Organizational and Individual Resources as Buffers of Work–Family Conflict Linkages to Affect: An Application of the Job Demands and Resources Health Impairment Model

Abstract

Work–family conflict is a phenomenon known to affect an individual’s well-being. However, its affective consequences are yet to be explored. In this study, we focus on understanding work–family conflict affective consequences on positive and negative affect. Our approach aims to refine the Job Demands and Resources model (JD-R model) by incorporating affect as a psychological mechanism in the health-impairment process and by exploring family-supportive organizational perceptions and psychological detachment as moderators. The final sample was composed of 195 couples, with men’s mean age around 46 years old (M = 46.85, SD = 0.34) and women’s age around 44 (M = 44.23, SD = 0.37). Men worked an average of 44.46 h per week (SD = 0.83), while women worked an average of 39.79 h per week (SD = 0.65). The majority of couples had full-time jobs (77.9% of men and 73.8% of women), worked fixed schedules (55.4% of men and 73.8% of women), were employed by others (75% of men and 82.8% of women), and worked for small companies (54.6% of men and 40% of women). Concerning education, most of the men (81.3%) and women (71.4%) attended high school or had less than 12 years of education. To test our moderation models, PROCESS version 4.1st macro for SPSS was used. Additional analyses included correlations and paired mean comparisons. Our findings indicate that work–family conflict correlates positively with negative affect and negatively with positive affect. Psychological detachment moderated the effect of work–family conflict on negative affect for women, but did not moderate the relationship with positive affect for men or women. Family-supportive organizational perceptions also did not moderate any of the proposed relationships. This study highlights how the ability to detach and separate family and professional domains is important and supports the health-impairment process of the Job Demands and Resources model through affective experiences.

1. Introduction

Family and work constitute the most significant facets of adult life shaping individual well-being and personal development. In recent decades, as the domains of work and family have been changing, growing challenges in managing the demands of work and family life have arisen (Mauno et al. 2007). The composition of the workforce, the intensification of work demands, and the widespread adoption of advanced communication and technology tools had a significant impact on the overarching context of work (Allen et al. 2015). The accessibility and the flexibility of information technologies, which allow for a constant connection between employees and their work blur the once well-defined boundaries between work and family life (Boswell and Olson-Buchanan 2007; Davis et al. 2016; Kossek et al. 2011a).

Another notable transformation relates to the increased presence of women in the workforce and the escalating economic demands for sustaining family livelihood. Dual-earner families, families where both partners work outside their homes, have become the norm (Andrade 2015). Nevertheless, the role of the mother within the family is still marked by traditional gender views (Matias 2019), with women performing the majority of the family chores (Perista et al. 2016). Consequently, achieving a balance between work and family demands has become a challenge (Byron 2005; Kossek et al. 2011a), especially for women, who are the ones expected to make the most effort to reconcile work and family (Matias and Fontaine 2012, 2015).

In the current literature, the work–family conflict (WFC) is often examined as an outcome stemming from workplace demands impacting the family domain. However, acknowledging the evolving landscape of the current work reality, the present study posits WFC as an undeniable demand in today’s work context. As defined by Greenhaus and Beutell (1985), WFC denotes the overlap of demands arising from both the work and family domains, resulting from simultaneous pressures in fulfilling their respective responsibilities (Andrade 2015; Allen and French 2023; Bakker et al. 2008; Byron 2005; Boswell and Olson-Buchanan 2007; Carlson et al. 2000; Golden 2021; Greenhaus and Beutell 1985; Kossek et al. 2011a).

In this study, we focus on dual-earner individuals and leverage the Job Demands and Resources model (Demerouti et al. 2001) to shed light on the health impairment process proposed by the model. Our analysis will delve into how work demands, faced by men and women, specifically work–family conflict, impact their well-being. This investigation will consider positive and negative affect and emphasize the role of individual factors—psychological detachment, organizational resources, and family-supportive organizational perceptions. By unraveling gender-specific pathways in the perception of demands and resources, we aim to deepen the applicability of the JD-R model and thus contribute to the existing body of literature.

2. Underpinning Literature and Hypotheses Development

2.1. Job Demands and Resources Model

Over the past few decades, numerous conceptual models have emerged to elucidate the impact of work characteristics on the health and well-being of workers (Cooper 1998). One of the recent approaches in this regard is the Job Demands and Resources model (JDR model) developed by Demerouti and collaborators (Demerouti et al. 2001), drawing on various models and theories, including Hobfoll’s (1989) Resource Conservation Theory (Bakker and Demerouti 2017). This model emphasizes the importance of situational specificity and promotes the use of general constructs to enhance adaptability across diverse occupational contexts (van Veldhoven et al. 2005). The JDR model, originally formulated to explain burnout, has evolved over the years and now encompasses the examination of various factors influencing worker well-being (Bakker and Demerouti 2007, 2017).

The core premise of this model posits that all professional activities entail distinct job characteristics, which can be categorized into two general types: job demands and job resources (Bakker and Demerouti 2007, 2017). Job demands encompass “those physical, psychological, social or organizational aspects of the job that require sustained physical and/or psychological (cognitive and emotional) effort or skills and are therefore associated with certain physiological and/or psychological costs” (Bakker and Demerouti 2007, p. 312). In contrast, job resources refer to “those physical, psychological, social or organizational aspects of the job that are either/or functional in achieving work goals, reduce job demands and the associated physiological and psychological costs and stimulate personal growth, learning and development” (Bakker and Demerouti 2007, p. 312). The second premise of the JDR model revolves around the idea that job demands and resources trigger two distinct processes: a health-impairment process and a motivational process (Bakker and Demerouti 2017). This study focuses on the health-impairment process, where tension arises when job demands are high and resources are limited, resulting in a disproportionate balance between the demands faced by workers and the available resources. Several studies suggest that job demands, including work–family conflict, are associated with physical and psychological strain and various health issues (Bakker and Demerouti 2017; Demerouti et al. 2001).

2.2. Work–Family Conflict and Affective Well-Being

The substantial transformations within the domains of work and family, as noted by Kossek et al. (2011a), limit the available time and energy for employees to dedicate to their family lives, inevitably originating role conflict.

Thus, this inter-role conflict manifests itself in two directions: the work role encroaching on the family sphere (WFC) or the family role intruding upon the work domain (FWC); however, this study primarily focuses on the intrusion of the work sphere into the family domain. According to Amstad et al. (2011), a strong association exists between WFC and outcomes that manifest within the family context and between FWC and outcomes that manifest within the work domain. However, the relationship is also found to be robust with non-specific outcomes, such as overall well-being.

Indeed, numerous empirical studies have substantiated the adverse impact of work–family conflict on individuals’ well-being (Andrade 2015; Boswell and Olson-Buchanan 2007; Kossek et al. 2011a; Obrenovic et al. 2020). These effects manifest across a spectrum of negative emotional and psychological outcomes, encompassing psychological distress, depressive symptoms, clinical diagnosis of mood disorders, and emotional exhaustion (Ahmad et al. 2021; Allen et al. 2000; Amstad et al. 2011; Dettmers et al. 2016; Frone et al. 1996; Greenhaus et al. 2006; Kinnunen et al. 2004; Netemeyer et al. 1996). Affect, as an indicator of psychological health, encompasses the emotions we experience and express, with an impact on various aspects of our lives, including goal attainment, interpersonal interactions, behavior orientation, and life satisfaction (Davis et al. 2016; Verzeletti et al. 2016).

According to Eby et al.’s (2010) review of empirical research on the role of affect in work–family interaction, three perspectives can be highlighted: one where trait-based affect, i.e., certain traits, are seen as predisposing individuals to experience more intense emotional reactions to work–family interactions; a second perspective, which examines state-based moods and emotions in relation to work and family life, finding that as the demands of work and family life increase, individuals tend to experience more negative affective states and fewer positive affective states; and a third perspective that considers state-based global affective reactions, typically as consequences of work–family experiences (e.g., job satisfaction, family satisfaction).

Affective well-being is understood to encompass both positive (e.g., enthusiasm and alertness) and negative (e.g., overall perceived distress and anger; Watson et al. 1988) affect. Affect refers therefore to a mental state involving evaluative feelings. It encompasses a wide range of dispositions, moods, emotions, and generalized affective reactions to events and experiences (Eby et al. 2010).

Watson and Tellegen’s (1985) two-factor model of affect encompasses positive affect (PA) and negative affect (NA) as primary dimensions describing affective experiences. PA reflects excitement, activity, and alertness, while NA represents subjective distress and unpleasant involvement. Both dimensions have high relevance, capturing essential qualities of affective experiences (Gisler et al. 2018). Studies have demonstrated that NA is associated with stress (Clark and Watson 1986), health-related complaints (Tessler and Mechanic 1978), and exposure to unpleasant situations (Warr et al. 1983). Conversely, PA appears to be linked with the frequency of pleasant situations and overall satisfaction (Clark and Watson 1986, 1988; Watson 1988).

Indeed, affect is influential to adaptive behavior as it influences motivational behavior: positive affect may signal approach motivation, while negative affect represents withdrawal motivation (Watson et al. 1999). The feeling states associated with positive affect mobilizes goal-directed behavior and the feeling states associated with negative affect leads to apprehensiveness and cautiousness. It is therefore relevant to understand how work–family conflict may be impacting the experience of affect and what moderators can be relevant to increase positive experiences and decrease negative ones.

Studies have shown a relationship between WFC and positive affect, indicating that when work and family demands clash, it can impact individuals’ positive affective experiences (Davis et al. 2016; Kafetsios 2007). Similarly, there is evidence indicating a connection between WFC and negative affect (NA), suggesting that this conflict can also influence negative affective experiences (Judge et al. 2006). However, these correlations do not consistently hold true across all studies (e.g., Davis et al. 2016; Kafetsios 2007). This lack of consistent linkages between WFC and affective outcomes underscores the need for a more comprehensive analysis.

In this work, we bring together the JDR model and PA and NA constructs to delve into the psychological mechanisms inherent to the health-impairment process of the JD-R model and thus understand how individuals navigate and manage their emotional experiences amidst work–family conflict. See conceptual model depicted in Figure 1.

Thus, the following hypotheses are proposed:

H1. 

High WFC is associated with an increase in negative affect.

H2. 

High WFC is associated with a decrease in positive affect.

Furthermore, as WFC involves engaging with various stakeholders across work and family domains, each governed by distinct rules for emotional expression (Speights et al. 2020), individuals need to exert control over the emotions they experience, including determining which emotions arise, when they occur, and how they are expressed (Aldao et al. 2010; Kraft et al. 2024). Thus, by regulating their emotions (Gross 1998), individuals may modify emotional triggers or cognitive processes before emotions are elicited or manage the emotional response after it has been triggered. This way, by comprehending the affective ramifications of experiences involving work–family conflict and exploring the underlying mechanisms shaping this relationship, we gain deeper insights into the resultant emotional responses. These insights, drawing upon frameworks such as Gross’s (1998) work, shed light on the path of how individuals adaptively respond to the emotional challenges arising from conflicting work and family demands.

2.3. The Moderating Effect of Family-Supportive Organizational Perception and Psychological Detachment

The Job Demands-Resources (JD-R) model illuminates how work and personal resources act as protective shields against the detrimental effects of job demands on employee well-being. This underscores the importance of considering situational and individual characteristics as potential moderators in the relationship between WFC and affective well-being (Bakker and Demerouti 2007, 2017).

Among the various job resources discussed in the literature, social support stands out, and multiple studies have demonstrated its moderating effect on the association between job demands and stress (Andrade 2015; Bakker et al. 2007; Bakker and Demerouti 2007, 2017; Eng et al. 2010).

The organizational support theory introduces the concept of perception of organizational support, which reflects employees’ overall belief in how their employers value and prioritize their well-being and needs, providing the necessary support and resources to help them manage job demands. Another specific form of support is family-supportive organizational perception (FSOP), which pertains to how employees perceive their organization’s support and sensitivity regarding family-related matters, as well as the organization’s efforts to balance and facilitate work–family commitments and responsibilities (Allen 2001). This support includes policies that promote work–life balance, such as flexible work arrangements, maternity leave, and assistance with dependent care, among others (Chambel and Santos 2009; Kossek et al. 2011b). It is important to note that the mere existence of measures aimed at reconciling work and personal life is insufficient. Employees must also perceive these measures as useful, relevant, and well-adjusted to address their specific job demands. Feeling supported by their employers gives employees the confidence that they will receive assistance when needed.

Fredrickson and Levenson (1998) have highlighted the precedence of social support in fostering positive emotions and its potential to mitigate the affective consequences of job demands faced by employees. Over recent decades, a wealth of evidence has accumulated regarding the significance of social support in the workplace (Ilies et al. 2011), and its role as a moderator in mitigating various negative effects, including negative affect (George et al. 1993; Peeters et al. 1995).

The literature emphasizes how resources gained from professional pursuits, such as social resources like FSOP, not only bolster positive affect but also amplify performance in familial roles (Aryee et al. 2005; Chambel and Santos 2009; Greenhaus and Powell 2006; Voydanoff 2004). As individuals dedicate considerable effort to managing work demands, countering their adverse effects and halting resource depletion becomes imperative.

Recovery experiences are means for restoring and replenishing consumed resources (Demerouti et al. 2001; Demsky et al. 2014). Among these experiences, psychological detachment, integral to the recovery process (Sonnentag and Fritz 2007), emerges as a key factor in moderating the relationship between work demands and stress, as noted by Safstrom and Hartig (2013). Detaching from work has been found to predict positive outcomes, such as reduced tension and heightened well-being (Sonnentag 2012; Sonnentag et al. 2008; Sonnentag and Bayer 2005; Zijlstra and Sonnentag 2006). Psychological detachment allows individuals to distance themselves from the constant stimulation of work demands, facilitating a return to a balanced state. This enables workers to feel rejuvenated and regain the capacity to fulfill their remaining social roles (Sanz-Vergel et al. 2011; Gu et al. 2020). High levels of psychological detachment from work are expected to be associated with increased positive affect, while reduced detachment is linked to heightened negative affect. Therefore, when psychological detachment is lacking, the tension generated by work demands and situations persists, leading to the accumulation of stress and potential health issues (Sonnentag et al. 2008). Conversely, high psychological detachment enables effective recovery, as workers can fully focus on replenishing their resources and engaging in family experiences (Amstad and Semmer 2009; Sonnentag et al. 2008). It serves as a recovery strategy that can be perceived as a restoration strategy aiding the management of demands and facilitating well-being (Dettmers et al. 2016; Gu et al. 2020).

Considering the third and the fifth premise of the JD-R model, personal resources and work resources can play a similar role. Thus, this study regards FSOP as a work resource and PD as a personal resource. Consequently, we expect the following:

H3. 

The perception of family support by the organization (FSOP) will reduce the effect of WFC on PA and NA.

H4. 

Psychological Detachment will reduce the links from WFC to NA and PA.

3. Materials and Methods

3.1. Participants and Procedures

This study used data collected from a larger research project dedicated to the analyses of work–family boundary management in families with children aged between 13 and 18 years of age. As such, participants had to have a working partner with whom they cohabitate and have at least a child in the defined age range. The larger research project obtained a positive review from the Ethics Committee of the Faculty of Psychology and Education Sciences of the University of Porto (Reference number 5-9/2015) and from the National Data Protection Commission (Deliberation nº681/2016). Recruitment sites were a public school and a sports club. Recruitment via the researchers’ contact network was also conducted. Participants were approached either directly or through their adolescent children. In both cases, the interaction started by clarifying the study’s objectives and the processes taken to ensure confidentiality and anonymity. If participants (parents) agreed to participate, they were instructed to return their filled-in questionnaires in a sealed envelope directly to the investigator or to the teacher/trainer, who later handed them over to the research team. A response rate of 80.6% was obtained.

The final sample includes the same number of men and women (n = 390 in total), with men’s ages varying from 34 to 62 years (M = 46.85, SD = 0.34) and women’s ages varying from 31 to 57 years (M = 44.23, SD = 0.83). Most men (81.3%) and women (71.4%) had a secondary level of education or less (i.e., less than 12 years of education). Men worked an average of 44.46 h per week (SD = 0.83) and women an average of 39.79 h per week (SD = 0.65). Participants were mostly working full-time (men—79.9%; women—73.8%), on fixed schedules (men—55.4%; women—56.9%), were employed by others (75% of men and 82.8% of women) and worked for small companies (54.6% of men and 40% of women).

3.2. Measures

3.2.1. Work–Family Conflict (WFC)

The Portuguese version of the Work–Family Conflict Scale was used, addressing both directions of conflict, from work to family and from family to work (Vieira et al. 2014). The scale was composed of 7 items assessed on a 5-point Likert-type Scale (1 = totally disagree, 5 = totally agree). The reliability coefficients were very good (men’s Cronbach’s α = 0.89; women’s α = 0.87).

3.2.2. Family-Supportive Organizational Perception (FSOP)

This variable was approached by using the Family-Supportive Organization Perception Scale (Allen 2001), translated to Portuguese by Chambel and Santos (2009). This measure is composed of 14 items assessed on a 5-point Likert type Scale (1 = totally disagree, 5 = totally agree). Eleven items were recoded so that higher scores correspond to a more positive perception of the organization’s support for family. Item 13 (“Providing employees with flexibility in carrying out their work is seen as a strategic action in doing business.”) was excluded to help achieve an acceptable reliability coefficient for both men’s and women’s reports: men’s Cronbach’s α = 0.69; women’s Cronbach’s α = 0.69.

3.2.3. Positive and Negative Affect

The Positive Affect and Negative Affect Schedule (PANAS-X) was used (Watson and Clark 1994) in accordance with the reduced version proposed by Schoebi and collaborators (Schoebi et al. 2012). Thus, 8 items tapping both negative and positive feelings and emotions were used, and participants were asked to rate their typically affective state after work on a 5-point Likert scale (1 = very slightly or not at all, 5 = extremely). A principal component analysis with varimax rotation was performed for men and for women to ensure that the items were organized into factors (KMO = 0.76, X² (28) = 522.76, p < 0.001; KMO = 0.83, X² (28) = 684.77, p < 0.001). All items had communalities higher than 0.50 and factor loadings higher than 0.40 on their respective factor. One dimension aggregated 5 items tapping negative affect (scared/shaky, nervous, angry, irritable, sad/downhearted, from the basic negative emotion scale), with a total explained variance of 35% for men and 42% for women. The second dimension combined 3 items (happy, joyful, and confident from the basic positive emotion scale) into a positive affect factor, explaining 27% and 29% of men and women shared variance, respectively. Reliability coefficients were high for both valences and for both genders: PA α = 0.79 (men) and 0.83 (women); NA α = 0.82 (men) and 0.88 (women).

3.2.4. Psychological Detachment

We used the subscale of psychological detachment from work from the multidimensional Recovery Experience Questionnaire by Sonnentag and Fritz (2007), translated into Portuguese by Gonçalves (2015). The subscale is composed of four items answered on a 5-point Likert scale (1 = totally disagree, 5 = totally agree) and captures the individual’s ability to mentally disconnect from work during their free time and to abstain from work-related activities. High reliability coefficients were found for both men (α = 0.85) and women (α = 0.82).

3.3. Data Analysis

IBM SPSS (Statistical Package for Social Sciences) version 28.1 was used to input missing values and to perform Pearson’s Correlations and paired samples t-Test for comparing means across men’s and women’s variables. The number of missing values ranged from 1 to 16, and thus, fewer than 5% of missing values were found. Little’s test (Little 1988) is a valuable tool for evaluating the nature of missing data within a dataset, distinguishing between missing completely at random (MCAR) and missing not at random (MNAR) patterns. In this study, Little’s test was employed to ascertain the underlying structure of missing values. The results of Little’s test (Little 1988) indicated that the missing values exhibited a pattern consistent with missing at random (MAR) conditions, with a significance level exceeding 0.05, granting more confidence to the use of data imputation.

Because the scale used to address positive and negative affect was not previously validated with Portuguese participants, we performed an exploratory factor analysis with varimax rotation to establish the existence of two latent affect factors. In line with Costello and Osborne’s (2005) suggestions, items were expected to have communalities higher than 0.400 and saturation in the intended factor superior to 0.400. Kaiser–Meyer–Olkin (KMO) test was used to determine if the data were suited for factor analysis.

Furthermore, as Multiple Linear Regressions are sensitive to the existence of outliers, an inspection rendered 8 severe outliers who were excluded. Predictor variables (WFC, FSOP, and PD) were centered before the interaction term was calculated to avoid multicollinearity, and the independence of errors was verified by the Durbin–Watson test. The variable hours per week (HW) was included as a covariate in all models. A normal distribution of the residuals was ensured by using the bootstrap procedure. The assumption of homoscedasticity was ensured by asking for the estimation using the HC4 estimator as available in PROCESS, 4.1st macro version model 2 (Hayes 2013). The bootstrapped resampling procedure by randomly drawing 5000 samples and calculating the 95% bias-corrected bootstrapped confidence intervals allowed us to obtain more robust estimates (Hayes 2013). Since all study variables were tested using self-reports, we performed a factor analysis to test for the threat of common method bias with Harman’s single-factor test (Podsakoff et al. 2003). The results indicate that more than one factor had an eigenvalue greater than 1, and therefore, the threat of common method bias is unlikely. Preventive actions of common method bias were also considered and included clear instructions and a pilot of the survey to prevent misinterpretations of scale items, use of varying scale response options and guaranteed anonymity to participants.

4. Results

4.1. Descriptive Results

As shown in

Table 1. Mean and standard deviation by variable according to gender.

	Men		Women
Variable	M	SD	M	SD
WFC	3.06	0.90	3.01	0.84
FSOP	3.07	0.44	3.23	0.46
PD	2.81	0.90	2.90	0.88
NA	1.61	0.62	1.75	0.70
PA	3.42	0.64	3.28	0.67

Note: WFC = work–family conflict; FSOP = family-supportive organizational perception; PD = psychological detachment; NA = negative affect; PA = positive affect.

and

Table 2. Correlations for study variables disaggregated by gender.

Variable	HW	WFC	FSOP	PD	NA	PA
HW	-	0.004	−0.041	−0.053	0.066	0.003
WFC	0.183 *	-	−0.301 **	−0.035	0.476 **	−0.309 **
FSOP	−0.019	−248 **	-	0.064	−0.177 *	0.148 *
PD	−0.061	−0.107	−0.098	-	−0.212 **	0.017
NA	−0.002	0.460 **	−0.146 *	−0.114	-	−0.291 **
PA	−0.175 *	−294 **	0.051	−0.018	−313 **	-

Note: The results for the female sample are shown above the diagonal. The results for the male sample are shown below the diagonal. WFC = work–family conflict; FSOP = family-supportive organizational perception; PD = psychological detachment; NA = negative affect; PA = positive affect. HW = hours per week. ** p < 0.01. * p < 0.05.

, a similar pattern of correlations across study variables was found for men and women. Regarding both genders, WFC reports are negatively correlated to FSOP and PA and positively correlated to NA. Both genders’ FSOP reports are negatively correlated to NA. Regarding PA, the pattern is different; men’s FSOP does not correlate with PA, but women’s FSOP has a positive correlation. Women’s PD is negatively linked to NA, but men’s PD is not associated with any other variable of the study. Reports on NA and PA are negatively correlated for both genders. Regarding covariates, HW was found to have a negative correlation with PA and a positive correlation with WFC but only in the case of men.

The variables were also compared between genders, and we found significant differences in FSOP (t (202) = −3.85, p < 0.001, d = −0.27), PA (t (202) = 2.64, p = 0.01, d = 0.19), and NA (t (202) = −2.77, p = 0.001, d = −0.19). No gender differences were found for WFC (t (202) = 0.29, p = 0.77, d = 0.02) or PD (t (202) = −0.95, p = 0.34, d = −0.07). In particular, women perceive more organizational family support (women: M = 3.23, SD = 0.46; men: M = 3.07; SD = 0.44), more NA (women: M = 1.78, SD = 0.76; men: M = 1.62, SD = 0.65) and less PA (women: M = 3.23, SD = 0.72; men: M = 3.40, SD = 0.67).

4.2. Moderation Effect of Perceived Organizational Family Support

The moderator role of FSOP in the linkages between WFC and PA and NA was tested in multiple moderation models. Regarding positive affect, the model was found to explain 11% of the variance for men and women, respectively (

Table 3. Moderator effect of FSOP on the links between WFC and positive affect.

	Positive Affect (PA)
	Men					Women
Model Summary	R2	MSE	F(HC4)	df1	df2	R2	MSE	F(HC4)	df1	df2
	0.112	0.382	4.857 *	4	190	0.105	0.412	5.342 *	4	190
Model	b	SE	t	LL	UL	b	SE	t	LL	UL
WFC	−0.197 *	0.059	−3.194	−0.309	−0.075	−0.231 *	0.066	−3.205	−0.359	−0.101
FSOP	−0.016	0.100	−0.145	−0.211	0.185	0.088	0.137	0.542	−0.200	0.338
WFC × FSOP	0.144	0.113	1.086	−0.100	0.334	0.128	0.125	0.954	−0.124	0.378
HW	−0.008 *	0.004	−2.322	−0.015	−0.001	0.001	0.005	0.121	−0.008	0.011

Note: WFC = work–family conflict; FSOP = family-supportive organizational perception; HW = hours per week. * p < 0.05.

), while for negative affect, the model explained 23% of the variance for both genders (

Table 4. Moderator effect of FSOP on the links between WFC and negative affect.

	Negative Affect (NA)
	Men					Women
Model Summary	R2	MSE	F(HC4)	df1	df2	R2	MSE	F(HC4)	df1	df2
	0.232	0.297	14.218 *	4	190	0.233	0.383	14.919 *	4	190
Model	b	SE	t	LL	UL	b	SE	t	LL	UL
WFC	0.319 *	0.045	6.746	0.235	0.412	0.386 *	0.055	6.728	0.279	0.491
FSOP	−0.059	0.090	−0.637	−0.236	0.113	−0.052	0.115	−0.376	−0.244	0.211
WFC × FSOP	−0.151	0.073	−1.928	−0.300	−0.014	−0.059	0.143	−0.286	−0.261	0.294
HW	−0.006	0.004	−1.315	−0.014	0.003	0.005	0.006	0.734	−0.006	0.020

Note: WFC = work-family conflict; FSOP = family-supportive organizational perception; HW = hours per week. * p < 0.05.

). The model also showed that FSOP does not act as a moderator for either gender and for either type of affect. In fact, for men’s PA, we found only a main effect of WFC and of hours per week (HW): a higher number of working hours and higher levels of WFC are associated with lower levels of positive affect. Regarding women’s PA, only a main effect of WFC was found. Additionally, regarding negative affect, WFC is the only predictor both for men and women; i.e., higher levels of WFC are linked with higher levels of NA.

4.3. Moderation Effect of Psychological Detachment

We then tested the moderation effect of PD, with our findings revealing that the full model explains 11% and 10% of the variance of positive affect of men and women, respectively (

Table 5. Moderator effect of PD on the links between WFC and positive affect.

	Positive Affect (PA)
	Men					Women
Model Summary	R2	MSE	F(HC4)	df1	df2	R2	MSE	F(HC4)	df1	df2
	0.105	0.384	4.945 *	4	190	0.098	0.415	4.489 *	4	190
Model	b	SE	t	LL	UL	b	SE	t	LL	UL
WFC	−0.200 *	0.056	−3.364	−0.306	−0.088	−0.253 *	0.057	−4.158	−0.361	−0.135
PD	−0.040	0.054	−0.694	−0.147	0.064	−0.000	0.058	−0.005	−0.115	0.112
WFC × PD	0.001	0.063	0.011	−0.120	0.126	−0.044	0.079	−0.472	−0.194	0.114
HW	−0.008 *	0.004	−2.374	−0.016	−0.002	0.001	0.005	0.143	−0.007	0.010

Note: WFC = work–family conflict; PD = psychological detachment; HW = hours per week. * p < 0.05.

). PD, however, was not found to be a significant moderator of the relationship between WFC and positive affect. In fact, and similarly to the previous model, the main effect of WFC and working hours per week were significant for men, and only WFC was found significant for women.

In a similar vein, we tested for the moderation effect regarding negative affect and the models explained a higher proportion of the variance, ranging from 22% to 30%, for men and women, respectively (

Table 6. Moderator effect of PD on the links between WFC and negative affect.

	Negative Affect (NA)
	Men					Women
Model Summary	R2	MSE	F(HC4)	df1	df2	R2	MSE	F(HC4)	df1	df2
	0.224	0.300	12.789 *	4	190	0.296	0.351	15.330 *	4	190
Model	b	SE	t	LL	UL	b	SE	t	LL	UL
WFC	0.323 *	0.046	6.992	0.235	0.417	0.366 *	0.050	7.075	0.265	0.463
PD	−0.047	0.046	−1.019	−0.138	0.042	−0.171 *	0.047	−3.348	−0.269	−0.081
WFC × PD	0.002	0.040	0.046	−0.083	0.078	−0.153 *	0.053	−2.437	−0.262	−0.055
HW	−0.006	0.004	−1.342	−0.014	0.003	0.006	0.006	0.848	−0.005	0.012

Note: WFC = work–family conflict; PD = psychological detachment; HW = hours per week. * p < 0.05.

). Once again, PD was not a significant moderator, and WFC was the single predictor for men. For women, PD was found to be a moderator in the association between WFC and NA. While the standardized effect of the interaction is small, it should be considered that, especially in non-experimental studies, effect sizes for interactions are typically small (Frazier et al. 2004).

The moderation effect was further analyzed, and psychological detachment distribution was divided into three percentiles: bottom 16%, 64%, and upper 16% (Hayes 2018). Significant effects were found in all levels of the moderator (

Table 7. Conditional effects of psychological detachment on women’s negative affect.

Negative Affect (NA)
Conditional Effects (PD)	B	SE	t	LL	UL
−0.898 (16%) Inferior	0.503 *	0.076	6.580	0.352	0.654
0.102 (64%) Medium	0.350 *	0.052	6.721	0.247	0.453
1.102 (16%) Superior	0.197 *	0.086	2.287	0.027	0.368

Note: PD = psychological detachment. * p < 0.05.

), and as psychological detachment increases, the strength in the association between WFC and NA decreases. Figure 2 depicts these associations graphically.

5. Discussion and Conclusions

The primary objective of this study was to investigate the psychological mechanisms underlying the health impairment process in the JD-R model. Given the significance of positive and negative affect in the workplace and home settings, the study examined the relationships between these affect states, work–family conflict, and the moderating roles of family-supportive organizational perceptions and psychological detachment.

The pattern of correlations was rather similar between men and women. The correlations between FSOP and PA, however, diverge between genders, being non-significant for men and positively correlated for women. These differences can be discussed in light of the Boundary Theory (Clark 2000), which suggests a greater tendency for women to integrate both domains of their adult life. In the case of men, as their social roles are more separated, there may be less conflict, and thus, the availability of supportive resources may be perceived as less essential.

Our findings supported our initial two hypotheses. Specifically, we found that high levels of work–family conflict were associated with an increase in negative affect and a decrease in positive affect. These results align with the JD-R model and are consistent with previous research highlighting the connections between work–family conflict and negative affect (Cetin et al. 2021; French and Allen 2020; Gryzwacz et al. 2004; Judge et al. 2006). Furthermore, our findings contribute to the relatively limited literature demonstrating a relationship between positive affect and work–family conflict (Cetin et al. 2021; Davis et al. 2016; Sandrin et al. 2020).

Another objective of our study was to examine the role of family-supportive organizational perceptions as a job resource and psychological detachment as a personal resource in our previously confirmed relationships. However, our third hypothesis, which posited that the perception of a family-supportive organization would mitigate the impact of work–family conflict on positive and negative affect, was not supported. This was somewhat unexpected as it contradicts the JD-R model and previous studies that established this relationship (e.g., George et al. 1993; Mauno et al. 2007). In line with the conservation of resources theory (Hobfoll 1989), it would be expected that during periods of heightened tension and personal resource investment, the disposition to regain, replace, and acquire resources would be amplified (Salanova et al. 2006). Consequently, organizational support should have been perceived as an avenue for resource acquisition. This raises the prospect that FSOP may not be acknowledged as a valuable resource in managing the work–family conflict relationship and its affective consequences.

Other explanations can be considered for these results. One potential justification relates to our sample composition. Our participants, both men and women, primarily work for small companies compared to medium and large organizations. The literature suggests that the perception of organizational support tends to be higher in larger organizations (Ortiz-Isabeles and García-Avitia 2022). Larger companies typically have more comprehensive feedback systems, better employee tracking, and a greater sense of recognition, likely due to their superior financial resources and resource allocation capabilities (Hayton et al. 2012; Hutchison and Garstka 1996). Consequently, in smaller-sized companies, the perception of family-friendly support may not be robust enough to buffer the effects of work–family conflict on affect.

Another interpretation pertains to the implications of receiving support from the professionals’ perspective. According to Walsh and Cormack (1994), while professionals often express their support needs, receiving support can generate feelings of inequity, vulnerability, and powerlessness, posing a threat to self-esteem (Buunk and Hoorens 1992; Coyne and DeLongis 1986; Fisher et al. 1982; Stewart 1989). Therefore, receiving support may also adversely affect individuals, hindering the reduction in the negative impact of work–family conflict on affect.

It is worth noting that the moderation effect for men was nearly significant, raising questions about the power of our analyses. As a result, future research may benefit from replicating this study to provide further insights into these relationships.

Our final hypothesis (H4) posited that psychological detachment would mitigate the impact of work–family conflict on positive and negative affect. Our findings partially supported this hypothesis, as psychological detachment was found to buffer the effect of work–family conflict on negative affect for women.

The Boundary Theory (Allen and French 2023; Ashforth et al. 2000; Clark 2000) offers insight into these gender differences observed in our results. This theory suggests that individuals establish and maintain temporal, physical, and psychological boundaries to simplify and structure their environment. The gender disparities in our study may be attributed to the greater tendency among women to integrate both their work and family roles, leading to blurred boundaries between these domains (Kossek et al. 1999). This inclination is influenced by differing societal expectations regarding how work and family responsibilities should be jointly managed for men and women. While gender roles have evolved to become more balanced over time, women continue to bear a heavier load in the family domain, including household responsibilities. This increased responsibility results in women’s roles being more intertwined, making the boundaries between work and family more permeable and flexible in terms of time and physical space. As a result, women are better equipped to meet the demands of both roles (Ghislieri et al. 2017; Nsair and Piszczek 2021). Consequently, women may place more value on the ability to psychologically detach from work, seeing it as a more critical need than men.

Conversely, the observed gender differences in our study may also be attributed to variations in emotional regulation strategies. Various theoretical models have identified specific strategies as either adaptive or maladaptive over time (Aldao et al. 2010). Following Gross’s model (1995), response-focused strategies concentrate on regulating emotions after the initial trigger has occurred, in this case, WFC. It is common for individuals to engage in rumination, a phenomenon characterized by repetitive focus on the emotional experience, its causes, and consequences and often stems from individuals’ intentions to understand and resolve their concerns. However, in distressing situations, rumination has been shown to hinder effective problem-solving, which, oppositely, would be an adaptive response-focused strategy (Aldao and Nolen-Hoeksema 2010). Thus, rumination, a maladaptive coping mechanism, is often more prevalent among women (Ando et al. 2020; Calderwood et al. 2018; Demsky et al. 2019). The existing literature has consistently highlighted that women tend to engage in negative rumination about work, which can disrupt the restoration of psychophysiological systems during non-working hours (Demsky et al. 2019; Calderwood et al. 2018). In this context, psychological detachment may serve as an opportunity for women to interrupt these rumination processes, promoting better recovery.

Furthermore, our results indicate that the number of hours worked per week significantly affects men, not women. This can be explained by the fact that men, compared to women, may be better able to compartmentalize their work and family domains, largely due to experiencing fewer family demands (Nsair and Piszczek 2021; Rothbard 2001). Men’s capacity to segment these domains may be strained when they work longer hours, which, in turn, can diminish their positive affect. It is essential to acknowledge that societal gender norms also play a role in shaping these dynamics. Men often face social expectations that prioritize work over family commitments, making it less socially acceptable for them to prioritize their family roles over work responsibilities (Rothbard 2001). Consequently, they might endure the demands of their jobs due to these societal pressures, ultimately leading to a reduction in their positive affect.

Incorporating individual strategies into the JD-R theory holds theoretical and practical significance, shedding light on behaviors that facilitate optimal functioning in specific work contexts. This insight can guide organizations in promoting or training these behaviors (Bakker and Demerouti 2017). Therefore, organizations must recognize the significance of psychological detachment. This can be achieved through adjustments to work schedules, the provision of more breaks, offering leisure opportunities and activities, imparting emotional management skills to employees (Moreno-Jiménez et al. 2009), and fostering a culture that supports the differentiation between work and non-work life (Sonnentag et al. 2008). These practices are particularly important for female workers, as highlighted by our study.

These findings hold important theoretical implications as well. By encouraging the integration of different psychological mechanisms, such as affect, into the relationship between job demands and strain, we contribute to the refinement and advancement of the JD-R model. Affective experiences have a profound impact on both psychological and physical well-being. When specific emotions are triggered due to the associated physiological arousal, they can also drive behavioral tendencies, a concept of great interest in organizational psychology theory (Galinha et al. 2014; Weiss and Cropanzano 1996) and work–family conflict theory.

This study also yields practical implications. As previously established, the JD-R model posits that each profession possesses distinct working characteristics, categorizable as demands or resources (Bakker and Demerouti 2007). This makes it an applicable model to a wide array of occupational contexts, with particular relevance to professions where work–family conflict is a well-established demand (e.g., police officers, firefighters, health professionals). Moreover, the escalating prevalence of work–family conflict (WFC), propelled by multifaceted factors and its impact on well-being, points to potential effects on both positive and negative affect. Understanding these associations may help shape interventions, policies, practices, and organizational cultures that enable employees to navigate the core domains of their lives (Davis et al. 2016; Kossek et al. 2011a). Additionally, the affective states of workers are highly pertinent to their behavior in the organizational context. Therefore, the study of affect is crucial in understanding the relationships between affect, work attitudes, and workplace behaviors (Fritz et al. 2010).

All organizations must proactively support their workforce to navigate these demands and develop resources for emotional regulation, promoting both mental well-being and optimal performance. This entails recognizing the importance of emotion regulation, which, as previously mentioned, encompasses conscious and unconscious processes through which individuals modulate their emotions to adaptively respond to environmental demands. Within this framework, reappraisal and problem-solving strategies are regarded as adaptive approaches to managing emotions across diverse contexts, with reappraisal involving generating positive interpretations of stressful situations, aiming to reduce distress, and problem-solving entailing conscious efforts to alter or mitigate stressors. Moreover, mindfulness, the practice of non-judgmental acceptance of emotions, has emerged as a promising regulatory strategy associated with positive outcomes, emphasizing the need for organizations to incorporate such strategies into their support systems (Aldao et al. 2010; Aldao and Nolen-Hoeksema 2010; Pico et al. 2024).

The current investigation presents certain limitations that merit consideration. As most participants hold a conventional 9 h to 17 h work schedule, this may limit the applicability of our findings to settings involving non-traditional work hours, such as shift-based work. This circumstance overlooks additional challenges associated with schedules that more directly conflict with family life. Subsequent research endeavors should strive to encompass a broader spectrum of work schedules, thereby fostering a more comprehensive understanding of the explored relationships.

Moreover, most of our participants have educational qualifications equivalent to or lower than the mandated twelve years of schooling. This may signal that our sample is involved in low-skilled jobs, thus experiencing limited access to psychological support and work resources. Consequently, alternative moderators beyond organizational support or the capacity to psychologically detach from work may hold relevance. Hence, these findings may not readily extrapolate to individuals possessing higher educational attainments and qualifications.

Additionally, our data do not allow for the inference of causality between variables, particularly regarding the circular relationships between affect and perceptions of conflict. For instance, emotions exhibited at home could also influence the perception of negative work–life balance. Longitudinal studies are a valuable tool to explore these dynamics further. Furthermore, the reliance on self-reported measures might introduce common method bias. Finally, our affect measurement focused on five negative affect and three positive affect items; incorporating a more diverse array of emotional states could be beneficial to attain a broader scope.

Work–family conflict is a bidirectional concept, and to gain a more complete understanding on this phenomenon, future research should explore both directions of this conflict and their relationships with personal and work resources. Considering the impact of teamwork and examining perceptions of work characteristics at the team and department levels may also contribute to a more nuanced understanding of work–family conflict. Thus, several variables should be considered in future research, such as other forms of support. Examples encompass support derived from direct supervisors and colleagues. Direct supervisors have a more immediate influence on team conflict resolution and workload management. The perceived support from supervisors could correlate strongly with individuals’ affective states. Conversely, the team and work environment are intricately linked to the formal and informal relationships among coworkers. Thus, the support extended by peers may significantly impact an employee’s affective states. Additionally, other contextual factors necessitate consideration, such as work hours, potential time zone disparities, monthly workload variations, the families’ financial resources, and the size and accessibility of support networks. These factors often represent commonplace constraints in the interface between work and family life. Their potential to either exacerbate or alleviate an employee’s affective states may depend on the support available across these dimensions. Additionally, further research into the concept of working from home could yield valuable insights into practices that either alleviate or exacerbate work–family conflict.

In conclusion, despite the aforementioned limitations, our study contributes to the existing literature on the JD-R health-impairment process. Understanding the potential mechanisms that link work–family conflict to health is crucial for improving policies and designing targeted interventions to alleviate this source of stress in adults’ lives. Enhancing our knowledge of work practices, support systems, and organizational policies can have a positive impact on employee well-being as they navigate the intricate balance between their work and family roles (Eng et al. 2010). By incorporating the concept of psychological detachment, we have made a valuable addition to the existing literature on recovery experiences within the context of work–family conflict and organizational psychology.