main effect of Employee Gender, F(1,967) = 2.56, p = .110, η 2 = .003, or Pandemic, F(1,967) = .28, p = .598, η 2 < .001, on culpability. This means that the participants did not attribute culpability differently based on employee gender, nor did they attribute culpability differently Pre‐Pandemic versus during the Pandemic. However, there was a significant two‐way interaction of Pandemic × Employee Gender on culpability, F(1,967) = 4.47, p = .035, η 2 = .01 (Figure 1). Follow‐up analysis showed that male employees (M = 6.30, SD = 1.86) were seen as more culpable during the Pandemic compared to female employees (M = 5.84, SD = 2.05), F(1, 487) = 6.92, p = .009, η 2 = .01, whereas male and female employees’ culpability did not differ Pre‐Pandemic, F(1,480) = .13, p = .716, η 2 < .001. That is, consistent with the proscriptive stereotype hypothesis, men who experienced family interruption to work were judged as more culpable compared with women in Pandemic, but not Pre‐Pandemic.

FIGURE 1

Interaction effect of pandemic and employee gender on culpability judgment

Follow‐up analyses examined the effect of Pre‐Pandemic versus Pandemic among male and female employees separately. Specifically, there was a trend that male employees were judged as more culpable in Pandemic compared to Pre‐Pandemic, F(1,493) = 3.74, p = .054, η 2 = .008, whereas female employees were not judged differently Pre‐Pandemic versus Pandemic, F(1,474) = 1.17, p = .279, η 2 = .002. Hence, the interaction effect of Pandemic × Employee Gender was primarily driven by increased culpability judgment towards men in Pandemic compared to Pre‐Pandemic. The pattern of findings still hold even after controlling for demographic factors (see Table S7 in Supplemental materials).

Punishment

Using PROCESS Model 7 (Hayes, 2015) with 10,000 bootstrapping iterations, we tested the moderated mediation of Pandemic × Employee Gender through culpability judgment on withholding bonus and putting on probation, respectively. The results on bonus allocation revealed that the moderated mediation was significant (b = –2.25, SE = 1.10, 95% CI [–4.50, –.19]). Specifically, with the Pandemic, more bonus was withheld to male employees compared to female employees through increased culpability judgment (b = –1.97, SE = .78, 95% CI [–3.59, –.50]) (Figure 2), whereas the indirect effect was not observed Pre‐Pandemic (b = .27, SE = .77, 95% CI [–1.23, 1.78]).

FIGURE 2

Moderated mediation analysis of employee gender on bonus allocation, probation, invitation to next project, and empathetic understanding through culpability in pandemic

Next, we also tested the moderated mediation of Pandemic × Employee Gender on probation. Results showed that the moderated mediation was significant (b = .15, SE = .08, 95% CI [.01, .31]). Specifically, the indirect effect of employee gender on probation through culpability was significant in Pandemic (b = .13, SE = .06, 95% CI [.03, .25]) (Figure 2) whereas it was not significant Pre‐Pandemic (b = –.02, SE = .05, 95% CI [–.12, .08]). This means that during the Pandemic, male employees who experienced family interruption to work were more likely to be put on probation compared to female employees because they were perceived as more culpable, but not Pre‐Pandemic.

Future support

We also tested the moderated mediation of Pandemic × Employee Gender on future support through culpability judgment. The moderated mediation effect was significant for the likelihood of asking to join their next project (b = –.25, SE = .12, 95% CI [–.48, –.02]). The patterns showed that the likelihood of inviting male and female employees to their next project through culpability judgment was not significant Pre‐Pandemic (b = .03, SE = .08, 95% CI [–.14, .20]). However, the indirect effect was significant with the Pandemic (Figure 2). Specifically, the male employees were less likely to be invited to join the next project compared to female employees through increased culpability (b = –.22, SE = .08, 95% CI [–.39, –.06]).

The moderated mediation was also significant for empathetic understanding (b = –.23, SE = .11, 95% CI [–.45, –.01]). Again, empathetic understanding towards male and female employees through culpability did not differ Pre‐Pandemic (b = .03, SE = .08, 95% CI [–.12, .18]). However, with the Pandemic, people demonstrated less empathetic understanding of male employees than female employees through increased perceived culpability judgment towards male employees (b = –.20, SE = .08, 95% CI [–.36, –.05]) (Figure 2).

Subgroup analyses

Although we did not plan to examine subgroup differences in the pre‐registered analyses plan, given the widespread impact of the pandemic on different regions and people with different backgrounds, we conduct additional subgroup analyses to explore whether the effects fared differently across regions and demographic backgrounds. The results showed that the Pandemic × Employee Gender effect has remained robust even after controlling for different demographic variables, such as the participants’ nationality, socioeconomic status, number of children, and gender (Table S7). Furthermore, none of these demographic variables interacted with the Pandemic × Employee Gender factor to influence culpability judgment. Although the analyses with demographic variables did not show higher order interactions with Pandemic and Employee Gender, they have yielded interesting insights for future study. For instance, male participants were more punitive compared to female participants in general (Table S11; Figure S5), and participants from the U.S. were more punitive compared to those from the U.K. or other countries (Table S8; Figure S1). These results are consistent with the prior literature that show men tend to be less agreeable than women (Feingold, 1994), and there is a strong norm in the U.S. to be professional and not have family interrupt work, stemming from protestant work ethics (Sanchez‐Burks, 2002). Furthermore, the socioeconomic status (SES) of the respondents influenced how they perceived male and female targets differently. Using individuals’ income as a proxy for socioeconomic status (SES), the analyses showed a marginally significant interaction effect of target gender and SES (p = .071) (Table S9). Specifically, whereas those with low SES do not perceive male and female targets differently when family interrupted work, participants from middle and high SES groups rated male targets to be more culpable compared to female targets (Figure S2). Moreover, there was also a marginally significant interaction effect of the Pandemic and SES (p = .092) (Table S9). Specifically, during (vs. before) the pandemic, those with high SES perceived employees to be particularly culpable when family interrupted work (Figure S3). Taken together, these effects suggest that the proscriptive stereotype hypothesis is especially stronger for those with more resources (higher SES), perceiving men (vs. women) as more culpable when family interrupted work. Moreover, they also showed the most drastic shift in their culpability judgment, becoming particularly intolerant toward employees experiencing family interruptions to work during (vs. before) the pandemic.

In addition, we also found that having children influence people's perception towards male and female targets (Table S10). The interaction patterns show that those who have at least one child perceived male targets as more culpable compared to female targets, whereas those without children did not evaluate male and female targets differently (Figure S4). This finding seems counterintuitive as one would expect participants with children would be more understanding of the challenges face by both parents regardless of gender. However, it is possible that the results reflect the social reality that for families with children, women are often the de facto caregivers. Hence, women experiencing family interruption to work is the norm, rather than an exception in families with children. Thus, those with children are more understanding of the female employee when family interruptions occur, but not for the male employee.

It is important to note that these analyses are exploratory that aim to provide some insights into how people at the crossroads of different demographic groups (i.e., gender and SES; gender and parental status) may influence and be influenced by gendered expectations. These exploratory findings can also point to potential future directions that might help understand the intersectionality of different demographic attributes. However, we need to be mindful that these analyses were not planned and might not be sufficiently powered to detect the effects. The findings of these exploratory analyses should be interpreted with caution. The results of these exploratory analyses are available in the Supplemental materials.

DISCUSSION

There has been much public discourse about how the COVID‐19 pandemic has changed work and family boundaries and how that would have impacted men and women. Before the COVID‐19, research has shown that the impact of traditional gender roles on gender biases has decreased over time (Eagly et al., 2020). The pandemic outbreak introduces much uncertainty and threat. Under such situations, individuals embrace valued cultural worldviews (Arndt et al., 1997). Accordingly, we have predicted that people would adhere more strongly to the traditional gender roles in the face of COVID‐19.

The prescriptive stereotype hypothesis suggests that when family interrupts work during the pandemic, men would be judged more positively compared with women as men are seen as more competent and committed to work, whereas the proscriptive stereotype hypothesis argues that when family interrupts work, men would be judged more negatively compared with women for violating the traditional gender expectations that men should not prioritize family responsibilities over work. Results from this study support the proscriptive stereotype hypothesis. Specifically, during the pandemic outbreak, men are evaluated more negatively than women when they experience family interruption to work, leading to more punishment and less support at work. These findings suggest that in the face of COVID‐19, individuals are more punitive toward those who violated traditional gender role expectation, which is a valued cultural worldview.

The results have important implications. First, although the reduction of negative gender stereotypes and women's advancements in society in the previous decades is worth celebrating, this study reminds us to be more mindful about being complacent with improvements made towards gender equality. Our findings show that traditional gender expectations and stereotypes can resurface in the face of a crisis, yielding negative consequences, which might eventually hold both men and women back to traditional stereotypical gender roles. The literature on gender equity has largely focused on studying how females are being disadvantaged by traditional gender expectations, especially in the work domain (Blau & Kahn, 2006; Heilman, 2012). Policies such as on‐site childcare facilities and flexible work schedules have been implemented with the goal to address women's needs in handling caretaking responsibilities (Ezra & Deckman, 1996). However, current study suggests that a more comprehensive understanding of equality issues can only be achieved by examining how traditional gender roles influence both men and women, and by considering whether and when the impacts of gender stereotypes might wax and wane over time. Accordingly, the findings point to the need to provide holistic solutions and supports for both men and women, not just either one or the other, in order to promote gender equality.

Second, the study explores the consequences when work and family boundaries are blurred involuntarily. Research in the work‐family interface has often examined work‐family boundary management as voluntary choices and focused primarily on antecedents and consequences of boundary management (Chen et al., 2009; Kossek et al., 2012). Findings from the extant research enrich our understanding of how men and women experience and decide to employ certain work‐family boundaries. For example, women utilize family‐friendly policies more compared with men (Reid, 2015) and working women take on more housework than men (OECD, 2021). Our study contributes to this discourse by exploring how men and women are perceived under involuntary family‐work interruptions. Our results demonstrate that during the pandemic when work and family boundaries are blurred for both men and women, people are more likely to hold men accountable when family interrupted work, but less so for women. This finding has important implications for understanding gender equality issues, particularly in the face of the changing work‐family dynamics shaped by the pandemics.

Recent reports show that the pandemic has widened gender inequality worldwide and presented challenges to women (Azcona et al., 2020). For instance, women face 1.8 times higher job loss rate compared to men globally, partly because women tend to occupy more temporary job types compared to men (Madgavkar et al., 2020). In academia, a report shows that female researchers’ publication outputs are impacted more negatively by the pandemic compared to that of male researchers (Inno et al., 2020). Our results, at a first glance, appear to contradict the argument that women are more disadvantaged than men during the pandemic. However, the results are largely consistent with the broader discourse on gender discrimination about how gender role expectations imposed on both men and women would propagate inequality by punishing those who violate such expectations and keeping them in their places.

Our findings call for actions beyond instituting policies that simply allow female employees to take family leaves or reduced work hours. It requires practices that not only support women but also provide support to men who might deviate from traditional gendered role expectations. This would require decision‐makers in government and organizations to recognize that gender equality issues are rooted in gender role perceptions that affect both men and women. In fact there are evidence suggesting that men are penalized for taking parental leaves (Cox, 2021; Lee, 2020). As a way to combat such an issue, the Swedish government has made 90 days, out of 240 days of total parental leave, to be mandatory leave for both male and female caretaker after a child is born (European Commission, n.d.). This guarantees at least 90 days of leaves and minimizes risks of negative judgements due to proscriptive stereotypes. Moreover, such a policy might increase awareness of and acceptance towards men who share caretaking responsibilities.

Limitations, implications, and future research

Our paper is not without limitations. First, our study was conducted during the earlier stages of the pandemic. Hence, our findings cannot speak to how and whether the gender role expectations would unfold over time with changing work‐family dynamics as the pandemics unfold. This calls for future studies to examine the impact of societal changes on gender role expectations, and their implications to gender equality. For example, as COVID‐19 has begun to establish a new normal in the interface between work and family, it would be interesting to capture how the effect has changed over time. As the world is adjusting to the new normal, the existential threat presumably has gradually reduced. According to our theory, with a reduced sense of threat, people's tendencies to hold onto traditional stereotypes would also decline. It would be meaningful to further examine how the impacts of gender role expectations evolve.

Second, although our supplemental analyses found that the effect predicted by the proscriptive hypothesis have remained robust among participants of different nationality (Table S8), socioeconomic status (Table S9), number of children (Table S10), and gender (Table S11), it is important to note that the study was administered in English. Respondents from the different regions who were able to have access to and complete the study might be more similar to one another compared with those who do not access the English language platform or are not proficient in English within the same regions. Arguably, the perceptions towards men and women might be more similar among the participants in our study compared with those who did not participate in our study from these various regions due to language proficiency issue, which is often confounded with factors like education, political orientation, and social identities. Hence, although our data consistent of respondents from different regions, the data is limited in examining potential variations between and within regions. Future studies could look into regional differences and the potential impacts of different governmental mandates on gender perceptions in the face of family interruptions to work.

Thirdly, it would be fruitful to explore how different social categories and their intersections may influence the effect of gender role expectations and work‐family interruption. As revealed in subgroup analyses, demographic factors can influence evaluations towards male and female targets when family interrupted work. For example, given that SES of the respondents is linked to diverging attitudes toward the male and female targets (Table S9; Figure S2), future studies can explore how the SES of the targets, in combination with other social categories such as gender, influence the evaluations that they receive. For example, previous research has argued that middle‐ and upper‐class men are able to fully dedicate themselves to their careers because they are able to fully delegate caretaking roles to women at home (Blair‐Loy, 2003; Williams, 2001). Moreover, with increased women's participation in the labor force, working women with more financial resources could alleviate the pressures from housework and childcare by hiring domestic help (Baxter et al., 2009; Cheung & Lui, 2017), whereas women who cannot afford external help continue to live under the pressures of managing work and family simultaneously, or quit working altogether (Gershuny et al., 2005). Therefore, employees from high SES are expected to have and use the resources to hire external help to take care of their family responsibilities regardless of their gender. When their family interrupts work, it might raise questions about their commitment and dedication to work, as they should have delegated family matters to others. Moreover, the gender of the employee may also intersect with the target's SES to influence their evaluations. As men are expected to be career‐oriented and women are expected to be family‐oriented (Prentice & Carranza, 2002), when family interrupt work, men and women among the same SES could be perceived differently depending on the expectations towards their work and family orientations. The current study only manipulated the gender of the target employees to understand the general impact of gender stereotypes, an important extension is to manipulate other demographic characteristics, such as SES, to understand the intersectionality of different attributes.

In addition, race is another important factor to explore in conjunction with socioeconomic status and gender. For example, in the U.S., the median income for Blacks and Hispanics are lower than Whites and Asians (U.S. Bureau of Labor Statistics, 2021), which means that social class is also strongly intertwined with race, and the intersection of gender and socioeconomic status can be further explicated based on people's racial categories. Gender stereotypes also vary according to one's race. For example, Asian men are perceived as less masculine in the U.S. (Lu & Wong, 2013; Shek, 2007). As masculinity is closely related to being career‐focused, Asian men might need to overcompensate at work to show more commitment and dedication. This will leave their spouses with a heavier burden of household duties, deepening the division of work and family roles between men and women among Asian families. Therefore, it is important that future studies empirically test how members at the intersection of different social group memberships experience the work‐family interface.

Gender equality has advanced across the globe prior to the pandemic. It started with women obtaining legal rights in areas that were once granted to men only. For example, in the U.S., a mother was ruled to have equal rights over land as the father of their deceased son in a state that specified that “males must be preferred to females” in 1971 (Reed vs. Reed, 1971), and in 1972, caregiver tax deduction was granted for a single man (Mortiz vs. Commissioner of Internal Revenue). In New Zealand, daughters were ensured equal inheritance rights as sons in the 1970s (The Administration Act 1976). In some regions, such as Iceland and Bangladesh, women have been appointed to legislative and managerial positions for as long as men have been in such positions (Charlton, 2021). In Spain, more than half of the appointed government officials were female in 2018 (Valdivia, 2019). Despite the progress in the past decades, in the face of a global pandemic that led to much uncertainty and threat, our findings revealed that traditional gender role expectations are gaining ground again, punishing individuals who violate traditional gendered expectations.

The U.S. Supreme Court Justice Ruth Bader Ginsburg once said, “Women will have achieved true equality when men share with them the responsibility of bringing up the next generation” (Bote, 2020). If our society punishes men for sharing the responsibility of bringing up the next generation, equality cannot be achieved. As long as individuals adhere to the gender role expectations that men be career‐oriented breadwinners and women be the de facto homemakers, and punish those who violate such expectations, it will inevitably lead women to occupy more flexible jobs, which are usually less prestigious with lower pay and lower job security (Blau & Kahn, 2006), propagating gender inequality.

CONFLICT OF INTEREST

The authors declare no competing interests.

Supplemental Materials

Click here for additional data file.