Patient Advocacy Barriers & Facilitators Article Discussion

HHS Public Access Author manuscript Author Manuscript Psychol Bull. Author manuscript; available in PMC 2019 March 01. Published in final edited form as: Psychol Bull. 2018 March ; 144(3): 284–314. doi:10.1037/bul0000120. A Meta-Analysis of Work-Family Conflict and Social Support Kimberly A. French, MS, Department of Psychology, University of South Florida, 4202 E. Fowler Ave., PCD 4118G, Tampa, FL 33620, 847-989-9564 Soner Dumani, PhD, American Institutes for Research, 1000 Thomas Jefferson St NW, Washington, D.C., 20007 Author Manuscript Tammy D. Allen, PhD, and Department of Psychology, University of South Florida, 4202 E. Fowler Ave., PCD 4118G, Tampa, FL 33620 Kristen M. Shockley, PhD Department of Psychology, University of Georgia, 125 Baldwin St. Athens, GA 30602 Abstract Author Manuscript Author Manuscript The relationship between social support and work-family conflict is well-established, but the notion that different forms, sources, and types of social support as well as contextual factors can alter this relationship has been relatively neglected. To address this limitation, the current study provides the most comprehensive and in-depth examination of the relationship between social support and work-family conflict to date. We conduct a meta-analysis based on 1021 effect sizes and 46 countries to dissect the social support and work-family conflict relationship. Using social support theory as a theoretical framework, we challenge the assumption that social support measures are interchangeable by comparing work/family support relationships with work-family conflict across different support forms (behavior, perceptions), sources (e.g., supervisor, coworker, spouse), types (instrumental, emotional), and national contexts (cultural values, economic factors). National context hypotheses use a strong inferences paradigm in which utility and value congruence theoretical perspectives are pitted against one another. Significant results concerning support source are in line with social support theory, indicating that broad sources of support are more strongly related to work-family conflict than are specific sources of support. In line with utility perspective from social support theory, culture and economic national context significantly moderate some of the relationships between work/family support and work interference with family, indicating that social support is most beneficial in contexts in which it is needed or perceived as useful. The results suggest that organizational support may be the most important source of support overall. Social support is one of the most popular constructs in psychological scholarship. In 2016 alone, over 2,500 articles in PsychINFO list “social support” as a key subject. Social support’s popularity stems from its integral theoretical role as a means for reducing strain and improving health and well-being (Cohen & Wills, 1985; House, Umberson, & Landis, French et al. Page 2 Author Manuscript 1988). One such strain is work-family conflict, which occurs when the demands of work or family make it difficult to fulfill demands in the alternative role (Greenhaus & Beutell, 1985). Given recent societal trends such as increased use of technology, cross-national work, and dual-earner couple households, work-family conflict is recognized as a prominent societal concern and is studied by researchers around the world who span multiple disciplines (Aycan & Eskin, 2005; Duxbury, Higgins, & Lee, 1994; French & Johnson, 2016; Mortazavi, Pedhiwala, Shafiro, & Hammer, 2009; Shockley, Douek, Yu, Dumani, & French, 2017). Patient Advocacy Barriers & Facilitators Article Discussion

Author Manuscript In recent years, hundreds of studies focusing on social support within the work-family interface have been published in academic journals and presented at professional conferences. Overall, this research shows informal social support at home or at work negatively relates to work-family conflict (Kossek, Pichler, Bodner, & Hammer, 2011) and positively relates to beneficial well-being outcomes such as work and family satisfaction (Ford, Heinen, & Langkamer, 2007), mental health (Lee, Sudom, & Zamorski, 2013), cardiovascular health (Uchino, Cacioppo, & Kiecolt-Glaser, 1996), and sleep quality and quantity (Crain, Hammer, Bodner, Kossek, Moen, Lilienthal, & Buxton, 2014). Author Manuscript Although the importance and overall benefits of social support are clear, social support is a complex construct. For example, social support has been defined in diverse ways (Cohen & Wills, 1985; House et al., 1988) and as such it can be categorized into different forms (e.g., behaviors, perceptions; Barrera, 1986) and types (e.g., instrumental, appraisal, emotional support; Cohen & McKay, 1984). Social support also can come from a variety of sources (e.g., co-worker, supervisor, organization, family, spouse) (Ford et al., 2007). In addition, research suggests that the use and effectiveness of social support depends on culturally shared norms and expectations (Taylor, Sherman, Kim, Jarcho, Takagi, & Dunagan, 2004; Taylor, Welch, Kim, & Sherman, 2007). Author Manuscript As research on social support and work-family conflict has evolved, the complexity of social support has taken on greater relevance, setting the stage for our comprehensive and integrative review of how variations in social support alter the strength of the relationships between social support and work-family conflict. Moreover, the time is now ripe to examine how the broader societal context in which these relationships occur impacts the strengths of associations. With this in mind, the current study represents the most comprehensive and indepth examination of the relationship between social support and work-family conflict to date. Using meta-analysis, we investigate the relationship between work-family conflict and social support emanating from both the work and the family domains. We further differentiate support by specific form (i.e., behaviors and perceptions), source (e.g., spouse, organization, coworker), and type (i.e., emotional and instrumental). Moreover, we examine national-level cultural and economic context as moderators of these relationships. Figure 1 displays the relationships examined in the current study. Our synthesis of the work-family conflict and social support literature makes several key contributions. First, we provide a comprehensive and integrative quantitative review of the vast literature that connects social support from both the work and the family domains with work-family conflict. Previous meta-analyses have helped to paint parts of the overall Psychol Bull. Author manuscript; available in PMC 2019 March 01. French et al. Page 3 Author Manuscript Author Manuscript picture that depicts social support and work-family conflict (see Table 1). Early metaanalyses focused on aggregated measures of general work and/or family support (Byron, 2005; Mesmer-Magnus & Viswesvaran, 2005; Michel, Mitchelson, Kotrba, LeBreton, & Baltes, 2009). More recent meta-analyses examined different sources of support within the work domain (Mesmer-Magnus & Viswesvaran, 2006; Ford et al., 2007; Kossek et al., 2011) and within the family domain (Michel, Kotrba, Mitchelson, Clark, & Baltes, 2011). With empirical expansion of the primary study database, meta-analyses have begun to invoke theoretical rationale for why examining different aspects of social support matters. For example, two previous meta-analyses provide an empirical test of the domain specificity hypothesis, which contends that support from a given domain should most closely relate to directional conflict that also originates in that domain (e.g., work support versus family support more closely relates to conflict in the work-to-family direction) (Byron, 2005; Mesmer-Magnus & Viswesvaran, 2006). Most recently, Kossek and colleagues (2011) were the first to theorize how the relationship between social support and work-family conflict may vary in strength according to support source (organizational perceptions vs. supervisor support). Despite these advancements, many theoretical complexities associated with work and family social support remain under-recognized and empirically under-explored. Author Manuscript Our quantitative review updates and expands our understanding of the relationship between social support and work-family conflict both empirically and theoretically. We challenge the notion that social support measures are interchangeable by examining three theoretically distinct aspects of social support: form, source, and type. In doing so, we test foundational theory in the social support literature regarding the distinction and relative contribution of support forms (behaviors vs. perceptions), sources (broad vs. specific), and types (instrumental vs. emotional). For each distinction, social support is purported to function in unique theoretical ways (e.g., Cohen & Wills, 1985; Cohen & McKay, 1984; House et al., 1988). Patient Advocacy Barriers & Facilitators Article Discussion

Yet, to our knowledge, there has not been a large-scale, parsimonious test of these distinctions. Our analysis is conducted with an updated and considerably larger number of studies compared to previous meta-analyses, which allows for more current and precise effect size estimates. Author Manuscript To create this holistic picture and expand existing meta-analytic work, we investigate the distinction between measures of support behavior and support perceptions. Scholars have long debated how and why supportive perceptions and behaviors differ in their relationships with strain outcomes (Barrera, 1986; Lakey & Cohen, 2000). Our review brings this important consideration to the fore and contributes to a long-standing discussion in the social support literature. As such, we are the first to empirically test whether supportive perceptions and behaviors are similarly related to work-family conflict (Table 1). This question has implications for both the theoretical rationale that connects social support and work-family conflict, as well as the evaluation and implementation of social support initiatives designed to reduce work-family conflict. Similarly, we distinguish between emotional and instrumental support. Previous metaanalyses have yet to tease apart emotional and instrumental support (Table 1). This effort is critical given the long history of theoretical distinction (e.g., Cohen & McKay, 1984; House et al., 1988). Further, empirical evidence suggests emotional and instrumental support Psychol Bull. Author manuscript; available in PMC 2019 March 01. French et al. Page 4 Author Manuscript differentially relate to work-family conflict (Lapierre & Allen, 2006; Shockley & Allen, 2015). Understanding the distinction is important for advancing our theoretical understanding of factors that influence the magnitude of the support-work-family conflict relationship and for developing support interventions that can be used by organizations and family therapists to alleviate work-family conflict. Author Manuscript We also reconcile previously mixed findings regarding the domain specificity hypothesis. The domain specificity hypothesis has been primarily applied to work-family conflict but has implications for cross-domain interactions beyond conflict (e.g., spillover, enrichment) and beyond the work and family domains (e.g., leisure, friend relationships). Previous research has often assumed the domain specificity hypothesis holds, although strong support has yet to be found. Our meta-analysis is equipped with sufficient power to detect differences that may have been previously masked, and examines nuances in social support which may explain discrepant findings. Author Manuscript We further make a unique contribution to the literature by examining the context within which work and family support occurs. National context is difficult to meaningfully take into account in primary work-family studies, despite theoretical significance (Ollier-Malaterre & Foucrealt, 2016; Ollier-Malaterre, Valcour, Den Dulk, & Kossek, 2013; Powell, Francesco, & Ling, 2009). Patient Advocacy Barriers & Facilitators Article Discussion

However, context is important as recent meta-analytic findings shed light on systematic differences in levels of work-family conflict across cultures (Allen, French, Dumani, & Shockley, 2015). We make a novel contribution to this literature by systematically investigating how national context shapes relationships between work-family conflict and correlates. National context is especially critical for social support, given that support is a relational, socially enacted construct shaped by societal norms (Kim, Sherman, & Taylor, 2008). Our meta-analysis examines two distinct mechanisms of contextual influence, cultural values and economic context, providing empirical evidence where littleto-none exists (Ollier-Malaterre, 2016). Further, we examine context moderation by testing alternative competing hypotheses derived from two plausible theoretical perspectives: the utility perspective (Cohen & Wills, 1985) and the values perspective (Oishi, Diener, Lucas, & Suh, 1999; Oishi, Diener, Suh, & Lucas, 1999b). By employing this strong inferences paradigm (Platt, 1964), we provide systematic, overarching theoretical insight and guidance to the cross-national literature. Work-Family Conflict and Social Support Author Manuscript Work-family conflict occurs when demands from work and family domains are incompatible, impeding domain performance (Greenhaus & Beutell, 1985). Conflict can occur in two directions: work can interfere with the ability to meet family demands (WIF) or family can interfere with the ability to meet work demands (FIW; Frone et al., 1997b). Previous meta-analytic research confirms WIF and FIW are moderately correlated, but distinct (e.g., Michel et al., 2009; Shockley & Singla, 2011). Throughout the paper, we use the umbrella term work-family conflict when we refer to conflict in general and we employ WIF/FIW when we refer to specific directional conflict. Psychol Bull. Author manuscript; available in PMC 2019 March 01. French et al. Page 5 Author Manuscript Social support is one of the most widely studied contextual antecedents of work-family conflict. Although the definition and operationalization of social support has historically suffered from a lack of clarity and consensus (e,g., Cohen & Wills, 1985), two commonalities exist among definitions. First, social support is derived from social relationships. Second, social support protects an individual’s well-being under adverse circumstances (Cobb, 1976; House et al., 1988). We define social support in the current study as psychological or material resources provided through social relationships that can mitigate strains. Furthermore, support can come from either the work or the family domain. The terms “work support” and “family support” are used throughout the paper to refer to support that originates in the work and family domains, respectively. Author Manuscript Author Manuscript Cohen (1992a) delineated three core components of social support: social networks (existence, quantity, and types of social relationships), perceived support (perception that social relationships have provided resources), and supportive behaviors (the receipt of behaviors that help individuals manage strains). Patient Advocacy Barriers & Facilitators Article Discussion

These components can be sorted into two measurement groups: structural (social networks) versus functional (perceived support and support behaviors). Structural measures describe the existence of social relationships in an individual’s social network (e.g., marital status). Functional measures describe the functions provided by these relationships (e.g., provision of emotional resources). Functional support measures directly assess social support as they capture the transfer of support resources and/or quality of support (Cohen & Wills, 1985; House et al., 1988). In contrast, structural support measures indirectly assess social support, as they capture availability of supportive connections (Cohen & Wills, 1985; House et al., 1988). In the current study, we are interested in the transfer and quality of social support resources. As such, we focus on functional operationalizations of social support, including support perceptions and supportive behaviors. Researchers have identified three theoretical roles social support may play in the stress process (LaRocco, House, & French, 1980). Social support may have a direct mitigating effect on stressors, or social support may directly mitigate strains (main effect hypothesis; Cohen & Wills, 1985; LaRocco et al., 1980). Social support may alternatively serve as a buffer between stressors and strains (buffer hypothesis; LaRocco et al., 1980). This buffering may occur either during the appraisal process, mitigating perceptions of stressors, or after appraisal has taken place by providing solutions, facilitating healthy coping strategies, or decreasing problem importance (Cohen & McKay, 1984; Cohen & Wills, 1985). Author Manuscript Tests comparing the direct versus moderating roles of support find social support is most appropriately modeled as an antecedent to strains, such as work-family conflict (Carlson & Perrewe, 1999; Seiger & Wiese, 2009). This direct relationship is most consistent with the main effect hypothesis (Cohen & Wills, 1985). The direct antecedent role is buttressed by resource-based stress theories that conceptualize support as a resource that can be used to meet demands (e.g., conservation of resources, Hobfoll, 1989; job-demands resources model, Demerouti, Bakker, Nachreiner, & Schaufeli, 2001) and therefore avert work-family conflict (van Daalen, Willemsen, & Sanders, 2006). Patient Advocacy Barriers & Facilitators Article Discussion

Consistent with these theoretical perspectives, meta-analyses confirm work and family support have direct, negative Psychol Bull. Author manuscript; available in PMC 2019 March 01. French et al. Page 6 Author Manuscript relationships with WIF and with FIW (Ford et al., 2007; Kossek et al., 2011; MesmerMagnus & Viswesvaran, 2006; Michel et al., 2009). Author Manuscript Although meta-analyses indicate WIF and FIW share common correlates, the strength of these relationships differ (e.g., Amstad, Meier, Fasel, Elfering, & Semmer, 2011). Patterns tend to follow the domain specificity hypothesis (Frone et al., 1992; Frone et al., 1997b). The domain specificity hypothesis proposes that WIF most strongly relates to work domain antecedents because WIF originates in the work domain, whereas FIW most strongly relates to family domain antecedents because FIW originates in the family domain. Although numerous primary studies and virtually all previous meta-analyses on social support and work-family conflict have invoked this theory when developing hypotheses, meta-analytic empirical support is surprisingly sparse. Four meta-analyses have empirically tested the domain specificity hypothesis. Of these, two meta-analyses failed to find support for domain specificity (Mesmer-Magnus & Viswesvaran, 2006; Michel et al., 2011). Patient Advocacy Barriers & Facilitators Article Discussion

Two meta-analyses found partial support for domain specificity, in that work support was more strongly associated with WIF than family support (Byron, 2005; Mesmer-Magnus & Viswesvaran, 2005), but work and family support were similarly related to FIW (Byron, 2005). Given the relatively small number of primary studies in these meta-analyses (ks ranged from 2 to 31), it is unclear if lack of support reflects a true null finding or a lack of power (Byron, 2005; Mesmer-Magnus & Visewesvaran, 2005; 2006). Despite the lack of statistical significance, effect sizes appear to align with the domain specificity hypothesis (e.g., Mesmer-Magnus & Viswesvaran, 2005; Michel et al., 2011). Author Manuscript Based on theoretical and empirical evidence for a direct relationship between social support and work-family conflict, we exa Patient Advocacy Barriers & Facilitators Article Discussion