Gender differences in the association between nonstandard work schedules and work-family conflict: A mixed methods analysis in France

Objective: This study aims to examine the association between nonstandard work schedules and time-based work – family conflict (WFC) among employed parents. Taking a gender perspective, it further considers whether job and family resources mediates this association. Background: The recent increase in nonstandard work schedules has proportionally affected women more than men in various EU countries. Previous research has established a link between nonstandard work schedules and WFC. However, limited attention has been given to directly investigating time-based WFC and exploring the factors that drive gender-specific effects. Method: Employing a mixed methods design, we use cross-sectional data from a large-scale French Working Conditions survey and qualitative interviews with couples who work nonstandard schedules. Results: Quantitative findings demonstrate that nonstandard work schedules increase time-based work – family conflict for women more than men. Early morning and evening work disrupts socially valuable time for women, while weekend work negatively affects both genders. Lack of family support explains a significant portion of the association, with work schedule unpredictability being crucial for women. The qualitative findings shed light on the gender-specific generation of perceptions regarding time-based WFC among couples and the interaction of job and family resources in their daily lives. Conclusion: The findings suggest that a comprehensive understanding of the gendered interferences between family and work dynamics is vital for informing policy decisions, reducing gender inequalities, and enhancing workers’ wellbeing .


Introduction
In the context of a 24/7 global economy, changes in paid work, including the proliferation of nonstandard work schedules, irregular shift times, and on-call work, have dramatically transformed the ways in which work and family interact (Clawson & Gerstel, 2014;Kalleberg, 2018;Presser, 2003).According to the EU Labour Force Survey, a substantial portion of employees in Europe (37%) and France (35%) regularly work nonstandard schedules, including early mornings, evenings, nights, and weekends.Notably, women are disproportionately affected in the US and several European countries, due to their employment in the healthcare and retail sectors and in low-skilled service jobs that require nonstandard work schedule patterns (Gracia et al., 2021;Henly & Lambert, 2014;Lambert & Langlois, 2022).
However, to our knowledge, no study has combined quantitative and qualitative approaches to investigate the association between nonstandard schedules and time-based WFC, despite its importance in addressing time inequality within a global 24/7 economy.Moreover, the gendered mechanisms underlying time-based WFC remain poorly understood, in spite of increasing gender differences in exposure to nonstandard work schedules.Finally, research on work-family conflict rooted in family research and with an explicit focus on family issues has been comparatively rare compared to a more work-centered view (Schulz & Reimann, 2022).
The present study addresses this research gap by: (1) examining gender differences in the association between various nonstandard work hours and days and increased levels of time-based WFC among employed parents who live with a partner; (2) testing whether this association is mediated through lack of contextual resources, in both the work and family domains.This study employs a mixed methods approach to achieve its objectives.The quantitative analysis assesses the relationship between time-based WFC and nonstandard work schedules (NSWS) at a population level and allows us to examine the mediating role of contextual resources.It offers broad coverage with a large sample size (encompassing all industry sectors).We supplement our analysis with a qualitative approach that delves into the experiences of employed parents with underage children living at home, who live with a partner and who regularly work nonstandard schedules.The qualitative work provides a comprehensive and multidimensional perspective of the gendered pathways shaping perceptions and experiences of time-based WFC in daily life, and the role played by different resources and family members.By employing a mixed methods analysis, we thus leverage the strengths of each approach while compensating for their respective limitations.

Nonstandard work schedules and WFC
The concept of WFC refers to "a form of inter-role conflict in which role pressures from the work and family domains are mutually incompatible in some respect, resulting in time, stress-, or behavior-based conflicts" (Greenhaus & Beutell, 1985).This study specifically focuses on time-based WFC, in contrast to existing research that often examines role-and strain-based conflicts (Lapierre & Allen 2006;Steiber, 2009).Individuals working nonstandard schedules face unique challenges, as they are often at work when their family members and relatives are not, resulting in limited time available for family-related and non-work activities.Moreover, according to social time theories (Gerstel & Clawson, 2018;Presser, 2003), employees with nonstandard schedules face the greatest impact from the desynchronization between working time and other social times, including leisure, school, and childcare.This study focuses on work-to-family conflict, which is more commonly reported than family-to-work conflict (for a comprehensive literature review, see Reimann et al., 2022).
Previous research consistently demonstrates a positive association between nonstandard schedules and WFC across nearly all European countries (Taiji & Mills, 2020;Tammelin et al., 2017).However, the strength of this association is influenced by national contexts, including variations in labor regulations, family policies, and gender roles and ideology.Notably, the only study to directly examine time-based WFC so far was conducted by Tammelin et al. (2017), who compared three countries and found that nonstandard work schedules were associated with increased time-based WFC among Finnish and British parents but not among those in the Netherlands.The authors attributed this difference to the presence of childcare facilities and public policies that are better suited to nonstandard working hours in the Netherlands compared to the other two countries.However, results remain inconsistent across Europe.In a comparative study of 32 European countries, Taiji & Mills (2020) found that while nonstandard work hours and days are associated with increased WFC in general, the strength of this association varied significantly by national context.The authors attributed part of the variation to the extent of collective bargaining coverage within a country.Consequently, further research is necessary to consider the influence of national context, including public policies and the provision of reconciliation policies (such public childcare, parental leave schemes), alongside with cultural norms.
Moreover, previous studies have indicated that different types of nonstandard work schedules, including variations in hours and days worked, may have varying impacts on time-based WFC.However, the findings across studies remain inconsistent due to the use of multiple measures of nonstandard work schedules.For instance, most previous studies have used aggregate measures of nonstandard work schedules, which encompass multiple work schedule characteristics such as rotating shifts, weekend work, and night shifts, making it difficult to identify the specific type of nonstandard schedule that poses the greatest challenge to work-family reconciliation (Tammelin et al., 2017).Cross-countries variations may also be related to classical methodological features (sample size, confounders taken into account, etc.).For example, the existing body of literature has not provided convincing evidence of the impact of weekend work on WFC among OECD countries.Using data from the 2004/05 European Social Survey for 23 European countries, Steiber (2009) found that weekend work strongly affects time-based WFC among dualearner couples, as weekends are typically when family activities take place.Similarly, Laβ & Wooden (2022) reported a consistent link between weekend work and higher levels of time-based WFC among parents in a longitudinal study of a large-scale Australian sample.Moreover, multilevel analyses have yielded interesting findings, suggesting that the effect of weekend work on time-based WFC is only weakly influenced by institutional or cultural effects.However, Davis et al. (2008) found no significant association between weekend work and work-family spillover in a large panel of US adults.
The existing literature provides conflicting evidence on the role of gender in the association between nonstandard work schedules (NSWS) and WFC, and the mechanisms underlying gender-specific effects remain understudied (Craig & Powell, 2011;Steiber, 2009;Taiji & Mills, 2020;Tammelin et al., 2017).In their cross-sectional study of dual earners living in Finland, the Netherlands, and the United Kingdom, Tammelin et al. (2017) found that the impact of NSWS on time-based WFC was significantly stronger for women than for men in the Netherlands and the UK, but not in Finland.In contrast, in their longitudinal study of weekend work among Australian parents, Laβ and Wooden (2022) found no gender differences in the association between weekend work and time-based WFC, after accounting for unobserved personspecific traits.Similarly, Taiji & Mills (2020) examined WFC in general without specifying type in their study of a large panel of European countries.Their findings suggest that gender differences in the negative consequences of nonstandard work schedules on WFC only emerge when examining more extreme cases, such as workers who exclusively or consistently work nonstandard schedules.These results highlight the limitations of previous studies that utilized a binary measure of nonstandard work schedules and neglected the frequency and the types of nonstandard work schedules.Therefore, further research is warranted to examine the effect of gender on time-based WFC while further taking into account the diverse national contexts in which these associations occur.
This present study examines various types of nonstandard work schedules (NSWS) among employed parents living with a partner in France.Building upon previous research, we anticipate that the detrimental impact of NWSW on time-based WFC will be more pronounced for women than for men, with the exception of weekend work, which is expected to have a stronger negative impact on men.These hypotheses are grounded in gender role theories and supported by prior empirical studies that highlight the persistence of gendered divisions in domestic labor (Champagne et al., 2015;Craig & Powell, 2011;Lyonette & Crompton, 2015).Despite evolving attitudes and women's increased workforce participation, women continue to allocate significantly more time to household duties and parenting, even when they themselves work nonstandard schedules among dual-earners (Kitzmann & Ensellem, 2023;Maublanc, 2009).
Furthermore, gender differences have been found in the timing and nature of domestic work.Women tend to perform tasks that are more time constraining and have little variation from day to day, such as preparing meals and bathing young children, which require them to be at home at specific times.In contrast, men's involvement in domestic activities is more occasional and concentrated on weekends.While men have more flexibility, their domestic participation is often seen to comprise semi-leisure activities such as gardening, shopping, and playing with children and other family members (Brousse, 2015).Additionally, previous research suggests that men who work long hours during the week often use the weekend to recover lost time with their children (Craig & Brown, 2014).These findings suggest that the subjective value of time may vary considerably by gender and that different types of nonstandard work schedules may be associated with distinct perceptions of time-based WFC for each partner within couples.

The mediating role of job and family resources
Individuals can utilize contextual resources within their environment to cope with negative issues such as stress and WFC that result from their exposure to nonstandard work schedules.The Conservation of Resources (COR) theory broadly defines these resources as external forms of support that originates outside of self, within various domains of life, including work and family (Hobfoll, 2002;Ten Brummelhuis & Bakker, 2012).Such support can come from employers, supervisors, coworkers, partners, and family members (Allen et al., 2020;French et al., 2018).Furthermore, some researchers argue for asymmetry in the impact of resources, meaning that a lack of support or loss of resources can have a greater impact on individual outcomes than receiving additional support or resource gains (Bakker & Demerouti, 2007;Haines et al., 2020).In other words, a lack of support or unhelpful support may lead to increased levels of WFC among workers with family responsibilities.
Building on the insights of the COR and gender role theories, our study examines whether the association between NSWS and time-based WFC is mediated by the lack of two types of support: workplace support and family support.Additionally, we explore whether these mediating roles vary by gender.Considering prior literature, we anticipate gender differences in the importance of each resource's role, given the enduring influence of gender roles and expectations both in the home and the workplace.
Regarding the work domain, we focus on an essential job resource -time predictability -and consider whether its absence plays a mediating role in the association between NSWS and time-based WFC.Time unpredictability has recently been identified as an important issue among female employees, particularly with the widespread adoption of just-in-time management principles in the service industry (Henly & Lambert, 2014;Lambert et al., 2019;Selvarajan et al., 2013).Work schedule unpredictability refers to employer's practice of assigning schedules with little notice, last-minute schedule changes, and/or requiring employees to work overtime or extra hours.Research has shown that work schedule unpredictability is associated with higher levels of WFC, including both time-based and strain-based conflicts among working parents.Furthermore, previous studies suggest that work schedule unpredictability may have a more detrimental impact on women compared to men, given women's roles as family managers and primary caregivers to children.For instance, Henly and Lambert (2014), in their study of low-skilled retail jobs, found that when employed mothers receive their work schedules with little advance notice and experience frequent last-minute schedule changes, they struggle to arrange childcare, participate in school activities, and plan family meals.Consequently, they tend to report greater WFC than their counterparts with more predictable work schedules.However, their study did not include a comparison with male employees.Similarly, Lefrançois et al. (2017), in their immersion study of a cleaning department in a transportation company, found that low-wage women working nights or evenings and facing frequent short-notice changes in their work schedules have to manage complex family planning that is both time-consuming and exhausting, unlike their counterparts with regular work schedules.The present study builds upon previous research by examining the role of work schedule unpredictability in the association between NSWS and time-based WFC at a national level, while also comparing male and female employees.
In the family domain, our study considers the lack of family support as a potential mediator.Extensive research has explored the association between family and partner support and reduced levels of workfamily conflict and most of studies have pointed a positive association (for a comprehensive literature review, see Reimann et al., 2022).Moreover, using longitudinal data from the "LEEP-B3" study, Reimann & Diewald (2022) found evidence for cross-domain buffering between the work and the family spheres (family resources mitigate the negative mental health effects of work-to-family conflict; work resources those of family-to-work conflict).
Moreover, studies have identified gender differences in the relationship between family support and various dimensions of work-family conflict.For instance, Adams & Golsch (2021) find among employed fathers a significant relationship between family support and stress-based work-to-family conflict only.In contrast, mothers appeared to benefit from family support only in the context of family-to-work conflict, suggesting that men and women experience inter-role conflict differently due to their distinct work and family commitments (Adams & Golsch, 2021;Hammer et al., 1997) and to their varying perceptions of WFC (Blanch et al., 2012;Schulz & Reimann, 2022).Moreover, previous research indicates that perceived family support has a stronger impact than actual support on individuals' experiences of WFC (Adams & Golsch, 2021).These results emphasize the importance of employing mixed methods and comprehensive analysis in order to gain a deeper understanding of the role played by each resource in dealing with timerelated issues.

The present study
Our current study employs a mixed methods approach to investigating gender differences in the relationship between nonstandard work schedules and time-based WFC among employed parents living with a partner.Furthermore, drawing upon the insights of the COR theory, we explore whether the association between nonstandard work schedules and time-based WFC is mediated by a lack of job and family resources.To achieve this, we use quantitative data from the French Working Conditions survey and qualitative data from 19 interviews conducted with couples living with children and in which at least one partner usually works nonstandard schedules.
While our qualitative sample does not derive from the quantitative sample, our qualitative sampling was designed with the intention of gaining a comprehensive understanding of the gendered roles of job and family resources in daily life.Our selection focuses on households that are similar in structure to the quantitative sample (employed parents living with a partner and at least one dependent child).However, we specifically target families with extensive exposure to nonstandard work schedules (e.g.working both atypical hours and days), which are commonly observed in the air transport sector.We conducted separate interviews with both partners to capture individual perspectives.Then we compared their viewpoints in order to identify gender variations within a specific family and work context.
The combination of quantitative and qualitative approaches offers several advantages in addressing our research questions.The quantitative sample provides an overview of gender differences in the association between NSWS, time-based WFC, and contextual resource indicators.On the other hand, the qualitative sample enables us to delve into individuals' real-life experiences of working time constraints and their subjective evaluations, which are crucial predictors of perceived WFC (Lee et al., 2014).Additionally, the qualitative sample allows us to explore partner interactions and enhance our theoretical understanding of family resources.This research design helps us answer key questions concerning who provides resources within the family (considering interactions between partners, children, and non-family members such as teachers, neighbors and significant others), what they provide (including emotional, affective, cognitive, and material resources), and how they provide it (to identify the circumstances under which contextual resources are most effective for recipients and how they vary by gender).Furthermore, survey data and time diaries are typically designed for workers with standard fixed schedules that often fail to capture the complex experiences of employees working nonstandard schedules, which can vary greatly from day to day (Clawson & Gerstel, 2014).By combining qualitative and quantitative approaches, we gain a more comprehensive and nuanced understanding of the experiences of parent employees working nonstandard schedules and of their strategies for managing the associated negative outcomes in daily life.
Our study focuses specifically on France, which presents an interesting national context for analyzing the impact of nonstandard work schedules on time-based WFC, given that France's family support policies differ from those in countries that have been the focus of prior research, such as the US, the UK, and the Netherlands.The long-standing labor laws and working time regulations in France are complemented by its comprehensive system of high-quality childcare facilities (Thévenon et al., 2014).However, recent decades have seen relatively few policy measures enacted for work-family reconciliation, with most initiatives originating at the company level and thereby leading to considerable variations across sectors, firms, and occupations (Brochard & Letablier, 2017).Consequently, French employees working nonstandard schedules may face great challenges in balancing their work and family responsibilities.Furthermore, although France's overall prevalence of nonstandard work schedules among employees aligns with the European average (35% in 2019), the percentage has recently increased for women and women are now proportionally more affected than men (Lambert & Langlois, 2022).Consequently, France offers a unique opportunity to consider gender differences in the negative impacts of nonstandard work schedules.

Data
Our quantitative analysis utilizes data from the French Working Conditions Survey, that has collected since 1978 information from a large and nationally representative sample of economically active individuals in France.Our study draws specifically on the 2016 cross-sectional survey (CT-RPS), which provides detailed data on various dimensions of work schedule characteristics, including different types of nonstandard hours (early mornings, late evenings, nights) and nonstandard days (Saturdays, Sundays).Additionally, the 2016 survey dedicates a section to addressing the psychosocial risks faced by workers, examining aspects such as the work environment and perceived support both at work and in the family sphere.The 2016 survey offers a unique opportunity for enhancing our knowledge of the gendered mechanisms underlying time-based WFC among employees who work nonstandard schedules at a national population level, while controlling for cofounding factors.
The survey was conducted face-to-face.For the purposes of this article, our analysis focuses specifically on individuals who are employed, reside in metropolitan France (excluding overseas territories), are aged 15-64, live with a partner, and have at least one under-aged child.We restrict our sample to employed parents with a partner, to ensure that our analysis captures the specific dynamics of job and family resources.Moreover, we exclude singles and lone parents with children to compare similar family structures.Our final total sample size was 10,844 respondents.

Variables
Outcome variable: Time-based WFC is measured with the question, "In general, does your work schedule fit in with your social and family commitments outside work?"We assign a binary indicator of time-based WFC that is equal to 1 when the respondents answer either "Not very well" or "Not well at all." Predictors: Our primary explanatory variable is a dichotomous indicator of working nonstandard schedules.Respondents were asked whether they occasionally or usually work: a) evenings (8 p.m.midnight), b) nights (midnight-5 a.m.), c) early mornings (5 a.m.-7 a.m.), d) Saturdays, and e) Sundays.This allows us to construct a precise measure of NSWS that accounts for nonstandard hours as well as days.For the baseline assessment, we classify employees as having nonstandard schedules if they usually work during any of these time periods: early mornings, evenings, nights, Saturdays, or Sundays.Further, we examine the impact of different types of NSWS on time-based WFC using four separate measures: nonstandard hours, weekends, weekends and nonstandard hours, and night shifts.
Mediator: Exploiting the survey design and following the COR theory, we construct two indicators to capture the lack of family resources (LOFR) and lack of job resources (LOJR).LOFR is measured by asking respondents, "Do your relatives complain that your work schedule makes you unavailable to them?" LOFR is coded as one if workers respond "Always" or "Often".Similarly, LOJR is measured by asking workers, "Do you know your work schedule…?"with potential responses being, "1) a month in advance, 2) a week in advance," 3) a day in advance, or 4) I don't know my schedule."LOJR is coded as one if workers select either responses 2, 3, or 4.
Control variables: Our multivariate regression models control for various individual and work characteristics that are known in the literature to influence WFC.Our variables are employee age, number of children, age of the youngest child, education level, contract type (permanent or fixed term), total working hours per week, employment sector (public or private), and spouse's employment status (employed, unemployed, or economically inactive).
Table 1 presents the descriptive statistics for NSWS, time-based WFC, and the explanatory variables.Roughly 36% of partnered employees in France work nonstandard schedules.While 12% of standard schedules workers report time-based WFC, this figure is much higher (32%) for their counterparts working nonstandard schedules.The highest levels of time-based WFC are reported by employees working both weekends and nonstandard hours, as well as those working night shifts.LOFR and LOJR are twice as high among employees working nonstandard schedules, regardless of the definition used.Notably, women are more frequently found among weekend workers, and employees working nonstandard schedules tend to have lower education levels.

Empirical strategy
To examine the association between NSWS and time-based WFC, we estimate a series of multivariate linear probability models.Initially, we run ordinary least squares (OLS) regressions, where we regress the binary indicator of time-based WFC against our measure of NSWS and a set of socio-demographic and employment controls.We run two separate models for male and female employees to allow the NSWS coefficients and control variables to differ by gender.Additionally, the Appendix includes a combined model in which we run the regressions with both genders together to identify any statistically significant gender differences.
In the next step, we explore the potential mediating effects of two variables: the lack of job resources (LOJR) and the lack of family resources (LOFR).To assess whether the association between NSWS and time-based WFC is mediated by LOJR and LOFR, we introduce these two variables separately into our estimation model.We then examine the extent to which the coefficient associated with time-based WFC changes.A substantial change in the coefficient implies that a portion of the association between NSWS and time-based WFC is attributable to either LOJR or LOFR.

Qualitative study
Our qualitative study makes use of open qualitative interviews to capture the experiences of couples with children, where at least one partner usually works nonstandard schedules.The objective here is to gain insights into their daily lives from both partners' perspectives and enhance our understanding of the gender-related mechanisms influencing time-based WFC.The interviews were designed as couple interviews in which all members responded individually.Consequently, semi-structured interviews were conducted separately with both partners.
Initially, we recruited 10 individuals (half men, half women) who were employed by the same French airline company.This allowed us to compare employees facing identical working conditions and identify gender differences.We chose the air transport industry due to it having the highest prevalence of NSWS among employees in France, including nonstandard days and hours.Our selection of the qualitative sample aligns with our quantitative analysis, which demonstrates that overlapping nonstandard hours and days have the strongest association with time-based WFC for both genders (24.2% for women, and 22.7% for men).
Next, to allow for independent narratives in the interviews and facilitate the expression of critical perspectives on partnership and family life, we conducted separate interviews with nine partners of the above-mentioned airline employees.The reason for not conducting 10 interviews is because one partner declined to participate.The final sample consisted of five stewards and air hostesses, and five pilots, all of whom were partnered with flight attendants, a teacher, a doctor, an osteopath, a university professor, and managers, representing the upper-and middle-classes.Additionally, all participants were in the mid-life stage, aged between 32 and 50, were parents of at least one dependent child, and residing in metropolitan cities in France.
The data were collected either in participants' homes or online due to pandemic restrictions.The interviews consisted of open-ended narrative-generating questions covering career, work schedules, family life organization, and coping strategies.The interviews were transcribed, anonymized, and coded using MaxQda software.
The analysis follows an iterative-cyclical¬ approach of the Grounded-Theory Methodology (Charmaz, 2011;Glaser, 1992), in which theoretical sensitivity guided our coding process to help identify concepts, similarities, and conceptual reoccurrences in data collected from partners who work nonstandard schedules.An iterative process was employed to compare the initial categories with the relevant literature and our quantitative study results, leading to further adjustments of categories and concepts.

Quantitative findings: Nonstandard work schedules and time-based WFC
We first regressed our baseline indicator of NSWS on time-based WFC while controlling for sociodemographic and employment factors.Model 1 in Table 2 reveals that employees working nonstandard schedules are more likely to report time-based WFC than employees with standard schedules.Notably, women exhibit a higher coefficient (20.6%) than men (15.5%).To confirm the statistical significance of this gender difference, we ran the same regression with both genders combined (see Appendix, Table A2).
Next, we examined the relationship between different types of NSWS and time-based WFC using four alternative NSWS definitions: nonstandard hours (column 2), nonstandard days (e.g., weekend work, column 3), nonstandard hours and days (column 4), and night shifts (column 5).The coefficients in Table 2 indicate that overlapping nonstandard hours and days have the strongest association with time-based WFC for both genders (24.2% for women and 22.7% for men), followed by night shifts and weekend work.Interestingly, while night shifts are equally detrimental to men and women, weekend work has a more negative impact on time-based WFC for women.Finally, nonstandard working hours (e.g., early mornings and late evenings) are associated with increased time-based WFC for women only, with no significant association found for men.
In a robustness test, we examined the impact of binarizing the outcome variable.The results of a multinomial logit analysis using four possible intensity categories of the time-based WFC variable are presented in Table A1 of the Appendix, which shows that NSWS leads to a tenfold increase for women (sixfold for men) in the probability of reporting that "work schedules did not fit at all with social and family commitments outside work."For simplicity, we have retained the binary results as our baseline.Next, we assess the two mediating variables.Table 3 presents descriptive statistics on time-based WFC, LOJR, and LOFR by gender.Roughly 33% of women and 29% of men report time-based WFC.NSWS exacerbates the gender gap in the probability of reporting time-based WFC, with a 17 percentage point difference for men and a 21 percentage point difference for women.To test the significance of this gender disparity in the gap between non-standard and standard work schedules, we employ a simple regression with an interaction term.Furthermore, employees working nonstandard schedules report significantly more LOJR and LOFR than those on standard schedules, with a more pronounced difference for women.Notably, within the nonstandard schedule group, men report higher LOJR, while women report higher LOFR.These statistics indicate gender-specific mechanisms are at play in the influence of nonstandard work schedules.
Table 4 examines the influence of LOJR and LOFR on the association between NSWS and time-based WFC.We run our linear probability model and progressively introduce the two variables.Models 2 and 5 demonstrate that LOJR is associated with increased time-based WFC for both genders.Including the LOJR variable in the baseline model decreases the NSWS coefficient, with a higher decrease for women (9.2% reduction, from 0.206 to 0.187) than for men (5.1% reduction, from 0.155 to 0.147).This result suggests that LOJR is more likely to explain the impact of NSWS on time-based WFC for women than for men.Models 3 and 6 reveal a strong association between LOFR and time-based WFC, as evidenced by the high and significant LOFR coefficient, as well as by the adjusted R2 doubling from columns 1 to 3 and from columns 4 to 6.While remaining significant, the magnitude of the NSWS coefficient diminishes.For women, the NSWS coefficient decreases by 23% (from 0.206 to 0.157), suggesting that 23% of the effect of NSWS on time-based WFC is caused by LOFR.For men, the NSWS coefficient decreases by 27% (from 0.155 to 0.113).This difference is statistically significant (see column 3 of Table A2).These results indicate that LOFR has a greater influence on time-based WFC for men than for women.Finally, to ensure that binarizing the two mediating variables does not influence our findings, we conduct additional tests using categorical variables for LOJR and LOFR.Table A4 replicates the results obtained, showing that the three categories of LOJR and the three categories of LOFR are all positively associated with increased time-based WFC (Appendix, Table A4, columns 3 and 6).

Qualitative results
Through interview analysis, we identified four influential categories that shaped how partnered employees perceived and experienced nonstandard work schedules.These categories, namely "others-related vs. selfrelated motives for conflict," "interaction with partner and significant others," "relation with the planning department at work," and "loss spiral," are presented below with supporting quotations.These analytical categories, emerging from the qualitative study, offer potential directions for future research.

Others-related vs. self-related motives for conflict
We begin by examining the subjective perception of NSWS among couples.The qualitative analysis reveals that both men and women value working nonstandard schedules, as they provide higher income, increased travel opportunities, the possibility of being away from home at certain times, and greater work autonomy.However, despite recognizing these benefits, both men and women also acknowledge the negative impact of nonstandard work schedules on time dedicated to family.Interestingly, the perceived impacts of nonstandard work schedules on family time vary by gender in terms of nature and intensity, depending on who and what activities are encompassed within the concept of "family." In our sample, female flight attendants are concerned about the impact of their nonstandard work schedules on their children's well-being, especially when they work overlapping days.Those with older children worry about lacking time for homework supervision and other school-related responsibilities.Consequently, female flight attendants generally prefer either short-haul flights to ensure being at home daily or long-haul weekend flights, sacrificing leisure activities with children in favor of concentrated weekday educational tasks.Their limited availability compels them to prioritize school over leisure.Furthermore, missing family events like birthdays or religious celebrations is a major source of dissatisfaction for these mothers, who typically bear the responsibility of organizing them.

The children don't like it. I think younger children need regularity. Which is something I cannot give them. […]
It's more reassuring for them to have their mother there every evening than have her be away often, even though she's maybe there longer.(Nathalie,49,stewardess,two children,49,pilot) In contrast, male flight attendants report having more diverse motives for time-based conflicts, focusing less on children and more on personal fulfillment.Unlike their female counterparts, they prefer working weekdays and being home on weekends in order to share leisure time with their families, while also avoiding being alone with their children during the week, which sometimes provokes in them feelings of insecurity or boredom, especially with young children.The men in our sample report that their absence from important family events like birthdays or holidays is compensated by the financial and symbolic benefits their work gives the children, such as having fulfilled parents and access to expensive leisure activities (Chatot, 2020).Furthermore, men express more frequent complaints about the impact of nonstandard work schedules on their leisure time with relatives and friends.Men thus prioritize their free time away from work differently than women do.Finally, the case of Olivier also highlights men's regret over depending on their wives' schedules, as it undermines their control over time-a privilege traditionally associated with hegemonic masculinity (Connell, 2015;Gerstel & Clawson, 2018).[laughs].From one month to the next, I know the days she'll be here and won't be here, and I adjust in kind.My life is synchronized to her schedule.I'm telling you this because, as a result, I had to give up playing music.(Olivier,47,schoolteacher,two children,married to Céline,43,stewardess).

Interactions with partners and significant others
Interestingly, women's perception of time-based conflict is connected with direct or indirect intervention of relatives (partners) and significant others (teachers, friends, or neighbors) who frequently remind them of their "failure" as mothers.This goes in line with the quantitative results showing that LOFR is a mediating factor of time-based WFC.Our qualitative findings indicate that female pilots and flight attendants often face criticism due to their nonstandard work schedules.Their partners and relatives rarely voice criticisms directly but instead exploit comments made by children, intensifying women's feelings of guilt and exacerbating their perception of time-based WFC.
My family complains a lot."Are you going to be here for Mother's Day?" "I don't know.""Are you going to be here for Dad's birthday?""I don't know."And there you have it: I don't know.Honestly, I think it will always be difficult for them to understand… My daughter made me feel awful about leaving.My son is resigned to it [...] but when he was younger, he often made me pay even when I came home.(Élisabeth,40,pilot,Didier's wife) The children love it when she [their mother] is here, but they sometimes make her pay for being away.Kids tend to whine, blackmailing a bit and doing the drama thing.[...] I would be lying if I said there were times when it doesn't bother me and when I haven't had it up to here, when I make her pay for it, even indirectly, with comments or complaints, even though I don't intend to.I sometimes lose control.(Didier,50,three children,married to Élisabeth,40,pilot) Men working nonstandard schedules also face criticism from their female partners, although the children's well-being is generally not the issue.Instead, these women tend to reproach their partners for not participating in household tasks, as exemplified by the case of Laure and Victor.Consequently, men find it easier to dismiss these criticisms, since household chores are not typically associated with masculine identity (Connell, 1995).
I was just so tired.I shouted at him, saying "You take care of it.You weren't here, now you are, so you take care of it.I can't handle it anymore."I hated seeing myself in that state and the way Victor shrugged me off, telling me no, he had to get some sleep, he was tired.(Laure,32,osteopath,one child,in a civil partnership with Victor,32,pilot) Before, I had almost zero constraints.And now my whole day is dictated to me [...]She took care of the kid all week.When I come home, she wants me to be present… But I just want some downtime, to sleep.(Victor,32,pilot,one child,in a civil partnership with Laure,32,osteopath) Contrary to their female counterparts, these men describe themselves as "better dads" than the highly qualified fathers with long hours and standard work schedules to whom they usually compare themselves (e.g., managers and higher-level occupations).Despite spending less time with their children, they view this time as higher quality, less interrupted by work obligations and low-gratification tasks (e.g., washing, cleaning, feeding children).They prioritize emotional connection and leisure activities.Positive comments from teachers about their presence during school drop-offs further reinforce these fathers' opinions.Teachers, as seen in the case of Geoffrey, assess male participation in daycare by comparing men with other men, not women: The teacher always tells me I'm the dad she sees the most.[…] When we're at home, we're really there all day.So we're not there to do homework, bathe them and all that.We are there the whole day long to play with the child, and do lots of fun stuff with them.(Geoffrey,34,pilot,one child,in a civil partnership with Aline,34, also a pilot)

Relationships with the work planning department and the spiral of loss
Men tend to accumulate more resources than women in terms of support from their work environment.These resources enable them to effectively cope with the negative impacts of nonstandard work schedules on their own personal lives, and proactively prevent future time-based work-family conflicts, confirming the link between LOJR and time-based WFC.For instance, male airline workers in our study sample frequently accept last-minute schedule changes requested by the company, yet never report any increased levels of WFC.In exchange for their flexibility, the planning department and their managers willingly accommodate their own requests for schedule adjustments, which in turn reduces time-based WFC and prevents any potential future conflicts.As illustrated by the case of Eric below, this relationship allows them to obtain a day off when needed.E: Until now, when I ask for something, I call my sector manager and he knows very well, he knows me, after 20 years, that I've never been absent too much, except for illness and all that.And so he gives it to me without any question.He says, "Yes, don't worry, I'll manage.Take the day off."On the other hand, if I ask for something and the company doesn't give it to me, I'll try a floor above, moving up in the hierarchy.(Éric,49,pilot,three children,49,stewardess) In contrast, female airline workers consistently decline last-minute changes to their work schedules in the name of their family commitments.As demonstrated by the case of Elisabeth, refusing such changes exposes them to the risk of receiving unfavorable schedules in the future, with their time preferences and needs being less prioritized.Élisabeth refers to this as a "win-win situation", and conversely as a "lose-lose" situation, that ultimately increases time-based WFC.
I: I'm not allowed to improvise.If the planning department calls me to say, "Can you help us out for three days?"I can never say, "Yes."It's impossible.[But] if I call one day and say, "I have a problem, could you find me another flight or reschedule me?" well, since I never do favors, they won't do me any favors either.It's a bit of a win-win situation.(Élisabeth,40,pilot,3 children,married to Didier,50,executive

in an international company)
These findings align with the COR theory, which proposes that individuals with greater resources are less prone to resource loss, while those lacking strong resources are more susceptible to an escalating "loss spiral" (Hobfoll & Shirom, 2000).Adding a gender perspective to Hobfoll and Shirom's theory, our study demonstrates how job and family resources interact differently by gender, thus influencing the likelihood of experiencing time-based conflicts based on gender roles.

Main findings
Our quantitative findings indicate that NSWS are associated with increases in time-based WFC for women and, to a lesser extent, for men.However, the detrimental impact varies across different types of nonstandard schedules.Overlapping nonstandard hours and days have the strongest association with timebased WFC for both genders, followed by night shifts.While nonstandard hours (early mornings, late evenings) exclusively affect time-based WFC among women, weekend work harms both genders.One explanation for this finding lies in gender role theory.Women tend to perform regular, daily, and less flexible domestic tasks more often than men, even when they are the ones working nonstandard schedules in dual earner couples (Brousse, 2015;Kitzmann & Ensellem, 2023).Consequently, women who work shifts or nonstandard hours from Monday to Friday experience higher levels of time-based WFC as they juggle their daily responsibilities, whereas men can compensate for their absence during weekends.
Our study's second contribution addresses the underexplored role of job and family resources in the association between NSWS and time-based WFC, thus helping to fill the gap in previous research (see Reimann et al., 2022).The quantitative analysis reveals that the absence of family support and, to a lesser extent, work schedule unpredictability significantly contribute to the relationship between nonstandard work schedules and time-based WFC.Our findings are in line with the recent study by Reimann and Diewald (2022), which focuses on the interdependencies between the work and private spheres in the German context.Using longitudinal data, the authors concluded that the family should be viewed not only as place where work-family conflict arises or where it is transferred to, but that it also provides resources to deal with conflict.
Our analysis further indicates that work schedule unpredictability disproportionately affects women, contradicting previous findings that suggested schedule control buffers the impact of weekend work for both genders (Laβ and Wooden, 2022).Our qualitative study provides additional insights into how the contextual resources at home and in the workplace are influenced by gender and interact.Interviews with dual-earner couples highlight that women's responsibilities at home make it challenging for them to adapt to last-minute changes in their work schedules, which have a knock-on effect on household management and the coordination of family caregivers (e.g., nurses, housekeepers), in line with previous ethnographic studies on gender, parenting, and family organization (Hochschild, 1989;Lareau, 2011).
Furthermore, our study reveals that working mothers experience conflicts primarily driven by the naturalized needs and well-being of their children.Their experience of time-based WFC is amplified by daily interactions with partners and significant others (e.g., teachers) who frequently remind them that their absence from home during the week neglects the children's needs.Whereas their professional-related absences from home are perceived as detrimental to children's well-being, irrespective of the father's presence, men's absence from home are seen as natural, necessary for career advancement, and nonnegotiable, (Lambert, 2018).These findings suggest that although the nonstandard work schedules of both partners exert pressure on the family, the value of time exhibits asymmetry within couples.Working fathers' experience of time-based conflicts encompass family time, leisure activities, and personal pursuits, in an extensive interpretation of the notion of the family, compared to their female partners.

Strength and limitations
Our study offers several key contributions.First, to our knowledge, this study is the first to employ mixed methods to investigate gender differences in the association between different types of nonstandard work schedules and time-based WFC.Our mixed methods approach combines representative quantitative analysis (covering French employed parents) with in-depth qualitative research to enhance the interpretation of our empirical findings and enrich the theoretical understanding of gendered mechanisms in time-based conflicts.While our qualitative analysis focuses on one NSWS-intensive air transport sector, it encompasses both high-and middle-income workers which can inform other sectors with high NSWS prevalence, such as the health sector (doctors and nurses).Moreover, by examining diverse nonstandard working schedules, and by exceptionally conducting interviews with both partners, as most qualitative research on WFC typically adopts a single-sex perspective or compares individuals from different families, this study provides a nuanced view of the association between time-based WFC and NSWS, from a gender perspective.Finally, we provide novel theoretical insights into the relationship between work schedules and WFC by expanding on the Conservation of Resources (COR) theory and integrating it with a gender perspective, thereby revealing how resources operate and interact in daily life for both genders.
Despite its insights, our study has some important limitations.First, due to data constraints, our quantitative analysis could not simultaneously examine the impact of both partners' working schedules on perceived time-based WFC.Previous research suggests that WFC may be mitigated when partners are able to coordinate their work schedules, whether standard or nonstandard.For instance, Steiber (2009) found that the worst combination for dual-earner couples in Europe is when one partner works evenings or nights while the other does not, whereas both partners working such unsocial hours ameliorates the effects.Similarly, a study by Täht and Mills (2016) among dual-earner couples in the Netherlands suggests that combined spousal working schedules predict family outcomes more accurately than individual working schedules.However, evidence from an Australian panel demonstrates that weekend work impacts WFC in dual-earner families, regardless of the other partner's working schedule (Laβ & Wooden, 2022).Thus, future research should adopt a comparative perspective to investigate the association between partners' working schedules and time-based WFC.
Additionally, the French Working Conditions Survey lacks information on whether or not nonstandard work schedules are self-selected by employees (a choice), which is a significant aspect of future research.Previous findings demonstrate that dual-earner parents may deliberately desynchronize their work schedules to allocate more time with children and equally distribute childcare responsibilities, thereby reducing WFC (Begallet al., 2015;Kim, 2021;Täht & Mills, 2012).Other studies suggest that nonstandard schedules can enhance work-family balance when accompanied by flexibility and predictability (Lozano et al., 2016).These findings highlight the importance of considering how work schedules are determined, as self-selection for nonstandard schedules may potentially mitigate any adverse effects on family life.
Likewise, the French Working Conditions Survey offer limited information on family support.Shockley and Allen's study (2015) indicates that instrumental support from family members is stronger than emotional support at predicting WFC among married women.Future research on time-based WFC thus faces the challenge of developing a more precise measure of family resources that distinguishes between various dimensions (e.g., instrumental, affective, emotional, and cognitive) (Reimann et al. 2022).It could also include a comparative perspective with socioeconomically disadvantaged families as disadvantaged workers are more prone to engage in nonstandard work schedules throughout their lives and face involuntary work schedule constraints (Han et al., 2022;Presser & Ward, 2011).

Conclusion
As European societies shift towards a 24/7 economy, the prevalence of shift workers and nonstandard work schedules increases for employed parents.Therefore, it is crucial that researchers enhance their understanding of the implications of nonstandard schedules for employees and their families, using a gender perspective.Moreover, our findings underscore the importance of differentiating between various types of nonstandard schedules, as they can present distinct challenges for each gender due to persisting gender roles in the work and family domains.Our findings indicate that NSWS exacerbate preexisting gender inequalities and offer new directions for future research.
Our study also offers valuable insights into the interplay between work and family dynamics that have practical implications for organizations employing NSWS.Given the greater negative impact of work schedule unpredictability on working mothers, organizations could improve scheduling practices by increasing time predictability and facilitating shift swapping among parents when changes are necessary (Swanberg et al., 2008).Supportive policy measures such as extended childcare hours should be adopted at the national level to help employees manage their family responsibilities in 24/7 global economy.

Table 1 :
Work-family conflict, lack of family resources, and lack of job resources by type of schedule Note:The last row indicates share of the total sample of 10,844

Table 2 :
Linear probability models of NSWS on WFC, by type of NSWS Note: Significance is indicated by *p < 0.1, **p < 0.05, and ***p < 0.01.Standard errors are robust and in parentheses.Each column represents a different OLS regression with work-family conflict as the dependent variable.The NSWS variable is measured differently in each column.Control variables comprise employee age, number of children, age of youngest child, education level (reference: no high school), contract type (reference: fixed-term contract), weekly work hours, employment sector (reference: private sector), and partner's employment status (reference: unemployed).Data comes from the 2016 CT-RPS dataset

Table 3 :
Work-family conflict, lack of family resources, and lack of job resources by schedule type and gender Total observations are 10,844.Data comes from the 2016 CT-RPS dataset.The last column reports the coefficient of regressing the row variable against an interaction term between gender and NSWS: *** indicates significance level of 1% Note:

Table 4 :
Linear probability models of NSWS on WFC, the role of LOJR and LOFR Note: Significance is indicated by *p < 0.1, **p < 0.05, and ***p < 0.01.Standard errors are robust and in parentheses.Each column represents a different OLS regression, with work-family conflict as the dependent variable.The NSWS variable equals 1 for individuals who often work non-standard days and/or hours.Control variables comprise employee age, number of children, age of youngest child, education level, contract type (permanent or fixed-term), weekly worked hours, employment sector (public or private), and partner's employment status (employed, unemployed, or inactive).See TableA1in the Appendix for the coefficients of all controls.Data comes from the 2016 CT-RPS dataset