Supervisory support in Slovenian and Dutch organizations: a contextualizing approach

Abstract

This paper builds on the influential work of Suzan Lewis examining how employees’ work–life experiences are shaped by different layers of context. Our approach is therefore a comparative one using data from four organizational contexts in two countries, the Netherlands and Slovenia. Within each organization, we examine the role of different types of supervisory support (specific family support and general support) and the quality of the relationship between supervisor and subordinate (LMX) in work-to-family conflict, enrichment, and work–life balance satisfaction. Findings indicate that in a context with a high level of national policy support, some dimensions of family support are seen as part of general supervisory support. Moreover, the role of supervisory support and the leader–member exchange relationship differs between organizational contexts and the outcomes considered.

Ce document se fonde sur le travail d'influence de Suzan Lewis sur la façon dont les expériences au niveau du travail et la vie des employés sont façonnées par les différentes couches de contexte. Par conséquence nous prenons une perspective comparative en utilisant des données provenant de quatre contextes organisationnels situés dans deux pays; les Pays-Bas et la Slovénie. Au sein de chaque organisation, nous examinons le rôle des différents types de soutien de surveillance (soutien spécifique familial et générale) ainsi que la qualité de la relation entre le superviseur et le subordonné (LMX) sur le conflit travail à famille, l'enrichissement et la satisfaction entre l'équilibre travail-vie. Les résultats indiquent que dans un contexte avec un niveau élevé de soutien de la politique nationale, certains aspects du soutien de la famille sont considérés comme faisant partie de l'appui général de surveillance. En outre, le rôle de supervision et le soutien de la relation d'échange superviseur – subordonné diffère entre les contextes organisationnels et les résultats qui sont considérés.

Keywords:

• supervisory support
• LMX
• work–family conflict
• work–life balance
• work–life policies

Mots-clés:

• soutien de surveillance
• LMX
• conflit travail-famille
• équilibre travail-vie
• politique travail-vie

Introduction

This paper builds on the influential work of Suzan Lewis examining the importance of contextualizing work–life experiences (Lewis, 1997; Lewis, Brannen, & Nilsen, 2009). Her research shows the importance of various layers of context and how they interact and influence each other. Work–life experiences are shaped by the workplace context and its work practices and cultures, the wider social context (public policies, economic factors, national culture, gendered and other norms and expectations of work–life issues), and the broader, rapidly changing global context. As Lewis wrote,

Policies and practices to support the reconciliation of work and family or ‘work-life balance’ in Europe, whether stemming from government regulation or voluntary organizational initiatives, are being implemented at a time when employing organizations are undergoing massive and rapid changes in a context of global competition and efficiency drives. (Lewis et al., 2009, p. 1)

Her work provides evidence of the gap between policy and practice and highlights the paramount importance of organizational culture and informal supervisory support even in countries with extensive social policies intended to help people combine paid work and non-work commitments (e.g. Lewis, 1997, 2003). Cultural assumptions that view long hours and face time in the workplace as a sign of commitment and work devotion may co-exist with flexible work hours, parental leave, and other work–life policies (Lewis, 2003). As argued by Lewis and others, both structural and cultural change is needed for organizations to become supportive (Kossek, Lewis, & Hammer, 2010).

Supervisory support is a critical aspect of organizational culture. Organizational cultures both reflect and shape managerial attitudes and practices (den Dulk & Peper, 2007; den Dulk & de Ruijter, 2005, 2008). Supervisors can enhance or inhibit individuals’ ability to exercise rights and utilize options for a better balance between paid work and private life (Kanjuo Mrčela & Sadar Černigoj, 2011). The crucial role of supervisory support has inspired scholars to study its effects on employees’ work–life experiences and to develop measurements for it (Hammer, Kossek, Zimmerman, & Daniels, 2007, 2009).

This study explores the impact of general and specific family support by supervisors and the quality of the relationship between supervisor and subordinate (LMX) on work-to-family conflict (WFC), work-to-family enrichment (WFE), and overall satisfaction with the way the work–life balance (WLB) is managed. Kossek, Pichler, Bodner, and Hammer (2011) argue in their meta-study on workplace support that specific supervisor behaviors supportive of employees’ WLB are more important than more general supportive behaviors. Their study was conducted in the US and whether their assertion also holds in other cultural contexts remains to be seen. Bagger and Li (2014) argue that the role of informal supervisory support is more salient in a context of few formal policies because there are no other options available. In contexts with more extensive policy support at both the national and organizational level, WLB support may be seen as part of being a supportive supervisor. This study responds to the need for more research in contexts outside the US and other Anglo-Saxon countries. We use the multidimensional concept of supervisory family support developed by Hammer and others (2007, 2009, 2013) to measure specific supervisory support for employees’ WLB, including emotional support, role modeling behaviors, instrumental support, and creative work–family management. General supervisory support refers to general expressions of concern for the well-being and effectiveness of workers (Kossek et al., 2011). In addition, we explore the concept of leader–member-exchange (LMX) (Graen & Uhl-Bien, 1995; Liden, Sparrowe, & Wayne, 1997) and how it relates to the work–life experiences of Dutch and Slovenian public and private-sector employees. LMX is rooted in social exchange theory and refers to the social quality of the relationship between supervisor and employee. Offering family support may be part of this relationship.

Following the work of Lewis, we take a comparative approach by using survey data collected from four organizations: a Dutch university, a Slovenian university, a Dutch consultancy firm, and a Slovenian hospital. The Netherlands and Slovenia are interesting contexts in which to study supervisory support because each one represents a different national policy context compared to the US. Slovenia represents a post-socialist context with a strong tradition of state support for reconciling work and family, although WLB programs and policies are only rarely upgraded at the organizational level. Within the Netherlands, WLB policies are relatively new and more modest, but the regulatory framework is much stronger than in Anglo-Saxon countries like the UK or the US. The presence of compulsory state policies is likely to affect enacted supervisory support. What are the effects of general and specific supervisory support and LMX on employee outcomes such as WFC, WFE, and WLB satisfaction within Dutch and Slovenian organizational contexts?

In the next section, we describe the two national contexts in more detail, followed by a brief discussion of our theoretical framework. In section three, we explain our research method. Our findings are presented in section four and we conclude this paper with a discussion and remarks.

Background and theory

Interest in combining work and family life or – more broadly – in integrating work and private life has grown in both the Netherlands and Slovenia in recent decades.¹ In the Netherlands, workforce diversity is increasing along with the number of people who combine tasks (Merens & Van den Brakel, 2014). Slovenia has a longer tradition of dual-earner families, but employment became less secure during the period of political and economic transition and work intensification increased. In both countries, the economic crisis has intensified this trend (Chung & van Oorschot, 2011; Kanjuo Mrčela & Ignjatović, 2012).

Data from the 5th European Working Conditions Survey (Eurofound, 2012) show that Dutch and Slovenian workers (especially women) spent more hours on paid and unpaid work than workers on average in EU27 (European Foundation, 2012). The same survey shows that the two countries differ considerably when it comes to how satisfied workers are combining their work hours with family or other non-work commitments. While workers in Netherlands are more satisfied on average than workers in EU27 (36% compared to 30% EU27 average), workers in Slovenia rate below the EU average (18%). In addition, only 61% of Slovenian workers report it being easy to take an hour or two off work for personal or family matters, compared to 85% of Dutch workers.

Worker access to formal work–life policies differs in the two countries. Slovenia has a long tradition of state-funded public childcare services and generous parental leaves. Workers have a right to fully paid maternity leave (15 weeks), paternity leave (first 15 days; basic insurance is paid for an additional 75 days), and parental leave (260 days). Parents can choose to work part-time until the child's third year, but this right is rarely exercised. These arrangements have resulted in a high participation rate among women in full-time employment, which did not decline during the transition to the market economy. The dual-earner family model has had little impact on the gendered division of unpaid care, however, with household work remaining highly feminized in Slovenia.

The Dutch national context is characterized by a welfare system based on the male breadwinner model and the traditional distribution of paid and unpaid work. Women are seen primarily as mothers and wives and not as breadwinners (Yerkes, 2009). The participation of women in the Dutch labor market increased in the eighties, mainly in part-time jobs. Part-time work has long been regarded as the main strategy for combining paid work and family life, and the one-and-a-half-earner family model – with the woman working part-time and the man full time – has become the dominant model for couples with children (Merens & Van den Brakel, 2014). In addition, statutory leave periods have been introduced, such as 16 weeks paid maternity leave, 26 weeks unpaid parental leave, and 6 weeks unpaid care leave. In 2000, employees were granted the right to reduce or increase their work hours over their working lives. Moreover, organizations are encouraged to introduce flexible work hours and telecommuting options. The latter are less common in the Slovenian context (Kanjuo Mrčela & Ignjatović, 2006).

Despite the relatively extensive legal entitlements, research indicates that such policies tend to have limited effect when not accompanied by supervisory support (Allen et al., 2014; Thompson, Beauvais, & Lyness, 1999). It is up to the supervisor to communicate, implement, and manage formal policies (Lewis, 2003). In addition, Behson (2005) found that supervisory support explains more variance in work–life conflict than work–life policies. Research has identified supervisors’ support as an important resource for WLB satisfaction (Abendroth & den Dulk, 2011; den Dulk & Peper, 2007) and the reduction of work–family conflict (Anderson, Coffey, & Byerly, 2002; Frye & Breaugh, 2004; Thomas & Ganster, 1995).

Hammer defines supervisory support as a source of social support at work (Hammer, Kossek, Yragui, Bodner, & Hanson, 2009, p. 4). When investigating supervisory support for employee WLB, most studies operationalize this as emotional supervisory support (Abendroth & den Dulk, 2011; Hammer et al., 2009). Supervisors provide emotional support when they show empathy for the employee's WLB situation (Abendroth & den Dulk, 2011) and when employees feel comfortable talking to them about work–life issues (Hammer et al., 2009). According to Hammer et al. (2009), supervisory support is not limited to emotional support. They developed a multidimensional measure of family supportive supervisor behaviors (FSSB) that consists of four dimensions: emotional support, role modeling behaviors, instrumental support, and creative work–family management. As explained earlier, emotional support is when employees feel that their supervisor cares about them and their personal life and when they feel comfortable talking about family and personal commitments. ‘Role modelling behavior refers to supervisors demonstrating how to integrate work and family through modelling behaviors on the job’ (2009, p. 5). Examples include a supervisor who leaves work at six to have dinner with his or her family, or supervisors who mentor employees by sharing ideas or advising them on how to combine work and family life. Instrumental support refers to practical assistance such as responding to the need for flexible work hours or requests to use workplace or statutory work–life policies. Finally, creative work–family management refers to proactive behaviors in which supervisors look for ways to redesign work to help workers balance work and family life. This fourth dimension is based on the dual agenda literature (Bailyn, 2011; Rapoport, Bailyn, Fletcher, & Pruit, 2002), which emphasizes that work can be redesigned to support both workers WLB and the effectiveness of the organization.

In this study we use this multidimensional concept to measure specific family support in the Dutch and Slovenian samples. So far, research indicates that specific support for combining work and family life has a stronger negative correlation to WFC than general supervisory support (Hammer et al., 2009; Kossek et al., 2011). Based on these findings, we would expect specific family support to be more closely related to WFC, WFE, and WLB satisfaction than general supervisory support in our four organizational contexts. However, compared to the US, workers in the Netherlands and Slovenia have more access to formal work–life policies (Gornick & Meyers, 2003; Moss, 2014). This is especially true for public-sector workers. According to Bagger and Li (2014), in a context with limited formal policies like the US, employees are more likely to notice family support provided by their supervisors. Employees in the Netherlands and Slovenia may therefore be less affected by specific work–family supervisory support. We thus explore whether FSSB is indeed more relevant for the work, family, and personal life interface than general support provided by the supervisor.

LMX theory states that leaders and their subordinates have different types of exchange relationships. A distinction is made between high-quality and low-quality relationships. The former are characterized by high levels of mutual trust, respect and liking between supervisors and their employees. Supervisors have high expectations of employees, invest in them, and offer challenging assignments and job autonomy. Employees reciprocate by working hard and sticking with the organization. As employees value their family life, they appreciate family support from their supervisor and in exchange show more commitment and work engagement. Hence, Bagger and Li (2014) contend that supervisory work–family support enhances the quality of LMX. In their study, they found support for this argument, although causality is an issue here. In contrast, other scholars argue that supervisory work–family support is the product of a high-quality exchange relationship. According to Major and Morganson (2011), in a context of high LMX, employees are more likely to utilize existing work–life policies and to receive family support from their supervisor.

Based on social exchange theory, it is possible that LMX and specific and general supervisory support form a positive cycle and positively affect and reinforce each other. We are not able to test this assumption or to shed light on this causality issue. However, both arguments assume that LMX and specific and general supervisory support are distinct constructs, and we can examine that using exploratory factor analyses. In this article, we (a) explore whether LMX, general supervisory support, and specific work–family supervisory support (FSSB) are distinct constructs and (b) analyze the effects of LMX, general and specific supervisory support on WFC, WFE, and WLB satisfaction in Dutch and Slovenian workplaces. To shed light on the importance of organizational context, we analyze each organization separately.

Research design

Sample

We collected data in four organizations: a Slovenian hospital and university and a Dutch university and consultancy firm. Data collection at the two universities and the hospital took place in 2013 using the same questionnaire (Study I). Data collection at the consultancy firm took place in 2014 using a slightly different questionnaire (Study II). Universities are very distinctive work settings. Academic work involves a large measure of independence and flexibility but also a lifelong investment in a career. Bailyn (2006) argues that the unbounded nature of the academic career makes it particularly hard to find a satisfactory WLB and to avoid WFC. The consultancy firm is a Dutch subsidiary of a global financial corporation. The firm is a typical post-industrial, knowledge-intensive company. It mainly employs professionals. Job insecurity has generally been low for these employees, but workloads have always been fairly severe. At the Slovenian hospital, only nurses, midwives, medical technicians, and technical support staff participated in the survey; they typically work in teams and shifts.

In Study I, we used both a digital and a paper version of the Leadership and Work–Life Balance questionnaire to collect data. At the Slovenian university, we invited the employees of three faculties – about 1200 persons in all – to take the survey. A total of 138 respondents filled in the questionnaire: 48.6% from the Faculty of Medicine, 43.4% from the Faculty of Social Sciences, and 8.0% from the Academy of Theatre, Radio, Film, and Television. At the Dutch university, the same questionnaire was distributed among employees of the Faculty of Social Sciences: a total of 94 people responded (response rate of 86%). In the Slovenian hospital, we invited nurses, midwives, medical technicians, and technical support staff to fill in the questionnaire, some 830 employees. A total of 292 respondents completed the questionnaire. In the Dutch consultancy firm, we distributed a larger survey on health and well-being online. The total number of employees is 3500, of which 198 participated in the survey.

Measures

Study I

The survey in Study I contained measurements of FSSB, LMX, and general supervisory support. The leadership variables were all measured using 5-point Likert scales, with 1 indicating very weak support and 5 very strong support for the item statement.

Family Supportive Supervisor Behaviors – FSSB was measured using the short version of the scale reflecting all four dimensions (see Table A1 for the relevant items) (Hammer, Kossek, Bodner, & Crain, 2013).

Leader–member exchange – We used the Liden, Wayne, & Stilwell (1993) 7-item LMX scale – based on Scandura and Graen (1984) – to measure the quality of the relationship between respondents and their leaders.

General Supervisory Support – GSS was measured using a 3-item scale developed by Yoon and Lim (1999). An example of an item is ‘My supervisor can be relied on when things get tough in my job.’

Work-to-family conflict (WFC) – We included the 4-item measurement of WFC by Bond, Thompson, Galinsky, and Prottas (2003), for example, ‘In the past 3 months how often have you not had enough time for your family or other important people in your life because of your job?’ The answer categories ranged from (1) never to (4) always.

Work-to-family enrichment (WFE) – We included the 3-item affect dimension of WFE, based on Carlson, Kacmar, Wayne, & Grzywacz, 2006. A sample item is ‘My involvement in my work puts me in a good mood and this helps me be a better family member.’

Work life balance (WLB) – Satisfaction with work–life balance was measured using the three items of the scale as developed by Valcour (2007). The answer categories ranged from (1) very unsatisfied to (5) very satisfied. A sample item is ‘How satisfied are you with the way you divide your time between work and personal or family life?’

Control variables – We also included commonly used control variables: gender, number of actual work hours, child living at home (yes/no), autonomy (5 items, Karasek & Theorell, 1990), work pressure (a sample item is ‘Does your job require you to work fast?’) (Karasek & Theorell, 1990), and work–life culture (WL culture) (6 items measuring time and career demands, for example ‘To be taken seriously in this organization, employees need to work long days and be available all the time’) (den Dulk & de Ruijter, 2005; Dikkers, Geurts, den Dulk, Peper, & Kompier, 2004).

Study II

The dataset of the consultancy firm contained neither the LMX scale nor the WFC and WFE measures. The analysis focused on FSSB, general supervisory support, and overall WLB satisfaction as a dependent variable (same scale as in Study I).

General supervisory support (GSS) – Measured using the scale derived from Karasek et al. (1998), and with items such as ‘My supervisor is a good coach’ and ‘My supervisor helps me to get the work done.’ GSS is thus measured in a slightly different way than in Study I.

Control variables – The control variables are measured similarly to Study I except for work–life culture, which was measured with a shorter version of the scale used in Study I (three of the six items).

Method

To explore whether LMX, GSS, and FSSB are distinct concepts, we conducted exploratory factor analysis on each of the four datasets using a principle components approach with an oblique rotation. We opted for principal component analysis and oblique rotation because the latter is the favored rotation method when factors are expected to be related (Field, 2005). We conducted multiple regression analyses for each organization to examine how supervisory support and LMX relate to work–life outcomes.

Results

Outcomes exploratory factor analysis

The factor solution for the Dutch university contained three factors (see Table 1). The factors were retained based on the scree plot, the Kaiser's criterion, and the theoretical meaningfulness of the factors (DeVellis, 2003). The items proposed for tapping LMX fell into one dimension. The FSSB items fell under a separate factor, confirming research by Hammer et al. (2009), although factor loading for the item measuring emotional support is not very strong given the sample size (.429). In addition, not all items belonging to GSS fell under a separate factor; one item also loaded on the FSSB factor.

Table 1. Overview of the measurements in Study I and II, Cronbach Alpha.

Scale	Study I	Study II
WFC	.87	n.a.
WFE	.94	n.a.
WFB	.91	.93
WF culture	.89	.76*
FSSB	.90	.85
GSS	.84	.90**
LMX	.92	n.a.
Autonomy	.79	.85
Work pressure	.71	.89

*Including three of the six items.

**Different items than in Study I.

The factor solutions for the Slovenian university and hospital are similar, showing two different factors. The FSSB items loaded on one factor together with the GSS items and one item intended to measure LMX, suggesting that Slovenian respondents in the two organizations do not make an explicit distinction between general support and FSSB. Six of the seven items proposed for tapping LMX fell under a separate factor.

At the Dutch consultancy firm, only items measuring FSSB and general support were included. As noted earlier, the items measuring general support differed from those used in the other three cases. The factor solution indicated that general support and FSSB are distinctive factors in this organizational context (see Table 2).

Table 2. Descriptives.

	Study I				Study II
	Dutch university		Slovenian university		Slovenian hospital		Dutch consultancy firm
Variables	Mean	S.D.	Mean	S.D.	Mean	S.D.	Mean	S.D.
WFC	1.99	.56	2.44	.62	2.38	.58	n.a.	n.a.
WFE	3.42	.81	3.33	.76	3.24	.81	n.a.	n.a.
WLB	3.43	.80	3.33	.91	3.33	.71	3.00	1.01
WL culture	3.43	.77	3.23	.89	2.78	.88	3.61	.76
FSSB	2.98	.79	2.89	.88	3.70	.82	3.05	.78
GSS	3.85	.90	3.54	.94	4.11	.74	3.50	.78
LMX	3.43	.71	3.27	.79	3.89	.63	n.a.	n.a.
Autonomy	2.97	.52	2.49	.61	1.98	.61	2.66	.62
Work pressure	2.59	.53	2.65	.51	2.84	.46	2.83	.64
Gender	.60	.49	.66	.47	.88	.32	.51	.50
Work hours	39.89	8.56	46.14	12.48	46.42	9.12	43.28	8.60
Child at home	.59	.49	.42	.50	.27	.44	.55	.50

Based on the outcomes of the explanatory factor analysis across the four organizations, FSSB and LMX appear to be distinct but related constructs (0.80). The outcomes with respect to general support were less conclusive. Based on these findings, we focus on LMX and FSSB as different constructs for the organizations in Study I. We omit GSS since this scale did not appear to be a separate construct in our exploratory analysis. In Study II, the consultancy firm, we include FSSB and GSS to study the impact of support on work–life outcomes (correlation between the two constructs is 0.63). Before discussing the regression analysis, we show the mean scores on the different variables in the analysis.

The means in Table 2 show that employees of the two Slovenian organizations work more hours on average per week than respondents from the Dutch organizations. Findings also indicate a higher score on WFC in the Slovenian organizations, confirming existing research (e.g. European Foundation, 2012). The highest average mean for WFE is found in the Dutch university sample, although the differences between the organizations are not significant. In contrast, the Dutch consultancy firm rates lowest on WLB satisfaction and has the highest score on work–life culture, indicating a strong emphasis on career and time demands. Regarding GSS and LMX, we find the highest mean scores on all three constructs at the Slovenian hospital. In contrast, we find the lowest mean score on FSSB at the Slovenian university, although that score differs very little from those of the Dutch university and consultancy firm.

Tables 3^{Table 4Table 5}–6 present the results of regression analyses for each organization. For each dependent variable, the first model contains the control variables, the second model includes FSSB, and the third model consists of LMX (Study I) or general support (Study II).

Table 3. Regression Study I: Dutch university.

	WFC			WFE			WLB
	Control	FSSB	LMX	Control	FSSB	LMX	Control	FSSB	LMX
Gender (Female = 1)	−.06	−.06	.05	.07	.03	.11	.03	.02	.03
Work hours	.15	.14	.14	.10	.05	.03	−.17	−.18	.19
Child at home	−.09	−.10	−.11	.03	−.01	−.04	.13	.12	.11
Autonomy	.00	.00	−.01	.01	.02	−.04	.09	.09	.08
Work pressure	.33***	.34***	.34***	.08	.13	.14	−.22*	−.21	−.21
WL culture	.41***	.42***	.41***	.04	.12	.05	−.23*	−.22*	−.23*
FSSB		.06			.42***			.70
LMX			.06			.40**			.09
Adj R^2	.39	.39	.44	−.05	.11	.08	.15	.14	.14
R^2 change		.003	.05		.16	.13		.01	.01

*p < 0.05.

**p < 0.01.

***p < 0.001.

Table 4. Regression Study I: Slovenian university.

	WFC			WFE			WLB
	Control	FSSB	LMX	Control	FSSB	LMX	Control	FSSB	LMX
Gender (Female = 1)	.02	.01	.01	.17	.17	.17	−.05	−.04	−.04
Work hours	−.02	.03	.03	.19*	.20*	.19*	−.12	−.21*	−.21*
Child at home	−.02	−.01	−.02	.02	−.02	−.01	.06	.06	.07
Autonomy	.02	.01	.01	.35***	.33**	.32**	.19*	.21*	.22*
Work pressure	.43***	.41***	.42***	−11	−.11	−.13	−.24*	−.16	.18
WL culture	.29**	.26**	.29**	−.17	−.10	−.11	−.17	−.17	−.19
FSSB		−.07			.17			.09
LMX			.00			.14			.02
Adj R^2	.32	.32	.36	.21	.22	.22	.25	.24	.24
R^2 change		.00	.04		.01	.01		.01	.00

*p < 0.05.

**p < 0.01.

***p < 0.001.

Table 5. Regression Study I: Slovenian hospital.

	WFC			WFE			WLB
	Control	FSSB	LMX	Control	FSSB	LMX	Control	FSSB	LMX
Gender (Female = 1)	.03	.04	.03	.05	.04	.04	−.05	−0.5	−.05
Work hours	.10	.09	.10	.06	.08	.06	−.13*	−.11	−.13*
Child at home	−.09	−.09	−.09	−.00	−.01	−.00	−.02	−.02	−.02
Autonomy	−.04	−.03	−.02	.03	.02	.01	.20**	.18**	.18**
Work pressure	.27***	.26***	.26***	−.23***	−.22**	−.22**	−.15*	−.14*	−.15*
WL culture	.29***	.25***	.26***	−.10	−.05	−.06	−.20**	−.15*	−.17**
FSSB		−.12*			.15*			.17**
LMX			−.12*			.13*			.12
Adj R^2	.24	.25	.25	.06	.07	.07	.14	.16	.15
R^2 change		.01	.00		.01	.01		.02	.01

*p < 0.05.

**p < 0.01.

***p < 0.001.

Table 6. Regression Study II: Dutch consultancy firm.

	WFB
	Control	FSSB	ALGSTEUN
Gender (Female = 1)	−.09	−.11	−.11
Work hours	−.13	−.11	−.13
Child at home	.07	.05	.04
Autonomy	.38***	.31***	.33***
Work pressure	−.38***	−.30***	−.35***
WL culture	−.14*	−.10	−.15*
FSSB		.27***
GSS			.19**
Adj R^2	.44	.49	.47
R^2 change		.05	.03

*p < 0.05.

**p < 0.01.

***p < 0.001.

Work-to-family conflict

With respect to WFC, the results of the two universities show that work pressure and work–life culture increase the degree of perceived conflict. Contrary to our expectations, FSSB and LMX do not show any significant effects. In an organizational culture that emphasizes individual performance, it apparently does not matter whether or not workers feel they have supervisory support, at least not for WFC. That is otherwise at the Slovenian hospital; there we find that FSSB and LMX help to reduce WFC. Similar to the universities, work pressure and work–life culture are important determinants.

Work-to-family enrichment

The picture that emerges for WFE differs at the two universities. FSSB and LMX are important determinants enhancing WFE at the Dutch university. That is not the case at the Slovenian university, where job autonomy and work hours relate positively to WFE. At the hospital, reported work pressure correlates negatively to WFE, while FSSB and LMX have a positive impact.

Work–life balance satisfaction

With respect to WLB, we find no effect for FSSB and LMX in the two university contexts. At the Dutch university, work–life culture appears to be the most important workplace condition. At the Slovenian university, work hours have a negative and job autonomy a positive effect on WLB satisfaction. At the Slovenian hospital, findings indicate that FSSB helps increase WLB satisfaction. Adding FSSB to the model also diminished the negative effect of work hours. The impact of job autonomy, work pressure, and work–life culture remained significant. We found no significant effect for LMX. Regression analysis for the Dutch consultancy firm shows significant effects for both FSSB and GSS, although the impact of FSSB appears to be larger and work–life culture is no longer significant in the model that includes FSSB. The data from the consultancy firm also show relatively large effects for job autonomy and work pressure.

Discussion and concluding remarks

Building on the influential work of Suzan Lewis, this paper examines the role of different types of supervisory support in work-to-family conflict, enrichment, and work–life balance satisfaction across national and organizational contexts. The findings of this exploratory comparative case study indicate that the role of specific family support, general support, and the leader–member exchange relationship quality differs across contexts and the outcomes considered.

Our findings are in line with previous studies that identified specific family support and LMX as different constructs (Bagger & Li, 2014), but they only partly confirm the work of Kossek et al. (2011) suggesting that FSSB and general support are independent constructs. This divergence might be related to differences in context and/or measurement issues. We used two different scales to measure general support and only on the scale derived from Karasek et al. (1998) does it appear to be distinct from FSSB. To our knowledge, however, no previous study has investigated the relationship between the two concepts outside the US. As argued earlier, compared to the US, the Netherlands, and Slovenia offer more formal support for WLB, especially in public-sector organizations (den Dulk, Groeneveld, Ollier-Malaterre, & Valcour, 2013). Consequently, empathy for employee WLB may be more common in the Dutch and Slovenian public-sector organizations in which we collected our data. It may be more normal in that context to express empathy regarding WLB issues and to talk about such issues than in contexts where WLB is seen as a private matter. Case study research in the Netherlands has indicated that even in organizational cultures in which career demands conflict with the take-up of work–life policies, for example leave arrangements or part-time work, employees report feeling comfortable talking to their supervisors about WLB problems (den Dulk & Peper, 2007; Peper, van Doorne-Huiskes, & den Dulk, 2009). Hence, in organizations operating in contexts with strong policy support for WLB, emotional and instrumental support might be seen as part of general supervisory support. Exploratory factor analysis including the full FSSB scale developed by Hammer et al. (2009) showed items measuring FSSB emotional and instrumental support loading together with GSS in some of our case studies (results not shown). More research is needed to draw definitive conclusions about this issue and the four dimensions of the FSSB scale.

Scholars have only now begun to investigate supervisory support for the combination of work and family life and how it relates to LMX, but without using the multidimensional measure proposed by Hammer and others. Moreover, they disagree about the relationship between supervisory support and LMX. While some scholars state that supervisory family support enhances the quality of LMX (Bagger & Li, 2014), others argue that supervisory family support is the product of a high-quality exchange relationship (Major & Morganson, 2011). Based on the social exchange theory, it is possible, however, that LMX and FSSB positively affect and reinforce each other. Longitudinal data are required to shed light on this issue. Nevertheless, both arguments assume that LMX and FSSB are distinct constructs, and our findings thus far concur. However, our study is limited to specific organizations and relatively small samples and future research should elaborate on this issue.

Our findings suggest that supervisory support and LMX do not enhance the work–life experiences of employees in every organizational context. The outcomes for the Slovenian hospital and Dutch consultancy firm confirm previous research showing that, in addition to organizational culture and job demands and resources such as autonomy and work pressure, family support, general support, and LMX influence WFC, WFE, and WLB satisfaction. The results at the two universities were contrary to our expectations. This might be due to the specific organizational context of universities, in which academics have relatively little direct supervision, and in which career expectations appear to be unbounded (Bailyn, 2006).

We end our study by mentioning a number of limitations. Due to data restrictions, we were unable to examine the relative importance of the different types of supervisory support and how they relate to the leader–member exchange relationship. Our study did not include measures of co-worker support, which Lewis’ comparative work shows to be highly relevant (Lewis et al., 2009). Lewis has further shown that qualitative cross-national comparative research can shed light on how different layers of context – workplace, national, and global – shape supervisory support and perspectives (den Dulk et al., 2011; Lewis et al., 2009) and employees’ sense of entitlement to that support (Lewis & Smithson, 2001). For this study, we only had cross-sectional survey data available from four different organizations in two national contexts. We compared a small number of organizations and the response rates were low in some cases. Hence, we are unable to reach any definitive conclusions about cross-national differences and differences across industries.

Nevertheless, following in Lewis’ footsteps, we have taken the first steps towards studying the role of supervisory support and LMX across diverse organizational and policy contexts. More research is required, especially comparative studies. Lewis has devoted her academic career to this kind of research. By coordinating transnational research teams involved in both quantitative and qualitative research, she set the agenda for understanding the impact of work–life policies on the national, organizational, and individual level: a contextualizing approach.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes on contributors

Laura den Dulk is associate professor at Erasmus University Rotterdam, Department of Public Administration and Sociology, the Netherlands. Her main area of expertise is cross-national research on work-life policies in organisations. Current research interests include the integration of work and personal life at the individual level (how women and men manage their work and personal/family life), the organizational level (the role of managers, workplace flexibility), and the country level (cross-national comparisons). She is co-editor of Quality of Life and Work in Europe (2011, Palgrave) and leading author of the paper “National context in work-life research: A multi-level cross-national analysis of the adoption of workplace work-life arrangements in Europe” which won the 2014 Rosabeth Moss Kanter Award for Excellence in Work-Family Research. She is co-editor of Community, Work and Family.

Bram Peper is Lecturer at the Tilburg University, Department of sociology, and Assistant Professor at the Erasmus Law School (criminology). He teaches on sociology, social theory gender, work-life issues, and welfare states. His main research is cross-national research regarding work–family arrangements in organizations in different welfare state regimes. Bram has co-edited of a number of books: Flexible Working, Organizational Change and the Integration of Work and Personal life (Edward Elgar, 2005), and Diversity, Standardization and Social Transformation. Gender, Ethnicity and Equality in Europe (Ashgate, 2011). He co-wrote with Joanne van der Leun and John Macionis a Dutch introduction to sociology (Pearson, 2010; new edition 2014). Currently, Bram's research interests include the well-being of employees, the work–life balance and role of managers, and boundary management in relation to work-life issues.

Aleksandra Kanjuo Mrčela is Professor of Sociology of Work and Economic sociology at the University of Ljubljana, FSS; from 1999–2001 she was a visiting scholar at the London School of Economics and Political Sciences. Since 2004 she is a member of the European Commission Networks of Experts in the field of gender equality. She was Vice-Dean at the Faculty of Social Sciences (2007–2011). She researches and publishes on economic position of women and work-life balance in Slovenia and abroad.

Miroljub Ignjatović (born in Belgrade in 1963) is a Researcher/Assistant Professor at the Faculty of Social Sciences, University of Ljubljana, Slovenia. His main areas of interest are labour market analysis, employment policies, labour market flexibilization, youth unemployment, vocational education and training, and social policy.

Notes

1. In this study we follow Suzan Lewis in using the term work-life instead of work-family. The term work-life is more inclusive in that it covers employees without children or family responsibilities, although the term itself is still debatable (see Gambles, Lewis, & Rapoport, 2006; Lewis & Rapoport, 2005 for a more detailed discussion). However, we continue to use common terms such as work–family conflict and enrichment in line with the existing literature.

Appendix

Table A1. Exploratory factor analysis for Study I.

	Dutch university			SLO university		SLO hospital
	Fac 1	Fac 2	Fac 3	Fac 1	Fac 2	Fac 1	Fac 2
FSSB
Your supervisor demonstrates effective behaviors in how to juggle work and non-work issues	−.155	.836	−.025	.950	−.314	.736	.102
Your supervisor makes you feel comfortable talking to him/her about your conflicts between work and non-work	.203	.429	.375	.752	.119	.656	.280
Your supervisor works effectively with employees to creatively solve conflicts between work and non-work	.236	.710	−.047	.868	.035	.750	.146
Your supervisor organizes the work in your department or unit to jointly benefit employees and the company	.089	.726	.120	.672	.193	.794	.086
LMX
My working relationship with my immediate supervisor is extremely effective	.711	.053	.164	.375	.525	.445	.444
I usually know how satisfied my immediate supervisor is with what I do	.611	.050	.215	.092	.756	−.169	.897
My immediate supervisor would use his or her power to help me solve my problems in my work	.491	.386	.030	.373	.578	.444	.509
I usually feel like I know where I stand with my immediate supervisor	.852	−.029	−.142	.371	.483	.137	.745
My immediate supervisor understands my problems and needs completely	.668	.040	.199	.652	.298	.364	.560
My immediate supervisor recognizes my potential	.854	−.213	.211	−.113	.907	.042	.848
My supervisor would ‘bail me out’ at his/her expense when I really need it	.739	.219	−.184	.314	.529	.294	.570
GSS
My supervisor can be relied on when things get tough on my job	.257	.484	.301	.671	.290	.833	.054
My supervisor is willing to listen to my job-related problems	−.003	.177	.811	.640	.283	.850	.004
My supervisor really does not care about my well-being	.053	−.086	.904	.752	.148	.778	−.193
Eigenvalue	6.89	1.37	1.06	8.53	1.03	8.30	1.19
R^2	0.49	0.09	0.08	0.61	0.07	.59	0.08
N	94	94	94	137	123	292	289

Table A2. Exploratory factor analysis for Study II.

	Dutch consultancy
	Fac 1	Fac 2
FSSB
Your supervisor demonstrates effective behaviors in how to juggle work and non-work issues	−.199	.921
Your supervisor makes you feel comfortable talking to him/her about your conflicts between work and non-work	.359	.606
Your supervisor works effectively with employees to creatively solve conflicts between work and non-work	.176	.788
Your supervisor organizes the work in your department or unit to jointly benefit employees and the company	.181	.700
GSS
I have a good relationship with my immediate supervisor	.817	.007
My immediate supervisor pays attention to what I have to say	.842	.035
My immediate supervisor is a good coach	.871	.036
My immediate supervisor manages to let people work together	.857	.020
My immediate supervisor helps me to get the job done	.819	−.032
Eigenvalue	5.26	1.22
R^2	0.58	0.13
N	198	198