The role of family support in work-family balance and subjective well-being of SME owners

ABSTRACT

Drawing on the conservation of resources (COR) theory, we examine the role of two types of family support—emotional and instrumental support—in work-family balance and subjective well-being of small and medium enterprise (SME) owners. Using a sample of French SME owners, our results show that work-family balance mediates the relationship between family support and subjective well-being of SME owners. Surprisingly, while emotional support has a positive association, instrumental support has a negative association with the subjective well-being of SME owners via work-family balance. We provide a theoretical explanation by integrating COR theory with the threat to self-esteem model.

KEYWORDS:

• Small and medium enterprises
• owners
• instrumental support
• emotional support
• work-family balance
• subjective well-being
• conservation of resources theory
• threat to self-esteem model

Introduction

Work and family are commonly regarded as the two most important and closely intertwined domains in people’s lives (Frone, Russell, & Cooper, 1992; Hirschi, Shockley, & Zacher, 2019; Michel, Kotrba, Mitchelson, Clark, & Baltes, 2011). Within the entrepreneurship literature, the importance of family for entrepreneurs has been highlighted by several seminal works (Aldrich & Cliff, 2003; Rogoff & Heck, 2003). In support of this view, empirical research on the link between family and entrepreneurship generally indicates that family plays a major role in the startup behavior and success of entrepreneurs (e.g., Eddleston & Powell, 2012; Edelman, Manolova, Shirokova, & Tsukanova, 2016). However, while the literature on work-family interface of entrepreneurs has blossomed over the past few decades, less research has been conducted to understand the role of family support in small and medium enterprises (SMEs).

The dearth of research on the work-family interface of SMEs can be partly explained by the implicit assumption that entrepreneurs are homogeneous. Specifically, there is no clear consensus in the literature on what entrepreneurship refers to. Empirical research in the entrepreneurship literature frequently uses the term entrepreneur to represent SME owner, self-employed, and Schumpeterian entrepreneur (Carland, Hoy, Boulton, & Carland, 1984; Henrekson & Sanandaji, 2014). This assumption is problematic, as the motivation and orientation of these “entrepreneurial entities” tend to be quite different. Hence, conclusions drawn from research on one type of business may lead to erroneous implications when generalized to other businesses.

In this study, we focus specifically on SME owners. SMEs play a vital role in the modern economy (OECD, 2017). We define SME owner as a person who creates and actively manages a business with less than 250 employees for the primary purpose of serving their personal goals (Carland et al., 1984). We argue that there are three primary reasons why the influences of family support on SME owners warrant further attention. First, SME owners tend to experience higher stress and receive lower support from work than salaried workers and managers (Tetrick, Slack, Da Silva, & Sinclair, 2000). They often run a high personal risk of failure, juggle with multiple roles in the business, and have a high level of commitment to the performance of the business (Parasuraman & Simmers, 2001). Due to this combination of a high demand and low support work environment, the support received from family is a vital resource to cope with the stress experience at work. Second, the psychological characteristics of SME owners are generally different from entrepreneurs and managers. For example, Stewart, Watson, Carland, and Carland (1999) found that SME owners are less risk taking and less achievement oriented than entrepreneurs, but more risk taking than managers. Since the effect of social support depends on the psychological characteristics of the support recipient (Chay, 1993), SME owners may perceive family support differently when compared to other occupations. Third, unlike Schumpeterian entrepreneurs who manage their businesses with the primary goal of profit and growth, many SME owners choose to start a business to achieve their personal goal (Loscocco, 1997; Walker & Brown, 2004) and perceive the business as an extension of their personality (Carland et al., 1984). Given the relatively small size of the business, many SMEs also have a close tie with the family members of the owner (Feltham, Feltham, & Barnett, 2005; Loscocco & Leicht, 1993). For example, SME owners often seek labor or financial resources from their family members to support their business (Aldrich & Cliff, 2003; Haynes, Walker, Rowe, & Hong, 1999).

Social support generally refers to “social interactions or relationships that provide individuals with actual assistance or a feeling of attachment to a person or group that is perceived as caring or loving” (Hobfoll & Stokes, 1988, p. 499). One area that has attracted much attention in the social support literature is the relationship between social support and subjective well-being (SWB) (Jakubiak & Feeney, 2016; Kahn, Hessling, & Russell, 2003; Mikulincer & Florian, 1998; Rau, Georgiades, Fredrikson, Lemne, & De Faire, 2001). However, a vast majority of these studies pertain to the psychology literature, where the target populations are not SME owners. In fact, only a handful of studies have attempted to understand the link between family support and subjective well-being of SME owners (Chay, 1993; Nguyen & Sawang, 2016; Tetrick et al., 2000), thus leaving us with little understanding about the influence of family support on the SWB of SME owners.

Besides, most existing studies on social support are theoretically guided by the stress-buffering hypothesis (Lakey & Orehek, 2011). This theory posits that social support acts as a protective factor that attenuates the adverse effect of stress or negative life events on SWB (Cohen & Wills, 1985). Despite the domination of the stress-buffering hypothesis in the social support literature, review studies that have focused on the social support and SWB relationship have found almost no consistent evidence of the stress-buffering hypothesis (Burton, Stice, & Seeley, 2004; Lakey & Cronin, 2008). For example, while Tetrick et al. (2000) showed that social support moderates the association between emotional exhaustion and job satisfaction of SME owners, Nguyen and Sawang (2016) found only a positive association between social support and well-being of SME owners. Clearly, the link between social support and SWB is complex, for which an alternative theoretical model is warranted to gain a deeper understanding of this relationship.

In our quest for a theoretical model that helps to better understand the relationship between social support and SWB, we turned to the conservation of resources (COR) theory, a theoretical framework widely used in the general management literature (Hobfoll, 1989; Hobfoll, Freedy, Lane, & Geller, 1990). The basic premise of the COR theory is that individuals seek to retain and protect their current resources while attempting to obtain new resources (Hobfoll, 1989). Resources in the COR theory includes objects (for example, car), conditions (for example, job security), personal characteristics (for example, self-esteem), and energies (for example, knowledge) that are valued by the person. Although proposed as a theory of motivation, the COR theory has been widely applied to explain various individual outcomes such as burnout (Halbesleben, 2006), in-role performance (Halbesleben & Wheeler, 2008), extra-role performance (Eldor & Harpaz, 2016), and work-family conflict (Crawford, Shanine, Whitman, & Kacmar, 2016; Halbesleben, Harvey, & Bolino, 2009). Yet the COR theory has, to our knowledge, never been used as a theoretical framework in the context of SME owners.

In the present study, we draw on the COR theory to propose a model that examines the main effects of two types of family support (that is, instrumental and emotional support) on the SWB of SME owners. As we explain further in the later sections of this article, we propose that family support can be conceptualized as a resource that promotes the SWB of SME owners via instrumental and affective means. We further argue that the relationships between these two types of family supports (instrumental and emotional support) and SWB are mediated by work-family balance (WFB). More specifically, we contend that these two types of support provided by family members create a gain spiral that fuels a sense of role accomplishment (that is, WFB) for SME owners, which in turn increases their feeling of satisfactions and reduces their strains.

We decided to focus on WFB rather than work-family conflict or enrichment perspective due to three principle reasons. First, rather than focusing on either the positive or negative influence between the work and family domains, work-family balance provides a more holistic view of an individual’s perceived role accomplishment between the work and family domains. Second, while the conflict perspective has received much attention, there is a call in the literature for research to examine other work-family construct that has received relatively less attention (Greenhaus & Allen, 2011; Grzywacz & Carlson, 2007). Third, although previous studies have examined the work-family interface of SME owners (Nguyen & Sawang, 2016) and suggested that work-family balance is the main motivation to become SME owners (Gorgievski, Ascalon, & Stephan, 2011; Walker & Brown, 2004), to our knowledge no study to date has examined the antecedents and consequences of work-family balance among SME owners.

We define SWB as the presence of low strain and high satisfaction across different life domains (Diener, 2000). This multi-indicator approach has been widely adopted in the literature to provide an accurate understanding of SWB in a positive and a negative sense and across work and family domains (Hahn, Frese, Binnewies, & Schmitt, 2012; Lyubomirsky, King, & Diener, 2005).

To test our hypothesized model, we collected questionnaire data from 213 French SME owners in the construction sector via phone interviews. Our findings generally substantiate our argument with one intriguing finding. In particular, our results show that while emotional family support is positively associated with SWB of SME owners via WFB, instrumental support is negatively associated with SWB via WFB. We discuss this unexpected finding in the discussion section of this article, by complementing the COR theory with the threat to self-esteem (ToSE) model.

The present study contributes to the SME literature in several ways. First, we examine the influences of family support and its link with the SWB of SME owners, an inquiry that has been by far largely neglected despite its importance within the SME literature. Second, by examining the effect of two types of family support (that is, instrumental and emotional family supports), our findings contribute to the emerging literature that acknowledges the complexity of social support and emphasizes the importance of fit between individual needs and different kinds of support. Third, by investigating the mediating role of WFB in the relationship between family support and SWB, we extend previous findings that indicate the important and positive role of WFB in the SME context. Fourth, we enrich the SME literature by bringing in the theoretical insights of the COR theory. Finally, by integrating the COR theory with the threat to self-esteem model, we propose a theoretical framework that can guide future studies that aim to contribute to the work-family balance and small business management literature.

Theory and hypotheses development

The following section is organized as follows. First, we introduce the main premises of the COR theory. Then, we discuss our main constructs; namely, family support, SWB, WFB, and their relationships with the COR theory. Finally, we develop our hypotheses on the relationship between family support and WFB, the relationship between WFB and SWB, and the mediating role of WFB in the relationship between family support and SWB. In the present section, we have not limited ourselves to the literature pertaining only to SMEs in order to open up the field to adjacent ones.

Conservation of resources theory

According to the COR theory (Hobfoll, 2001), humans are motivated to retain, protect, acquire things that they find valuable, referred to as “resources,” and “position themselves so that they are less vulnerable to future resource loss” (Hobfoll, 2002, p. 317). The primary resources proposed by the COR theory are material resources (for example, any object of value), conditions (for example, tenure), energies (for example, time, knowledge, money), and personal resources (for example, personal traits). In addition, the COR theory proposes that there are secondary resources (for example, work, family), which help to gain or protect the primary resources. Since the value of a resource depends on the subjective evaluation of a person, the perceived value of each resource can vary significantly across individuals and cultures (Halbesleben, Neveu, Paustian-Underdahl, & Westman, 2014; Hobfoll, 2001).

The COR theory also suggests that, when not confronted with stressors, people develop resource surpluses, a “resource reservoir” that they draw from to facilitate resource gain and minimize resource loss in the future (Hobfoll & Shirom, 1993). One example is social relationships. Particularly, people who contribute to their social relationships (for example, friendship) would expect to gain resources (for example, self-esteem, relatedness, and sense of security) in the long term.

Another assumption of the COR theory is that the loss or gain of resources does not occur in isolation. In other words, initial loss or gain of resources leads to vulnerability or predisposition to future loss or gain of resources (Hobfoll, 1988). The loss and gain cycles are referred to as the loss and gain spirals, respectively (Hobfoll, 2001). Since resource loss is more salient than resource gain, the loss spirals are more salient in terms of strength and speed than the gain spirals (Hobfoll, 2001).

Social support

As noted above, social support is a form of interpersonal transaction that is generally provided with the goal of helping and caring for the recipient. Social support can be provided in a tangible form (for example, financial aid) or in an intangible form (for example, care or love). According to the COR theory, social support is an important means to expand personal resources (contained within oneself) to the broader environment (Hobfoll et al., 1990). In other words, people are motivated to maintain social support to meet their needs and to retain and protect their resources and identity; social support has also been shown to be a resource related to individual well-being in the face of stress (Hobfoll, 2009).

It is worth noting that although social support is often seen as a means to facilitate the accumulation of personal resources, it may hinder the accumulation of personal resources under certain situations (Gudmunson, Danes, Werbel, & Loy, 2009). For example, when a shared stressor runs in the family (for example, chronic diseases), social ties with family members may become a psychological burden that leads to resource loss. This complex nature of social support has been acknowledged in the COR literature (Hobfoll, 2002; Hobfoll et al., 1990). Thus, it seems that a more nuanced perspective is required to better understand the effect of social support, from both empirical (Brown, Nesse, Vinokur, & Smith, 2003; Deelstra et al., 2003) and theoretical points of view (Hobfoll, Halbesleben, Neveu, & Westman, 2018).

In the present study, we aim to contribute to this emerging literature by differentiating the effect of two types of family support: instrumental support and emotional support. While the differentiation between different types of social support have been examined in some studies (for example, Edelman et al., 2016; Kirrane & Buckley, 2004), to our knowledge no existing study has been conducted in the context of SMEs. Given the complex nature of social support and the uniqueness of SME owners as an occupation group, we argue that the effect of family support experienced by SME deserves further attention.

Our choice of focusing on instrumental support and emotional support was guided by three main reasons. First, these two types of support have received considerable attention and empirical validation in the literature (Beehr, Jex, Stacy, & Murray, 2000; King, Mattimore, King, & Adams, 1995; Shakespeare-Finch & Obst, 2011). Second, instrumental and emotional supports can be commonly found in the daily interaction between entrepreneurs and their family members (Aldrich & Cliff, 2003), and are closely related to the outcome of the work-family interface (Adams, King, & King, 1996; Michel et al., 2011). Third, to our knowledge, no known study has investigated the effects of these two types of family support separately in the context of SME owners.

Subjective well-being

Subjective well-being refers broadly to a person’s evaluation of the wellness and positivity of one’s life or other specific domains (for example, family and work). Given the overarching nature of SWB, various approaches have been adopted to define SWB in the literature. One common distinction is the use of a global (for example, Andrews & Robinson, 1991; Andrews & Withey, 1976) versus a multidimensional SWB scale (e.g., Parasuraman, Greenhaus, & Granrose, 1992; Parasuraman, Purohit, Godshalk, & Beutell, 1996). Another has been to define SWB as a state-like (for example, Conway, Clinton, Sturges, & Budjanovcanin, 2015; Kross et al., 2013) versus a trait-like construct (for example, Srivastava, Locke, & Bartol, 2001).

Since we intend to examine the association between social supports and SWB that are stable over time in the work and family domains, we adopted the multidimensional approach and define SWB as the combination of low strain and high satisfaction in the work and family domains (Diener, 2000). Specifically, we assess the SWB of SME owners with four different indicators: work satisfaction, family satisfaction, burnout, and work stress.

There are several advantages of adopting the multidimensional approach in this study. First, these indicators allow us to examine SWB dimensions in family and in work domains. Second, these indicators reflect positive (for example, satisfaction) and negative (for example, burnout) dimensions of SWB (Grant, Christianson, & Price, 2007). Finally, these indicators have been shown to be highly relevant when examining the SWB of SME owners (for example, Parasuraman & Simmers, 2001; Prottas & Thompson, 2006).

According to the COR theory, SWB is a primary resource that is commonly valued across cultures and individuals. Given the fundamental value of SWB as a resource, most previous studies on the COR theory have examined SWB as an outcome of their research models (Avey, Luthans, Smith, & Palmer, 2010; Newman, Tay, & Diener, 2014).

Work-family balance

The idea of balancing between work and family roles has gained much impetus from the popular press and the scholarly literature in the past few decades (Maertz & Boyar, 2011; Wayne, Butts, Casper, & Allen, 2017). Within the SME literature, there is consistent evidence that achieving work-family balance is one of the primary motivations to start one’s own business (Loscocco, 1997; Smith, 2000; Walker & Brown, 2004). Besides, prior studies consistently show that WFB is positively associated with various desirable personal and organizational outcomes such as organizational commitment, work satisfaction, family satisfaction, and performance (Brough & O’Driscoll, 2010; for a review, see Eby, Casper, Lockwood, Bordeaux, & Brinley, 2005).

In this study, we view WFB as the global evaluation of the role-related accomplishment in work and family domains that is unique for each individual (Voydanoff, 2005). Specifically, we define WFB as the “accomplishment of role-related expectations that are negotiated and shared between an individual and his or her role-related partners in the work and family domains” (Grzywacz & Carlson, 2007, p. 458).

Unlike other definitions of WFB that focus on only certain dimensions (for example, satisfaction, time allocation), this definition includes all role-related expectations that are considered to be relevant by the person after negotiating with the parties involved in each domain. Moreover, rather than emphasizing the actual equality between work and family domains, this definition focuses on the subjective equality perceived by the person (Greenhaus, Collins, & Shaw, 2003). Finally, by focusing on the global assessment of work and family domains, this definition clearly distinguishes WFB from other work-family constructs that specify the work-family linking mechanisms such as work-family conflict and enrichment (Edwards & Rothbard, 2000; Wayne et al., 2017).

Family supports and work-family balance

Emotional support and work-family balance

Emotional support can be defined as behaviors or attitudes of family members that are aimed to provide encouragement, understanding, attention, and positive regard (King et al., 1995). Research indicates that social support is negatively associated with work-family conflict (Grzywacz & Marks, 2000) and positively associated with integration of work and family roles (Greenhaus & Parasuraman, 1999). Furthermore, family support is positively associated with WFB among salaried workers (Ferguson, Carlson, Zivnuska, & Whitten, 2012; Russo, Shteigman, & Carmeli, 2016). Within the entrepreneurship literature, Gudmunson et al. (2009) found spousal emotional support to be indirectly linked to WFB via satisfaction with business communication among family business owners. However, no known study has examined the association between emotional family support and WFB of SME owners.

Drawing on the COR theory, we contend that emotional support of family is generally viewed by SME owners as a resource that nurtures their sense of WFB (Hobfoll, 1989). For instance, family member may fuel the sense of WFB by acknowledging the contributions and accomplishments of the SME owner in family and work domains. In particular, the understanding and encouragement expressed by family members may promote WFB of SME owners by enhancing their positive experience in managing the work-family interface (Shelton, 2006). Furthermore, emotional support may create a sense of intimacy, trust, and closeness, which in turn fuels positive self-image (for example, perceived role accomplishment) in the SME owners (Hobfoll et al., 1990). Besides, since the focus of emotional support is to provide a feeling of belonging and reassurance, we argue that emotional support is more likely to be perceived as a resource (rather than a threat) for the preservation of other resources like WFB. Accordingly, we hypothesize that:

Hypothesis 1 (H1): Emotional support provided by family members is positively associated with WFB of SME owners.

Instrumental support and work-family balance

Instrumental support can be defined as behaviors and attitudes of one’s family members that are aimed to facilitate their everyday operations (King et al., 1995). Unlike emotional support, instrumental support is provided with the aim of completing certain tasks for the support recipient (Beehr & Bhagat, 1985; House, 1981).

Previous studies suggested that the operation of SMEs depends significantly on the instrumental support of family members (Aldrich & Cliff, 2003; Loscocco & Leicht, 1993; Olson et al., 2003). On the one hand, family members may instrumentally support the SME owner by doing the household chores or taking care of children and older people. On the other hand, family members may provide business advice as well as labor and financial resources to support the SME owner’s business. Supporting the importance of instrumental support, Eddleston and Powell’s (2012) study with entrepreneurs found that satisfaction with WFB is nurtured by instrumental family-to-business enrichment (for example, spillover of positive mood from home to work) and instrumental support at home (for example, help with routine household tasks). Edelman et al. (2016) found that while family social capital is positively associated with the scope of startup activities, family financial capital is negatively associated with the scope of startup activities. However, no existing research has examined the association between instrumental family support and WFB among SME owners.

Guided by the COR theory, we argue that instrumental family support is generally perceived by SME owners as a resource that promotes the preservation of other resources (Ferguson et al., 2012; Hobfoll, 1989). Specifically, by aiding the SME owners to complete certain tasks, instrumental support limits the loss of their energy and time (thus preserving it), which leads to their sense of role accomplishment in work and family domains. Hence, we hypothesize that:

Hypothesis 2 (H2): Instrumental support provided by family members is positively associated with WFB of SME owners.

Work-family balance and subjective well-being

Achieving balance between work and family life has become a major concern for today’s workforce (Hirschi et al., 2019). This is particularly true for SME owners, as their choice to start a business is often motivated by nonfinancial factors such as a flexible life style and personal satisfaction (Walker & Brown, 2004). Supporting the importance of WFB, previous studies consistently show that WFB is associated with various positive outcomes in work and family domains (Aryee, Srinivas, & Tan, 2005; Carlson, Grzywacz, & Zivnuska, 2009; Ferguson et al., 2012). However, most existing studies have been conducted using salaried workers; research examining the link between WFB and SWB among SME owners remains sparse.

Extending previous findings in the management literature, and hinging on the COR theory, we argue that WFB is a condition (hence, a resource), that promotes the SWB (that is, high satisfaction and low strain) of SME owners. Specifically, the experience of WFB acts as a “resource reservoir” (Hobfoll, 2001) that can be used to facilitate resource gain and avoid resource loss. For instance, SME owners who experience WFB may have more time and energy to explore a new business opportunity or plan for their family life. As suggested by the COR theory, any resource gains or surplus are likely to result in “eustress,” which is improved SWB (Cohen & Edwards, 1989; Hobfoll, 1989). Accordingly, we propose that:

Hypothesis 3: WFB of SME owners is positively associated with their (H3a) work satisfaction and (H3b) family satisfaction, and negatively associated with (H3c) burnout and (H3d) work stress.

The mediating role of work-family balance in the relationship between family supports and subjective well-being

In the previous sections, we proposed that the two types of family support – emotional and instrumental – are associated with WFB, and that WFB is associated with SWB. By putting these hypothesized relationships together, we argue that the relationships between these two types of family support and SWB are mediated by WFB.

According to the COR theory, while social support is valued by itself, it is also important because it contributes to the maintenance of resource reservoir (Hobfoll, 2001). In this study, we contend that social support contributes to the well-being of SME owners by creating a sense of balance between their work and family role. Particularly, WFB resulting from emotional support and instrumental support (that is, resources gain) would initiate a gain spiral (Hobfoll, 2001) that facilitates further resources gain and contributes to positive SWB. For example, SME owners who are encouraged and comforted by their partners may have a better view on their role accomplishments (that is, work-family balance) and thus feel more satisfied with their family and work. Similarly, by helping with the household chores or work tasks, family members may help SME owners to manage their work and family role demands and thus feel less stress.

Surprisingly, the literature remains silent about the mediating role of WFB in the relationship between social support and SWB. One exception to this is the study by Ferguson et al. (2012), which found the relationships between social supports and both job and family satisfaction of salaried employees to be mediated by WFB. In light of the COR theory, we hypothesize that WFB mediates the relationship between emotional support and SWB as well as the relationship between instrumental support and SWB. All the hypotheses of the present study are summarized in Figure 1.

Hypothesis 4: The relationship between emotional support and (H4a) work satisfaction, (H4b) family satisfaction, (H4c) burnout, and (H4d) work stress is mediated by WFB of SME owners.

Hypothesis 5: The relationship between instrumental support and (H5a) work satisfaction, (H5b) family satisfaction, (H5c) burnout, and (H5d) work stress is mediated by WFB of SME owners.

Figure 1. Graphical summary of the hypothesized paths.

Method

Data collection procedure

A list containing contacts of 4,000 SME owners was obtained from a insurance company in France. SME owners on this list were contacted one at a time until we obtained a pool of 350 SME owners who agreed to participate in our study. This pool of SME owners was then contacted throughout the year to complete a total of five questionnaires via telephone interviews (conducted by Amarok France¹). The data pertaining to the present study were included in the fifth questionnaire. The response rate dropped from 349 responses in the first questionnaire, to 245 responses in the fifth questionnaire.

Sample

In total, 245 SME owners responded to our survey. Since most of the SMEs in our sample belong to the construction industry, we decided to focus only on them and excluded SME owners from other industries. This resulted in our final sample of 213 SME owners from the construction industry.

The mean number of employees in our final sample was 30, with a standard deviation of 50. The average age of the respondents was 52 years old, with a standard deviation of 8.02. The majority of the respondents were men (78.4 percent); most were married (74.6 percent) or lived together with a partner (8.9 percent). Only 4.2 percent of the respondents had no children while 85.9 percent had one to three children. Most of the respondents (63.8 percent) had a secondary education degree or above.

Measures and items

In this section we describe the variables used in the present study. Since the data was collected in France, all the measures were translated into French by professional translators using the forward-backward translation process. A full list of the translated items can be found in Appendix A. The correlation coefficients and Cronbach’s alphas are listed in Table 1.

Table 1. Correlations and descriptive statistics of all the variables.

Variables	Mean	SD	1	2	3	4	5	6	7	8	9	10	11	12	13
1. Sex^a	–	–	–
2. Age	51.55	8.02	−.00	–
3. Education^b	–	–	−.00	−.21**	–
4. Number of children	2.31	.97	.00	.17*	.06	–
5. With family partner^c	–	–	.14*	−.16*	.04	.11	–
6. Work support	3.40	.91	−.06	.13	−.05	.19**	−.13	(.88)
7. Emotional support	3.51	.93	.11	−.03	−.01	−.05	.12	.16*	(.79)
8. Instrumental support	3.69	.95	.28**	−.04	.02	.08	.28**	.02	.29**	(.75)
9. WFB	3.80	.65	−.17*	.17*	−.15*	.03	−.10	.24**	.20**	−.08	(.88)
10. Work satisfaction	4.25	.74	−.05	.08	.04	.10	.03	.31**	.19**	−.02	.32**	(.82)
11. Family satisfaction	3.95	.96	.06	.08	−.11	−.01	.12	.16*	.44**	.28**	.39**	.22**	(.93)
12. Work stress	3.90	.87	.13	−.19**	.04	−.10	.02	−.27**	.03	.09	−.25**	−.32**	−.20**	(.86)
13. Burnout	2.91	1.27	−.06	−.11	.02	−.07	.00	−.29**	−.05	.09	−.39**	−.47**	−.29**	.55**	(.90)

N = 213. Cronbach’s alphas are showed in parentheses along the diagonal.

^a 0 = female, 1 = male.

^b 1 = primary school or below, 2 = vocational diploma, 3 = high school, 4 = bachelor’s degree, 5 = master’s degree, 6 = PhD.

^c 0 = No family partner, 1 = With family partner.

*p < .05, **p < .01.

WFB = work-family balance.

Family support

Family support was measured by 17 items developed by King et al. (1995). This measurement contains items on instrumental and emotional support, and has been validated and widely adopted in previous studies (for example, Adams et al., 1996; Wayne, Randel, & Stevens, 2006).

Respondents were asked to indicate the extent to which they agreed or disagreed with the 17 statements on a 5-point scale (1 = strongly disagree, 5 = strongly agree). Sample items of instrumental support and emotional support are “When I’m having a difficult week at my job, my family members try to do more of the work around the house” and “Someone in my family helps me feel better when I’m upset about my job.”

Since we sought to examine instrumental and emotional support as two separate constructs, exploratory factor analysis (EFA) was conducted using SPSS 23 to extract items corresponding to each of the two constructs. Because the Shapiro-Wilk test showed that the assumption of normality was violated and it was unlikely that instrumental support and emotional support were uncorrelated, principal axis factoring with Promax rotation was used for the EFA (Fabrigar, Wegener, MacCallum, & Strahan, 1999).

Results showed that the five factor solution had eigenvalues greater than one (Kaiser, 1960), and the scree plot suggested a solution with three factors (Cattell, 1966). To confirm the factor structure, confirmatory factor analyses were conducted to test between the two-factor and three-factor models. Specifically, items with factor loading larger than 0.4 were selected (Corry, Merritt, Mrug, & Pamp, 2008) to construct the two-factor and three-factor models. Results showed that the two-factor model had a better fit than the three-factor model ($\mathrm{\Delta }{\chi }^2$ (32) = 77.47, p < .001). Accordingly, we selected the two-factor model as our final model. This model includes four items for instrumental family support (that is, Items 2, 3, 5, 9) and four items for emotional family support (that is, Items 4, 12, 15, 16). None of the selected items had a cross-factor loading with other factors.

Work-family balance

WFB was assessed by six items developed by Carlson et al. (2009). As in the case of family support, respondents were asked to indicate the extent to which they agree or disagree with the statements on a 5-point scale (1 = strongly disagree, 5 = strongly agree). Sample items included: “I am able to negotiate and accomplish what is expected of me at work and in my family” and “People who are close to me would say that I do a good job of balancing work and family.”

Burnout

Burnout was assessed using a shortened burnout measure (BM) developed by Malach-Pines (2005). The shortened version consists of ten items adopted from the original BM developed by Pines and Aronson (1988). The original BM has been shown to be reliable and valid across various occupations (Pines, 1993; Pines & Aronson, 1988), and used in different countries including France (Malach-Pines, 2005). Similar to the original BM, the shortened version defines burnout as a state of physical, mental, and emotional exhaustion (Pines & Aronson, 1988). Respondents were asked to rate how they feel when thinking about their work on a 7-point scale (1 = never, 7 = always). Sample items were “hopeless” and “depressed.”

Work stress

Work stress was assessed using a four-item measurement (Motowidlo, Packard, & Manning, 1986) with a 5-point scale (1 = strongly disagree, 5 = strongly agree). Sample items included: “My job is extremely stressful” and “I feel a great deal of stress because of my job.”

Family satisfaction

Following Edwards and Rothbard (1999), we measured family satisfaction by adopting three items from the Job Diagnostic Scale developed by Hackman and Oldham (1975). Respondents were asked to rate these items on a 5-point scale (1 = strongly disagree, 5 = strongly agree). Sample items included: “In general, I am satisfied with my family life” and “My family life is very enjoyable.”

Work satisfaction

Work satisfaction was assessed using the four-item Michigan Organizational Assessment Questionnaire Job Satisfaction Subscale (MOAQ-JSS), developed by Cammann, Fichman, Jenkins, and Klesh (1979). Previous studies have demonstrated that MOAQ-JSS is a reliable and valid measure of job satisfaction (Bowling & Hammond, 2008). Respondents were asked to report the degree to which they agree or disagree on a 5-point scale (1 = strongly disagree, 5 = strongly agree). One of the items was “All in all, I am satisfied with my job.”

Control variables

We controlled for five individual characteristics of SME owners: age, education, number of children, presence or absence of a partner, and support at work. These characteristics were selected based on previous studies (Aryee et al., 2005; Eddleston & Powell, 2012; Powell & Eddleston, 2013). Support provided by subordinates and coworkers were measured by the five items developed by Goldenhar, Swanson, Hurrell, Ruder, and Deddens (1998). Respondents were asked to report how often their subordinates or coworkers showed the behaviors described on a 5-point scale (1 = never, 5 = always). One of the sample items was “show support when you have a life problem.”

Results

Means, standard deviations, and zero-order correlations of all variables are presented in Table 1. To test our hypotheses with structural equation modeling (SEM), we adopted the two-step modeling procedure suggested by Anderson and Gerbing (1988). We first examined the model fit of our measurement model (the relationships between the latent variables and their items). After that, we assessed our structural model (the relationships between latent variables) by testing the hypothesized model and compared it with an alternative model, in which the paths between two types of support and SWB were removed. Finally, we examined the significance of the path estimates as well as the indirect effects using the bias-corrected bootstrapping approach (Mackinnon, Lockwood, & Williams, 2004). All the SEM based analyses were conducted using Mplus 8.0 (Muthén & Muthén, 2017).

Measurement model

We first estimated our measurement model (eight-factor model) using all the latent constructs. These factors are instrumental family support, emotional family support, WFB, support received at work, burnout, work stress, family satisfaction, and work satisfaction. All items were loaded on their respective latent factor. Results of confirmatory factor analysis (CFA) with maximum likelihood (ML) estimation showed that the eight-factor model was acceptable, χ² (674, N = 213) = 1021, p < .01, confirmatory fit index (CFI) = .92, root mean square error of approximation (RMSEA) = .05, and standardized root mean square residual (SRMR) = .06. To test whether instrumental and emotional supports are indeed separate phenomena, we tested an alternative model (seven-factor model) in which all the items of instrumental support and emotional support were loaded on a single factor, family support. Results showed that the seven-factor model had a poorer fit with the data, χ² (681, N = 213) = 1170.31, p < .01, CFI = .86, RMSEA = .06, and SRMR = .07, than the eight-factor model. The result of chi-square difference test confirmed that the fit of the eight-factor model was indeed significantly better than that of the seven-factor model ($\mathrm{\Delta }{\chi }^2$ (7) = 149.31, p < .001). Furthermore, all the items had significant loading on their respective latent factor (p < .001) with standardized loadings ranging from .50 to .91 in the eight-factor model. Thus, the eight-factor measurement model, which defines instrumental support and emotional as two separate constructs, was adopted to test our structural model and hypotheses.

Structural model

To test our hypotheses, we first tested our initial model, M₀, by estimating the hypothesized paths together with the paths between two types of support and SWB as well as the paths between all the control variables and dependent variables (that is, WFB and SWB).

Results indicated an acceptable model fit as recommended by Vandenberg and Lance (2000), χ² (810, N = 213) = 1241.99, p < .001, CFI = .90, RMSEA = .05, and SRMR = .06. In fact, although the CFI of the hypothesized model was lower than the recommended value, the RMSEA and SRMR met the required thresholds.

To assess whether M₀ best captured the proposed relationships, we compared it to an alternative nested model, M₁, in which the direct paths between two types of support and SWB were removed. The alternative model showed poorer model fit than the initial model, χ² (818, N = 213) = 1298.17, p < .01, CFI = .89, RMSEA = .05, and SRMR = .08. This was supported by the chi-square differences test between M₀ and M₁ ($\mathrm{\Delta }{\chi }^2$ (8) = 56.18, p < .001). Therefore, we further examined the hypothesized paths with the results of our initial model (M₀).

Emotional support (H1) and instrumental support (H2) were both hypothesized to be positively associated with WFB. We found that indeed emotional support was positively associated with WFB (β = .29, p < .01). Instrumental support, however, was negatively associated with WFB (β = −.19, p < .05). We hypothesized that WFB is positively associated with work satisfaction and family satisfaction (H3a and H3b) and negatively associated with burnout out and work stress (H3c and H3d). Results provided support for these hypotheses. We found WFB to be positively associated with work satisfaction (β = .30, p < .001) and family satisfaction (β = .37, p < .001), and negatively associated with burnout out (β = −.39, p < .001) and work stress (β = −.27, p < .01).

To test H4 (that is, the relationship between emotional support and SWB is mediated by WFB) and H5 (that is, the relationship between instrumental support and SWB is mediated by WFB), we followed the recommendation of Preacher and Hayes (2004) and Shrout and Bolger (2002) to test the mediation effect using the bias-corrected (BC) bootstrapping with 1,000 bootstrapped samples. Results provided support for H5a to H5e and partial support for H6a to H6e. In particular, the relationship between emotional supports and SWB (work satisfaction BC bootstrap 95% CI [.03 .20]; family satisfaction BC bootstrap 95% CI [.04 .22]; burnout BC bootstrap 95% CI [−.22 − .04]; work stress BC bootstrap 95% CI [−.17 − .03]) was mediated by WFB. Similarly, the relationship between instrumental supports and SWB (work satisfaction BC bootstrap 95% CI [−.17 − .00]; family satisfaction BC bootstrap 95% CI [−.17 − .00]; work stress BC bootstrap 95% CI [.00 .12]) was mediated by WFB except for burnout (BC bootstrap 95% CI [−.00 .17]).

In addition to the hypothesized path, we examined the direct effect between family supports and SWB. Results showed that only the effect between instrumental support and family satisfaction (β = .26, p < .05), as well as emotional support and family satisfaction (β = .31, p < .01) were significant.

The results in terms of path coefficients, standard errors, and BC bootstrap confidence intervals are summarized in Table 2.

Table 2. Standardized path coefficients, standard error, and BC bootstrap confidence interval.

			BC bootstrap 95% CI
Hypothesis	β	SE	Lower	Upper
Hypotheses 1 to 3
H1. Emotional support → WFB	.29**	.09	–	–
H2. Instrumental support → WFB	−.19*	.09	–	–
H3a. WFB → work satisfaction	.30***	.08	–	–
H3b. WFB → family satisfaction	.37***	.07	–	–
H3c. WFB → burnout	−.39***	.07	–	–
H3d. WFB → work stress	−.27**	.08	–	–
Hypotheses 4 to 5 (Indirect paths)
H4a. Emotional support → WFB → work satisfaction	.09	.04	.03	.20
H4b. Emotional support → WFB → family satisfaction	.11	.04	.04	.22
H4c. Emotional support → WFB → burnout	−.11	.05	−.22	−.04
H4d. Emotional support → WFB → work stress	−.08	.03	−.17	−.03
H5a. Instrumental support → WFB → work satisfaction	−.06	.04	−.17	−.00
H5b. Instrumental support → WFB → family satisfaction	−.07	.04	−.17	−.00
H5c. Instrumental support → WFB → burnout	.08	.04	−.00	.17
H5d. Instrumental support → WFB → work stress	.05	.03	.00	.12
Direct paths
Instrumental support → work satisfaction	−.08	.09	–	–
Instrumental support → family satisfaction	.26*	.11	–	–
Instrumental support → burnout	.03	.10	–	–
Instrumental support → work stress	.05	.11	–	–
Emotional support → work satisfaction	.15	.10	–	–
Emotional support → family satisfaction	.31**	.10	–	–
Emotional support → burnout	.04	.11	–	–
Emotional support → work stress	.15	.11	–	–
Control variables
Age → WFB	.13	.08	–	–
Age → work satisfaction	.02	.07	–	–
Age → family satisfaction	.04	.07	–	–
Age → burnout	−.04	.07	–	–
Age → work stress	−.10	.07	–	–
Education → WFB	−.12	.06	–	–
Education → work satisfaction	.10	.07
Education → family satisfaction	−.04	.07	–	–
Education → burnout	−.07	.07	–	–
Education → work stress	−.06	.07	–	–
Number of children → WFB	.00	.07	–	–
Number of children → work satisfaction	.05	.06	–	–
Number of children → family satisfaction	−.05	.06	–	–
Number of children → burnout	−.01	.06	–	–
Number of children → work stress	−.04	.07	–	–
With family partner → WFB	−.05	.07
With family partner → work satisfaction	.06	.09	–	–
With family partner → family satisfaction	.07	.07	–	–
With family partner → burnout	−.08	.08	–	–
With family partner → work stress	−.08	.09	–	–
Work support → WFB	.18*	.09	–	–
Work support → work satisfaction	.25**	.09	–	–
Work support → family satisfaction	.02	.08	–	–
Work support → burnout	−.22*	.09	–	–
Work support → work stress	−.27**	.09	–	–

*p < .05, **p < .01, ***p < .001.

BC bootstrap = bias-corrected bootstrap; WFB = work-family balance.

Discussion

Work and family are two inextricably intertwined domains of our life (Aldrich & Cliff, 2003). While over the previous decade there has been mounting evidence that family plays a crucial role in entrepreneurship, less is known about the work-family link of SME owners. Our article aims to contribute to the SME literature by investigating the role of two types of family support in WFB and SWB of SME owners. Drawing on the COR theory, we hypothesized that emotional and instrumental support provided by family members will be perceived as resources, and positively associated with WFB of SME owners. We further hypothesized that by initiating a gain spiral, family support would be indirectly associated with (four measures of) SWB via WFB.

Our results based on the responses of 213 SME owners provide empirical evidence that generally substantiates our hypotheses, with one unexpected finding. Consistent with the COR theory, emotional family support is positively associated with WFB. This provides empirical support to the argument that emotional support provided by family members is perceived by SME owners as a resource, which thereby contributes to their sense of WFB.

However, surprisingly, we found that instrumental support is negatively associated with the WFB of SME owners, albeit with a relatively small effect. This suggests that unlike emotional support, instrumental support provided by family members is not likely to be perceived as a resource by SME owners. This resonates with an emerging literature suggesting that social support is not always perceived as helpful by the support recipient (Brown et al., 2003; Camara, Bacigalupe, & Padilla, 2017; Deelstra et al., 2003). For example, Edelman et al. (2016) found that family financial support was negatively associated with the scope of startup activities of young entrepreneurs.

While the notion that social support may not always be perceived as helpful is not new (Brown et al., 2003), it is unclear when support is perceived as helpful, and when it is not. To address this important issue, Hobfoll et al. (2018) suggest that the COR theory should be integrated with other theories to gain a deeper understanding of this often contradictory phenomena. Hence, to make sense of our unexpected finding, we complemented our existing theoretical framework with the threat to self-esteem model (Fisher, Nadler, & Whitcher-Alagna, 1982).

The ToSE model argues that help is neither all good nor all bad. Instead, help has a “mixture of self-threatening and supportive elements” (Fisher et al., 1982, p. 38). Importantly, the ToSE model proposes that whether help is perceived as self-threating or self-supportive depends on the situational conditions and the personal characteristics of the recipient (Fisher et al., 1982). Below, we briefly discuss how each of these two factors may play a role in the perception of support for the SME owners.

Situational conditions refer to the characteristics of the help, the provider, and the context. Since emotional support typically involves the expression of care and love, it is more likely to be perceived as recognition (that is, resource) rather than threat to the role accomplishments of the SME owners. However, instrumental support that focuses mainly on support with tasks may pose a threat to the perceived role accomplishment and self-esteem of SME owners. For example, having a partner who takes care of all household chores may convey the impression to the SME owners that they are not fulfilling their family role. Furthermore, people tend to feel uneasy when receiving help from their loved ones or a less/equally experienced person (Fisher et al., 1982; Nadler & Fisher, 1986). In other words, instrumental support provided by a less experienced family member, especially in a work-related context, may provoke a sense of inferiority for the SME owner.

Personal characteristics of the recipient generally refers to the self-related concepts and traits of the help recipient. Previous literature suggests that recipients who have a high need for achievement and autonomy, and are ego-involved in their work, are more likely to perceive support as a threat (Nadler, Sheinberg, & Jaffe, 1981; Tessler & Schwartz, 1972). Since these personal characteristics are not uncommon among SME owners (Ladzani & Van Vuuren, 2002; Watson & Newby, 2005), they may be more likely to perceive instrumental support provided by their family members as a threat to their perceived role accomplishment.

Another key objective of this study was to examine the role of WFB among SME owners. Consistent with the COR theory and previous findings of Ferguson et al. (2012), we found that both types of family support are linked indirectly to SWB through WFB. More importantly, the mediating effects of WFB are significant across most indicators of SWB (that is, work satisfaction, family satisfaction, work stress). This shows that supports provided by family member could have important implications for the SWB of SME owners in work and family domains when mediated by WFB. It is worth noting that although the mediation effect of WFB between instrumental support and burnout is not significant in the strict sense, it can be considered as marginally significant with effect size that is comparable to other SWB measures. Therefore, the insignificance of the mediation effect of WFB between instrumental support and burnout should be interpreted with caution and requires further empirical examination. Taken together, our finding extends prior research on this subject to the SME literature by demonstrating the importance of WFB in the relationship between family supports and SWB of SME owners.

A closer examination of the link between the two types of support and SWB reveals three interesting findings. First, we found that emotional support has a relatively strong indirect association with SWB when compared to instrumental support. This provides further support to the notion that emotional support is particularly effective in promoting WFB and, thus, SWB of SME owners.

Second, our exploratory analyses indicate that most of the (direct) associations between support and SWB are not significant except for family satisfaction. One reason may be that the effect of family support has a strong within- rather than cross-domain effect (Amstad, Meier, Fasel, Elfering, & Semmer, 2011). Hence, the influence of family support can be observed in satisfaction only with family, but not with work.

Third, we found that burnout and family satisfaction have a relatively strong association with WFB among all four indicators of SWB. Similarly, the indirect effects of instrumental support and emotional support on SWB are strongest on burnout and family satisfaction. These findings provide further evidence that family support has a relatively strong effect on the nonwork dimension of SWB (Ferguson et al., 2012; Halbesleben, 2006).

Theoretical implications

Our findings have several implications for the WFB and SME literature. First, our results underscore the importance of family support on WFB and SWB of SME owners. While the importance of family has been adequately addressed in the entrepreneurship literature (Aldrich & Cliff, 2003; Rogoff & Heck, 2003), less is known about the role of family in the context of SME owners (Powell & Eddleston, 2013; Tetrick et al., 2000). Recognizing the complex nature of social support (Deelstra et al., 2003; DiMatteo, 2004), we took this issue a step further by examining instrumental and emotional support separately. Our results show that instrumental and emotional support play clear yet differentiated roles in SWB of SME owners via WFB, thus demonstrating the importance of studying SME owners as a unique occupational group. By providing a nuanced understanding of the role of family support in the WFB and SWB of SME owners, our work also contributes to the literature on small business management.

Second, our study enriches the literature on SMEs by bringing in insights from the COR theory. Particularly, we drew on the COR theory to argue that social supports can be seen as resources that promote the SWB of SME owners indirectly through WFB. While there is a large body of research adopting the COR theory within the organization literature (see Halbesleben et al., 2014; Hobfoll et al., 2018), limited research, if any, has applied the COR theory to the SME literature.

More importantly, our unexpected findings related to instrumental support provoked us to find a theoretical explanation for why a seemingly helpful support, that can be categorized as a “resource,” could lead to contradictory outcomes. The insight that the context of help and the recipient’s characteristics could have different implications on the recipient’s internal feelings and self-cognition, which we drew from the ToSE model, helped us to theoretically make sense of our empirical findings. The integration of the COR theory with the ToSE model that we proposed also addresses early calls for research to better understand the “normative evaluation of resources” (Hobfoll, 1989, p. 520) by integrating other theories with the COR theory (see also Hobfoll et al., 2018).

Given the capacity of the COR theory to make a broad range of specific hypotheses (Hobfoll et al., 2018), we believe that as a theoretical framework, it has the potential to offer a wealth of insights to better understand a variety of outcomes in the SME literature. Moreover, the integration of two theoretical approaches – the COR theory and ToSE model – that we propose in this article offers a promising theoretical framework that can be useful for future research to gain a more nuanced understanding of social support.

Third, recent literature suggests that WFB is a “global evaluation of the interplay between work and family” (Wayne et al., 2017, p. 168) that explains additional variance over other work-family constructs (Carlson et al., 2009). Our work lends support to this notion, and contributes to this literature by showing that WFB mediates the relationship between two types of family support and SWB. More importantly, the mediating relationships are significant across both positive (for example, family satisfaction) and negative indicators (for example, work stress) of SWB with the exception of the relationship between instrumental support and burnout.

Fourth, while considerable research has been devoted to the moderating role of social support (that is, stress-buffering hypothesis), less attention has been given to the direct effect of social support on SWB. Our findings provide evidence that social support not only mitigates the detrimental effect of stress but, by itself, can lead to resources gain or loss.

Limitations and future research directions

The present study has several limitations. First, our sample was limited to SME owners in the French construction industry. Thus, we cannot be sure if SME owners from other cultures and industries would yield identical results. Previous studies on the COR theory that examined the value of resources across cultures generally suggest that most, but not all, resources are valued universally (Morelli & Cunningham, 2012; Pines, Ben-Ari, Utasi, & Larson, 2002). Hence, it is likely that our findings can be generalized to at least some other cultures or countries. However, more studies are warranted to empirically examine the generalizability of our findings to other cultures. Furthermore, as we argued in this article, the SME owners are quite unique as an occupational group; thus, it is not clear whether our findings can be generalized to other occupations (for example, salaried worker). Future research may test our model with different occupations and compare the results with our findings.

Second, since we relied on cross-sectional data, causality could not be demonstrated between the variables and the mediation mechanism examined in this study. The cross-sectional nature of the data also limits the possibility of empirically examining the pattern of gain and loss spirals between the proposed variables in our theoretical model. We encourage future study to make use of cross-lagged data to examine the accumulation effect of the loss and gain spirals.

Third, since only survey data of SME owners were used, our study may suffer from common method bias (Podsakoff, Mackenzie, Lee, & Podsakoff, 2003).² Thus, we recommend future studies to replicate our work with data collected from different sources (for example, family members, employees, friends) using different methodologies.

Our findings provide three additional avenues for future research. First, conforming with earlier studies suggesting that social support is a complex construct (Deelstra et al., 2003; DiMatteo, 2004), our findings show that instrumental and emotional support can play differentiated role in WFB and SWB of SME owners. To take our findings a step further, future research could investigate the optimal match between different types of support and specific situations or problems encountered (Cutrona & Russell, 1990). For example, while instrumental support may not be helpful in enhancing the WFB of business owners, it could be useful for promoting the performance of their business. We also suggest future research to replicate our findings and extend our work by investigating the effect of other types of supports (for example, informational support) on the SWB of SME owners.

Second, while the COR theory has been widely adopted in the literature, future research could integrate the COR theory with other theories to provide a fine-grained understanding of the implications of social support on SWB or other outcomes (Halbesleben et al., 2014; Hobfoll et al., 2018). This could be particularly useful when examining various types of social support. Indeed, several recent studies have combined the COR theory with other theoretical models or perspectives to provide a better understanding of the variables of interest (Kiazad, Holtom, Hom, & Newman, 2015; van Woerkom, Bakker, & Nishii, 2016).

Third, future research could explore factors that may moderate the relationship between different types of support and WFB.³ One such avenue for research could be to explore the influence of individual preference or orientation. For instance, the work of Livingston and Judge (2008) shows that individuals with traditional gender role orientation experience more guilt from family to work interferences than individuals with egalitarian gender role orientation.

Practical implications

Our findings suggest that the benefits of family support are not always straightforward. On one hand, emotional support provided by family members can promote the WFB and SWB of SME owners. On the other hand, family members’ support in daily operational tasks may have an adverse effect on the WFB and SWB of SME owners. Thus, family members of business owners need to be aware of the costs and benefits of the different types of support that they can provide.

Our findings also imply that achieving WFB may hold the key to increased family and work satisfaction, and diminished work stress and burnout among SME owners. This has important implications for practitioners and policymakers, as well as mental health professions, who should encourage and support business owners to achieve WFB.

Conclusion

In the present study, drawing on the conservation of resources theory, we examined the association of family support with work family balance and subjective well-being of small and medium enterprise (SME) owners. Our findings show that instrumental and emotional support play a clear, yet differentiated, role in the WFB and SWB of SME owners. Consistent with previous work (Aldrich & Cliff, 2003; Ferguson et al., 2012), our study demonstrates the importance of work-family balance by showing that it mediates the relationship between two vital types of family support and multiple indicators of SWB of SME owners. By doing so, our study not only extends the work-family balance literature to the domain of small and medium enterprises, but also shows – based on theoretical and empirical considerations – that the conservation of resources theory can be integrated with other theories, such as the threat to self-esteem model, to advance the field and our current understanding of the role of resources in the SWB of individuals.

Acknowledgments

We acknowledge the support of Olivier Torrès and Florence Giuliani of the research institute AMAROK in Montpellier for making available and explaining the data to us. We thank Arnold Bakker, Joern Block, Indy Bernoster, and Martin Obschonka for their advice and comments. Roy Thurik is member of the Entrepreneurship and Innovation chair, which is part of Labex Entrepreneurship (University of Montpellier, France) and is funded by the French government (Labex Entreprendre, ANR-10-Labex-11-01).

Notes

¹ Amarok is an organization that conducts research and promotes the physical and mental health of self-employed workers. See http://www.observatoire-amarok.net/fr.

² Four different tests were conducted to examine common method bias. For the Harman’s single factor test, our results showed that the first factor explains only 25 percent of the variance, which is below the 50 percent threshold suggested by Podsakoff and Organ (1986). This suggests that common method bias is not a concern. For the CFA with one single latent factor, our results showed that the single-factor model fits significantly poorer than our eight-factor measurement model (p <.001). This suggests that common method bias is not a concern. For the unmeasured common latent method factor, we found that while the measurement model with the unmeasured common latent method factor fits significantly better than the measurement model without the unmeasured latent method factor (p <.001), the effect of 23 out of 39 indicators on the common latent factor are not significant. Furthermore, the R-square of these indicators have an average improvement of .01 only (from .56 to .57) with a maximum improvement of .1. This does not rule out common method bias entirely. For the CFA with a marker variable, we used body size as the marker. Following Williams, Hartman, and Cavazotte (2010), four models were estimated. A comparison of the Method-C and Method-U Models showed that there is a potential bias (p < .05). However, the Method-R model did not converge, so it is not clear if the effects are biased by the marker variable.

³ It is worth exploring the role of sex since it is positively associated with instrumental support (p <.01) and negatively associated with work-family balance (p <.05). However, it is not significantly associated with emotional support. See Table 1. We found that instrumental support is negatively associated with work-family balance for female SME owners, but not male SME owners. There is no moderating effect for sex on emotional support and work-family balance. One reason could be that female SME owners are more likely to accept the role as a “supporter” and male SME owners are more likely to accept the role as a “receiver” regarding family to business support (Akinola, Martin, & Phillips, 2018; Eddleston & Powell, 2012).

Appendix AList of all translated measurement items

Family emotional and instrumental support (17 items)

King et al. (1995). Family Support Inventory for Workers. Journal of Organizational Behavior, 16, 235–258.

1. Vous souhaiteriez que les membres de votre famille se soucient plus de ce que vous faîtes au travail. (R)

2. Les membres de votre famille coopèrent avec vous pour les tâches domestiques à faire

3. Si votre travail devient très exigeant, un membre de votre famille prendra des responsabilités domestiques supplémentaires.

4. Quelqu’un dans votre famille vous aide à vous sentir mieux quand vous êtes en colère à propos de votre travail.

5. Les membres de votre famille font leur juste part des tâches ménagères.

6. Les membres de la famille vont souvent au-delà de ce qui est normalement prévu pour contribuer à votre succès professionnel.

7. Vous passez trop de votre temps à la maison à ranger après les membres de la famille. (R)

8. Vous pouvez compter sur les membres de votre famille pour vous aider quand vous êtes en retard pour le travail.

9. Quand vous avez une semaine difficile au travail, les membres de votre famille essaient de faire plus de tâches domestiques.

10. Les membres de votre famille vous donnent souvent des commentaires et des idées utiles pour votre travail.

11. Si vous deviez voyager pour votre travail, votre famille aurait du mal à gérer les responsabilités domestiques. (R)

12. Les membres de votre famille sont compatissants quand vous êtes en colère à propos de votre travail.

13. Les membres de votre famille semblent s’ennuyer quand vous parlez de votre travail. (R)

14. Vous avez des difficultés à discuter de vos activités au travail avec les membres de votre famille. (R)

15. Quand quelque chose au travail vous tracasse, les membres de votre famille vous montrent qu’ils comprennent ce que vous ressentez.

16. Vous vous sentez mieux après avoir discuté de vos problèmes professionnels avec un membre de la famille.

17. Les membres de votre famille vous aident souvent avec votre travail.

Work and family balance (6 items)

Carlson et al. (2009). Is work-family balance more than conflict and enrichment?. Human Relations, 62, 1459–1486.

1. Je suis en mesure de négocier et d’accomplir ce qu’on attend de moi au travail et dans ma vie familiale.

2. Je parviens à assumer mon travail et ma vie de famille pour les personnes clé.

3. Les personnes qui me sont proches diraient que je parviens à bien concilier travail et famille.

4. Je suis en mesure d’accomplir les attentes que mes clients et ma famille ont pour moi.

5. Mes collaborateurs et les membres de ma famille peuvent dire que je satisfais leurs attentes.

6. Il est clair pour moi, grâce aux commentaires de mes collègues et des membres de ma famille, que j’accomplis à la fois mes responsabilités professionnelles et familiales.

Family Satisfaction (3 items)

Edwards and Rothbard (1999). Work and family stress and well-being: An examination of person-environment fit in the work and family domains. Organizational Behavior and Human Decision Processes, 77, 85–129.

1. Votre vie de famille est très agréable.

2. Dans l’ensemble, votre vie de famille est genial.

3. Globalement, vous êtes satisfait de ma vie de famille.

Work Satisfaction (3 items)

Cammann et al. (1979). The Michigan organizational assessment questionnaire. Unpublished manuscript, University of Michigan, Ann Arbor.

1. Globalement, vous n’aimez pas votre travail. (R)

2. Dans l’ensemble, vous êtes satisfait de votre travail.

3. Globalement, vous aimez votre travail actuel.

Work stress (4 items)

Motowidlo et al. (1986). Occupational stress: Its causes and consequences for job performance. Journal of Applied Psychology, 71, 618–629.

1. Mon travail est extrêmement stressant.

2. Très peu de choses stressantes m’arrivent au travail. (R)

3. Je ressens beaucoup de stress à cause de mon travail.

4. Je ne me sens presque jamais stressé à cause de mon travail. (R)

Burnout (10 items)

Malach-Pines (2005). The burnout measure, short version. International Journal of Stress Management, 12, 78–88.

1. Vous sentez-vous fatigué

2. Vous sentez-vous déçu par certaines personnes

3. Vous sentez-vous désespéré

4. Vous sentez-vous coincé

5. Vous sentez-vous impuissant

6. Vous sentez-vous déprimé

7. Vous sentez-vous physiquement faible ou malade

8. Vous sentez-vous sans valeur, comme un “échec”

9. Avez-vous des difficultés à dormir

10. Dites-vous “j’en ai marre”

Job support (5 items)

Goldenhar et al. (1998). Stressors and adverse outcomes for female construction workers. Journal of Occupational Health Psychology, 3, 19–32.

A quelle fréquence, vos subordonnés ou collaborateurs

1. font un effort supplémentaire pour rendre votre vie de travail plus facile pour vous?

2. Sont aptes à vous aider lorsque survient une situation difficile au travail?

3. montrent leur soutien lorsque vous avez un problème personnel?

4. s’adaptent quand vous avez un problème personnel?

5. font preuve de compréhension quand vous parlez de vos problèmes personnels?