Journal of Sociology

http://jos.sagepub.com Influences on work/ non-work conflict
Cameron Allan, Rebecca Loudoun and David Peetz Journal of Sociology 2007; 43; 219 DOI: 10.1177/1440783307080104 The online version of this article can be found at: http://jos.sagepub.com/cgi/content/abstract/43/3/219

Published by:
http://www.sagepublications.com

On behalf of:

The Australian Sociological Association

Additional services and information for Journal of Sociology can be found at: Email Alerts: http://jos.sagepub.com/cgi/alerts Subscriptions: http://jos.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.co.uk/journalsPermissions.nav Citations http://jos.sagepub.com/cgi/content/refs/43/3/219

Downloaded from http://jos.sagepub.com by Cristina Cristina on April 12, 2009

Influences on work/ non-work conflict
Cameron Allan, Rebecca Loudoun and David Peetz
Department of Industrial Relations, Griffith University

Abstract
Work/non-work conflict is important because it tells us about the well-being of individuals and more generally of a particular workplace or organization. Important progress has been made in research literature on the importance of structural policies designed to assist workers to meet competing demands to be at work and at home. More information is needed into organizational influences on the emotional aspects of work/non-work conflict. Based on a survey of over 900 employees, we use factor, correlation and multiple regression analyses to find that exacerbation in work/non-work conflict is a result of high workload pressure, long working hours, unsupportive management and weak employee control, especially control over workload and when employees can take time off. Keywords: employee control, work pressure, work/life balance, work/life collision, work/non-work conflict, working time

Ideological, political, economic and social developments have led to changes in the structure of the labour market and the industrial landscape more generally over the last few decades. In turn, these changes have resulted in reforms at the workplace level that have long raised concern amongst individuals, families and researchers. Until recently, governments and private and public sector organizations have shown little concern about the impact of changes at the workplace level on employees. The limited interest that has been shown has focused on relatively objective measures of wage rates, earnings dispersions, employment status changes and institutional protection levels. Recently, however, assessments of workplace change are being made using broader, more subjective terms of outcomes for workers. In particular the
Journal of Sociology © 2007 The Australian Sociological Association, Volume 43(3): 219–239 DOI:10.1177/1440783307080104 www.sagepublications.com

Downloaded from http://jos.sagepub.com by Cristina Cristina on April 12, 2009

220 Journal of Sociology 43(3)

work/home divide is receiving growing attention as a measure of workplace relations although, to date, attention has largely focused upon ‘family-friendly’ workplace changes at the expense of the ‘family unfriendliness’ of other changes (Pocock, 2001). There is a growing body of literature indicating a relationship between work/non-work conflict and diminished physical and psychological health (Duxbury, 2003; Loudoun and Bohle, 1997; National Centre for Epidemiology and Population Health [NCEPH], 2003). Research on work/non-work conflict is also particularly timely given the hyperbole surrounding family-friendly policies in contemporary public debate. Current research on work/non-work conflict in Australia generally concentrates on traditional, formal work/family policies and benefit packages. Researchers have recognized only recently that the nature of jobs and the workplace environment may be the key variable determining workers’ ability to reconcile their work and non-work lives. At present, however, although there are compelling arguments about the influence on work/nonwork conflict of variables that relate to the organizational context, there is limited empirical evidence available to assess their importance (Berg et al., 2003). This article addresses this shortcoming by using data from a largescale survey of male and female employees in Queensland to assess the influence of work pressure, management support and control at work on work/non-work conflict. The article is divided into three sections. The first reviews the relevant literature and establishes the research question in more detail. A detailed discussion of the method used and results found from the study is then presented. The last section discusses the findings.

Perceptions of work/non-work conflict: some findings from the literature
Work/non-work conflict is generally defined in the literature as occurring when the emotional and behavioural demands of work and non-work roles are incompatible, such that participation in one role is made more difficult by virtue of participation in the other (Greenhaus and Beutell, 1985). The main model guiding current research on work/non-work conflict is the ‘spillover’ model (Loscocco and Roschelle, 1991). In this model, a positive relationship is proposed between work and non-work roles to the extent that satisfaction or dissatisfaction in one role spills over into the other (Bond et al., 1998). Studies highlighting the link between work/non-work conflict and fatigue, stress, burnout, psychological well-being, depressed mood and physical symptoms are well documented in the research literature (Barton and Folkard, 1991; Bohle and Tilley, 1989; Duxbury, 2003; Loudoun and Bohle, 1997; NCEPH, 2003). Work/non-work conflict has also been found to influence the health and well-being of workers’ family members such as partners and children. For example Pocock’s (2006) qualitative research

Downloaded from http://jos.sagepub.com by Cristina Cristina on April 12, 2009

Allan et al.: Influences on work/non-work conflict 221

found a link from ‘bad’ jobs and unmet parental time-use preferences to negative spillover in family relationships, while a recent Australian epidemiological study found that children of parents who work consistently long hours or come home stressed were more likely to develop psychological problems and physical illness (NCEPH, 2003). In Canada, Duxbury (2003) found that work/non-work conflict affects workers’ ability to enjoy and nurture their family, resulting in lower levels of family well-being and stability. While ethically these findings alone should be sufficient reason to make work/non-work conflict an important area of investigation, there is another reason researchers are interested in work/non-work conflict. Evidence indicates that policies designed to assist work/non-work conflict can promote employee behaviour that is beneficial to the firm. For example, researchers have found that ‘family-friendly’ policies can result in improvements in return to work after childbirth (Squirchuk and Bourke, 1999), retention rates (Squirchuk and Bourke, 1999), morale and productivity (McCampbell, 1996) and absenteeism (Kossek and Nichol, 1992). Given the strong links found between work/non-work conflict and health of workers and their family members, and the links between work/non-work conflict and organizational performance, it is likely to be an area of growing interest in the future. Indeed, some argue that it is one of the most pressing social problems facing most developed economies (Zetlin and Whitehouse, 1998). Although much is known about the structural causes of work/non-work conflict for specific groups of workers such as mothers with young children, we argue that more discussion is needed about the impact of work/non-work conflict arising from organizational variables on a broader range of workers. These variables include the approach of management, the control exercisable by employees and the pressure of workload. We argue this view for three reasons. First, looking at organizational variables, the bulk of research to date on work/family interactions has been completed by psychologists using job stress as an underpinning framework. This means that much of the research has focused on the individual level, at the expense of considering organizational influences. This shortcoming is paralleled in stress management research where the focus is predominantly on individual mechanisms and strategies people can develop to cope with stress (see Kahn and Byosiere, 1992; Topf, 1992). This leaves the responsibility to cope with work/nonwork firmly upon the individual worker. It suggests that if the worker shows the appropriate amount of commitment then their problems will be overcome. It assumes that individuals can choose their response to stimuli or change the nature of the stimuli by acting on their environment. Unfortunately, this is not always the case. There is strong evidence indicating that individual coping is difficult and relatively ineffective in dealing with complex stressors (Menaghan and Mervis, 1984; Shinn et al., 1984). Indeed, Parker and DeCotiis (1983) concluded that,

Downloaded from http://jos.sagepub.com by Cristina Cristina on April 12, 2009

222 Journal of Sociology 43(3)

at best, individual differences have a mediating effect on reactions to potentially stressful situations and, consequently, individual differences are not the most appropriate perspective from which to study stress in organizations. They argue that in work settings, the organizational perspective deserves more theoretical and empirical attention. In a similar vein, Menaghan and Mervis (1984) argue that collective efforts to solve work-related problems are more effective than relying on individual coping strategies. The lack of information on organizational variables is particularly concerning in Australia, given the trend towards deciding on working conditions at the enterprise level. Roman and Blum (1993) and Quinlan (1993) argue that the responsibilities of organizations to acknowledge that characteristics of the work environment exacerbate problems have not been fully explored or reinforced. If workplaces are expected to decide on work issues, then it is imperative that decisions are made in light of sound empirical knowledge. Second, looking at the sample populations used in research to date, attention has largely focused on formal organizational policies designed to assist particular groups of workers such as women with children. This is not surprising given that there is considerable evidence indicating that work/non-work conflict is most acute for female workers (Charles and Brown, 1981; Gadbois, 1981) as they usually perform an uneven distribution of family and household duties (Gutek et al., 1988; Leslie et al., 1991; Loudoun and Bohle, 1997; Robson and Wedderburn, 1990). However, it is important to know the factors affecting work/non-work conflict for a wider range of workers. For example, more Australian children live in one-parent families than ever before and the majority of children do not have a stayat-home resident parent (ABS, 2000; Buchanan and Thornwaite, 2001), while people with care-giving responsibilities for elderly parents or family members with disabilities are also likely to experience work/non-work conflict. Dempster (2003), in particular, argues that work/non-work conflict should be promoted as an issue that affects workers in an ongoing way at different ages and at different stages of their life. She cautions against singleissue debates that focus on one aspect of family-friendly work arrangements such as paid maternity leave, as this approach narrows discussion about possible workplace strategies to alleviate work/non-work conflict. Third, looking at different forms of work/non-work conflict, most studies to date have focused on structural conflict, which arises from the conflicting demands for time in work and family roles (Voydanoff, 1988). Work and family duties usually cannot be performed simultaneously, a problem that is aggravated for many workers because the increasing span of workers’ hours means that work frequently conflicts with the most valued times for family activities – weekends and evenings (Staines and Pleck, 1984). Given this dilemma, the structural interventions considered in much of the literature to date on work/non-work conflict provide valuable

Downloaded from http://jos.sagepub.com by Cristina Cristina on April 12, 2009

Allan et al.: Influences on work/non-work conflict 223

assistance designed to assist workers to meet their competing demands (Berg et al., 2003). These schemes include paid maternity leave, paternity leave and the option to ‘buy out’ work-time. At the same time, however, there is limited research on emotional interference resulting from time spent ‘recovering’ on rest days (Jackson et al., 1985). The evidence suggests that emotional interference reduces both the quantity and quality of family contact time because workers do not feel capable of participating in family activities (Pisarski et al., 1998). In summary, work/non-work conflict is important because it tells us about the well-being of individuals and, more generally, of a particular workplace or organization. Important progress has been made to date about the importance of structural policies designed to assist workers to meet competing demands to be at work and at home. More research is needed, however, about organizational influences on work/non-work conflict. If these work process variables influence workers’ health, or if the labour market and workplace relations system more broadly allow or even encourage work practices that inhibit the ability of workers to balance their work/non-work lives, then it is an area that should concern both researchers and those involved in policy development. The next section outlines the conceptual model used to examine work/non-work conflict in the present study.

The model of high-demand jobs
In this article we present the results of an analysis of the question: do workers with high-demand jobs experience high work/non-work conflict? There are many ways to measure job demand. There are also many terms used in the literature to categorize job demand, including job strain, high-performance jobs and high-commitment jobs (see for example Berg et al., 2003; D’Souza et al., 2003; Janssen and Nijhuis, 2004). Although it is worthwhile to be aware of these conflicting interpretations, we argue that more progress will be made by focusing on relationships between the variables rather than the terms used to categorize the job dimensions. In this study we draw on Theorell and Karasek’s (2000) influential demand-control model to argue that highdemand jobs can result in lower-quality family interactions as the worker is recovering from time spent at work and thus emotionally unavailable for their family. Consistent with the model, we use four work variables to measure job demands: working hours, management support, employee control and workload pressure. These four variables, measured by self-report, are included in the analyses. This approach was chosen because it is less invasive than so-called ‘objective’ indices of the variables of interest. Although it is important to recognize that data using self-report measures can be influenced by factors such as personality, researchers have found similar results using self-report and

Downloaded from http://jos.sagepub.com by Cristina Cristina on April 12, 2009

224 Journal of Sociology 43(3)

alternative measures (see Sparks et al., 1997 for a review of these studies). The next section briefly considers relationships between the four variables of interest and work/non-work conflict.
Working hours

The amount of time that people spend at work will have a strong influence on work–life balance. The more time spent at work the less time available for participation in non-working life. Long working hours reduces opportunities for socially productive leisure by restricting time available ‘for being an effective marriage partner, parent and citizen’ (Golden and Figart, 2000: 26). Therefore, in our study, we measured respondents’ weekly working hours so we could gauge the effect of hours worked on work–life conflict. Existing research indicates that long weekly hours and involuntary overtime have a negative effect on work–life balance (Berg et al., 2003) as it reduces the quality and quantity of workers’ participation in family and social life (Pocock, 2001; Pocock and Clarke, 2004). People working long hours report lower levels of satisfaction with their hours of work and their work–life balance than other workers (Watson et al., 2003: 87). In this study, we aim to extend these research findings by examining the interaction between work–life balance and employee control over such issues as hours worked, workload pressure and management support.
Workload pressure

Workload pressure encompasses the amount and pace of work. In essence this variable is important because if someone spends more time doing one thing, then they must find that time by spending less on something else or they must do it at a faster pace. Accordingly, we asked questions about the amount of time workers have to rest during breaks, whether they work to tight deadlines, leave on time, take work home, have adequate employees to complete the jobs, have a backlog of work if they are sick and whether working late taken is taken for granted at their workplace. There is considerable research examining these aspects of work pressure using case study research (Allan, 1998; White and Bray, 2003; Willis, 2002), but limited research that draws them together using a large sample. For example Pocock (2003) found that workers who reported work intensification also reported exhaustion, frustration and guilt over their inability to meet parental and spousal expectations. Although such studies provide valuable insight into the changing nature of modern workplaces and the outcomes for families, a focussed, large-scale study is also needed.
Management support

There is considerable research examining the importance of management support for workers but limited research that focuses directly on work/

Downloaded from http://jos.sagepub.com by Cristina Cristina on April 12, 2009

Allan et al.: Influences on work/non-work conflict 225

non-work conflict. Peetz et al. (2003), for example, found that management support is particularly important for counteracting pressures to work longer hours. Their results show that workplaces’ policies to prevent overwork and increased safety risks from extended working hours are ineffective if managers do not support and monitor the policy. In a recent longitudinal study, Janssen and Nijhuis (2004) also highlighted the importance of management support for workers when they found that positive changes in perceived support from managers and co-workers are associated with reductions in emotional exhaustion and fatigue. Research that examines work/non-work conflict more narrowly defined tends to be limited to narrow groups of workers such as shiftworkers. For example Pisarski et al. (1998) extended preliminary work by Loudoun and Bohle (1997), who found a positive relationship between supervisor support and work/non-work conflict amongst nurses. Pisarski et al. (1998) found that supervisor support was most important for allowing workers sufficient control over shifts to diminish structural conflict. Although these studies contribute to our understanding of work/non-work conflict, it is important to see whether the findings extend to workers in different industries and workers on more standard work hours.
Employee control

Consistent with Karasek and Theorell (2000), we use employee control to refer to an employee’s ability to make decisions about how and when they perform their work, as well as the extent to which their job entails using and developing their skills. However, we diverge from Karasek’s and Theorell model in the sense that we have used the more recent approach of asking employees about their perceptions of control at work rather than specific issues over which they were allowed to exercise discretion (Berg et al., 2003; Duxbury, 2003). Accordingly, we asked questions about how much control employees have over the number and timing of hours they worked each week, including overtime, meal breaks and holidays, as well as their workload.

Method
The data reported in this article are drawn from a broader study into the types and effects of work-time change in Queensland. The broader study comprised case studies in 17 organizations and a survey of 15 organizations. The survey research was conducted in a wide range of organizations, including a retail outlet, a theme park, a public utility, a construction firm, a public sector department, two manufacturing firms, a mine, a hospital, a law firm, a community organization, a bank, a repair company, an educational institution and a public-sector enforcement agency. While the sample was not intended to be fully representative of the Queensland population,

Downloaded from http://jos.sagepub.com by Cristina Cristina on April 12, 2009

226 Journal of Sociology 43(3)

it was designed to include organizations in most of the key sectors of Queensland industry. The study was also designed to include a balance of female-dominated, male-dominated and mixed-gender workplaces, and a mixture of strongly, weakly and non-unionized workplaces. While the study included a blend of small, medium and large organizations, the latter were over-represented in the sample. Hence, while the findings are not generalizable to the entire population, they do provide interesting insights into workplace phenomena, particularly in larger, often unionized workplaces where management systems and procedures are well developed. We selected case study organizations in a number of ways. First, we selected some case studies from a database of enterprise agreements involving work-time change in Queensland, commissioned from the Australian Centre for Industrial Relations Research and Training (ACIRRT). Only some of the firms we selected from this database agreed to participate in our study. Second, we approached organizations that were known to be experimenting with work-time change. We found out about these organizations by approaching personal contacts in industry, unions, employers’ bodies and the Queensland Government and asking them to recommend to us organizations they knew to be experimenting with work-time change. We also made contact directly with some organizations at conferences and industry functions. Third, after exhausting these two methods, we examined the composition of our cases and identified that segments of the Queensland economy were under-represented. We then elected to specifically recruit organizations in nominated industry sectors. We directly contacted prominent organizations in these sectors and asked them to participate in our study. The survey was administered between March and May 2002. The study site usually corresponded to either a whole workplace or the entire organization. However, in a small number of cases, a division of an organization was surveyed rather than a single workplace. In one organization, only a particular occupation was surveyed. At study sites with fewer than 200 employees, all workers were surveyed. At study sites with more than 200 employees, a sample of 200 employees was selected using systematic random sampling. In total, 963 usable questionnaires were returned, an overall response rate of 42 per cent. For this particular article, we draw on the responses of a sample of some 886 respondents that includes only those persons who answered all the questions used for this analysis of work/nonwork conflict. The data are unweighted. We provide details about the demographic characteristics of our sample in Table 1. As can be seen, the bulk of respondents were in the age range 20 to 49. Our case study firms included a considerable number of large organizations and several of them were professional organizations – a public sector department, a hospital, a law firm and an educational institution and a public sector enforcement agency. As a result, our

Downloaded from http://jos.sagepub.com by Cristina Cristina on April 12, 2009

Allan et al.: Influences on work/non-work conflict 227

Table 1: Demographic features of the total sample (n = 963) Item Age < = 19 20–29 30–39 40–49 50–59 60+ Occupation Managers Professionals Associate professionals Tradespersons Clerical sales and service Intermediate transport and production Labourers and related workers Employment status Permanent Casual Fixed-term contract Total Gender Men Women Union membership Yes No Total (%) 3 21 31 28 15 2 7 27 14 15 31 4 2 88 7 5 100 50 50 50 50

sample included a high proportion of professionals and associate professionals, and a relatively low proportion of blue-collar workers. This distribution of occupations is likely to have arisen due to case selection, plus the greater preparedness of white-collar workers to participate in research of this type. As can also be seen in Table 1, we captured few casual workers in our study with the overwhelming majority of respondents being employed on a permanent basis. Casual employment comprises about a third of employment in Queensland and, as such, our sample clearly under-represents this important segment of the labour market. Our sample also includes a high proportion of trade union members: about double the national average. Again, this is due to our selection of large organizations, where union density is higher than in smaller workplaces. In addition to the questions about the demographic characteristics of respondents, the survey instrument also elicited respondents’ views about

Downloaded from http://jos.sagepub.com by Cristina Cristina on April 12, 2009

228 Journal of Sociology 43(3)

a range of workplace matters such as work-time arrangements, work/nonwork conflict, perceptions of management, work culture, workload issues and other features of work. These items were measured on two five-point Likert scales and one four-point scale. One of the former ranged from ‘strongly agree’ to ‘strongly disagree’, the second from ‘very satisfied’ to ‘very dissatisfied’. The four-point scale, used for measures of employee influence, ranged from ‘none’ to ‘a great deal’. We used some of these questions to construct scales of work/non-work conflict, employee control, supportive management and workload pressure.

Factor analysis
We used exploratory factor analysis to delineate the four main constructs in this study: work/non-work conflict, employee control, supportive management and workload pressure. As we included the dependent and independent variables in our factor analysis, we expected the constructs to be correlated. Accordingly, we used oblique rotation with principal axis factoring. We commenced the factor analysis using 27 items we considered were components of the main constructs. The Kaiser-MeyerOlkin measure of sampling adequacy was 0.89 indicating that the items were factorable. We checked the sampling adequacy of the individual variables. The initial examination of the scree plot suggested a four- or five-factor solution. The four-factor solution was chosen because of theoretical interpretability and because it had a more clearly defined, simple structure. We eliminated items that loaded at below 0.4. The analysis included some 866 cases although the n was reduced due to missing values. As a cross-check, a separate factor analysis was run using means instead of missing values. The same four-factor solution loading on identical variables was derived, indicating that the missing data, due to missing values, did not affect the outcome of the factor analysis. The four factors accounted for some 51 per cent of the total variance and 43 per cent of the common variance. The mean, standard deviation and Cronbach’s alpha for each factor and factor intercorrelations for each facTable 2: Means, standard deviations, Cronbach’s alpha and factor intercorrelations for main factors Factor 1. 2. 3. 4. Work/non-work conflict Employee control Management support Workload pressure M 3.10 2.37 2.88 2.92 SD 0.86 0.78 0.86 0.82 Alpha 0.65 0.82 0.81 0.84 1 1 2 .217 1 1 3 .456 .251 4 .510 .043 .113 1

Downloaded from http://jos.sagepub.com by Cristina Cristina on April 12, 2009

Table 3: Factor loadings Factorb Work/non-work conflict 1 0.142 −0.176 −0.030 −0.026 −0.015 0.098 0.006 0.020 0.011 −0.058 0.002 −0.027 0.013 0.071 −0.650 0.797 0.643 0.785 0.753 0.710 0.525 -0.083 −0.014 −0.149 −0.028 0.067 0.022 −0.038 0.004 0.031 0.127 −0.149 −0.043 −0.129 0.233 −0.022 0.462 0.589 0.817 0.834 0.696 0.010 0.043 0.000 −0.098 0.123 0.043 0.089 2 3 Employee control Management support Workload pressure 4 0.494 −0.421 −0.614 −0.544 0.127 0.034 −0.082 −0.075 0.080 −0.047 −0.192 0.011 0.150 0.035 −0.034 −0.097 (continued)

Itema

I leave on time most days [reversed] Long hours taken for granted I often take work home If you take time off or get sick, your work just builds up while you’re away

Downloaded from http://jos.sagepub.com by Cristina Cristina on April 12, 2009

Performance targets set by management are reasonable Employees are treated with equal fairness Management can be trusted to tell things the way they are Management tries to cooperate with employees Employees here have enough say if a problem arises with management

[How much say [How much say [How much say [How much say appointments)? [How much say

over] over] over] over]

how many hours you work a week? your starting and finishing times? when you have a meal break? when you take time off (e.g. holidays,

over] your workload?

Allan et al.: Influences on work/non-work conflict 229

I work more hours each week than I would like [Satisfaction with] balance between your work and personal life [reversed]

Table 3: (continued) Factorb Work/non-work conflict 1 −0.560 −0.783 −0.638 −0.015 −0.047 0.017 2 Employee control Manage-ment support 3 0.086 0.031 −0.099 Workload pressure 4 −0.284 −0.076 −0.016

230 Journal of Sociology 43(3)

Itema

I get told at home that I am working too much My work responsibilities interfere with my social life more than they should I am often too tired to properly enjoy my time away from work

Downloaded from http://jos.sagepub.com by Cristina Cristina on April 12, 2009

Note a The wording on some questions has been abbreviated slightly. b Factor 1 = Work/non-work conflict; factor 2 = Employee control; factor 3 = Management support; factor 4 = Workload pressure.

Allan et al.: Influences on work/non-work conflict 231

tor are presented in Table 2. The final items used in the four derived factors are shown in the factor loading table (Table 3).

Regression analysis
To explore the relationship between work/non-work conflict and other variables, we conducted multiple regression. The number of observations for the regression was slightly lower (n = 793) than the factor analysis due to a missing value problem at one of our case study organizations (where, at management’s direction, we were unable to ask the full range of questions). Our principal method of analysis was ordinary least squares (OLS) regression. We devised the following equation:
WORK/NON-WORK CONFLICT = b0 + b1WORKLOAD PRESSURE + b2SUPPORTIVE MANAGEMENT + b3EMPLOYEE CONTROL + b4HOURS WORKED IN WEEK + b5NUMBER OF CHILDREN + b6COHABIT WITH EMPLOYED PARTNER + b7TRADE UNION MEMBERSHIP + b8CASUAL EMPLOYMENT + b9FIXED-TERM CONTRACT + b10SEX + b11AGE + b12OCCUPATION + e

Work/non-work conflict was the dependent variable. The independent variables included the three scales we constructed using factor analysis: workload pressure, supportive management and employee control. Only items with a loading above a threshold of 0.4 (shown in the shaded sections of Table 3) were included in the factors. To control for other characteristics, we included in the equation dummy variables for cohabitation with employed partner, casual employment, fixed-terms contract, age, occupation, sex and trade union membership. We also included in the equation two numeric variables: hours worked per week and number of children. The results of the regression analysis are shown in Table 4. Consistent with our expectations, the results of the regression analysis indicated that work/non-work conflict was negatively correlated with supportive management and employee control, and positively correlated with workload pressure. The results also showed a statistically significant relationship between work/non-work conflict and hours worked and number of children. To check to see if our model had equal salience for men and women we ran the same regression equation separately for men only and women only (results not shown). These two separate equations showed the same statistically significant pattern of association of work/non-work conflict with supportive management, employee control, workload pressure and hours worked. The variable for number of children was significant for women and men.

Downloaded from http://jos.sagepub.com by Cristina Cristina on April 12, 2009

232 Journal of Sociology 43(3)

Table 4: Regression of work/non-work conflict Unstandardized coefficients B (Constant) Workload pressure Supportive management Employee control Hours worked in week Number of children Cohabit with employed partner Trade union status Casual employment Fixed-term contract Sex Age 20–24 Age 25–29 Age 30–34 Age 35–39 Age 40–44 Age 45–49 Age 50–54 Age 55–59 Age 60+ Professionals Assoc. professionals Tradespersons Advanced clerical and service Intermediate clerical, sales and service Intermediate transport and production Elementary clerical, sales and service Labourers and related workers Adjusted R2 = 0.45; N = 793. 3.503 0.462 −0.244 −0.170 −0.018 −0.056 −0.083 −0.023 0.025 0.041 −0.058 0.201 0.227 0.241 0.260 0.245 0.140 0.207 0.204 0.148 −0.073 0.074 −0.070 0.070 −0.097 −0.334 0.131 −0.189 t-stat. 11.870 14.158 −8.065 −5.194 −6.826 −2.234 −1.715 −0.461 0.186 0.390 −1.012 1.211 1.408 1.502 1.596 1.507 0.856 1.247 1.165 0.715 −0.752 0.727 −0.644 0.536 −0.903 −1.911 0.739 −1.087 Sig. 0.000 0.000 0.000 0.000 0.000 0.026 0.087 0.645 0.853 0.697 0.312 0.226 0.160 0.133 0.111 0.132 0.393 0.213 0.244 0.475 0.452 0.467 0.520 0.592 0.367 0.056 0.460 0.277

Discussion
The most important factor influencing work/non-work conflict in Table 4 is workload pressure. When employees are in organizations where working long hours is taken for granted, they do not leave on time, they often take work home and work builds up while they are away, they are likely to show the signs of work/non-work conflict: they are more dissatisfied with the balance between their work and personal lives, are often too tired to properly enjoy their time away from work, get told at home they are working too much, find their work responsibilities interfering with their social life, and would prefer to be working fewer hours. As might be expected, the more hours people work, the more likely they are to experience work/non-work conflict. But hours worked is less important, in

Downloaded from http://jos.sagepub.com by Cristina Cristina on April 12, 2009

Allan et al.: Influences on work/non-work conflict 233

explaining work/non-work conflict, than workload pressure and hence the extent to which the organization expects employees to do more than can be fitted into a working day. Household arrangements are also important. People who have children or cohabit with partners who are working are also more likely to experience work/non-work conflict. Work/non-work conflict is also related to employee control, but on this issue the relationship is more complex. Our index of employee control revealed by the factor analysis regresses significantly against the dependent variable, but when we look inside it we find that some of its five components do not significantly correlate with most of the components of work/non-work conflict. Certainly, employee control over workload, and over when they can take time off, significantly reduces work/non-work conflict. However, employee control over the number of hours worked each week does not significantly correlate with most of the components of work/non-work conflict. Employee control over start and finishing times, and over when they can take meal breaks, generally loses significance after we hold constant employees’ ability to control workload or when they can take leave. Employee control matters for containing the work/non-work conflict, but what particularly matters is the issues over which employees have control. Having some control over working hours, or starting and finishing times, is of little value if employees have no control over their workload or they cannot control when they can take time off for holidays or appointments. Supportive management was at least as important as the above two factors. Workers were less likely to experience work/non-work conflict if management could be trusted, tried to get on with employees, set reasonable performance targets and treated all groups of employees with equal fairness. Our findings indicate the importance of organizational variables in explaining the ability of workers to balance their work and non-working lives. Exploring the nature of the relationship between workers and management is vital to explaining the capacity of workers to achieve a greater work/non-work balance. As Lewis (2001) has observed, while formal organizational policies may have the potential to reduce work/non-work conflict, such initiatives are often undermined by workplace culture. The role of management is critical here. Within limits, managers, for the most part, allocate workload and set working hours. The scope of managerial prerogative also circumscribes workers’ ability to exercise control over aspects of their working lives. Management style in terms of high-trust or low-trust employee relations is also important. Supportive ‘high-commitment’ styles of management are considerably more likely to provide workers with opportunities to achieve some work/non-work balance than low-trust management styles (Evans, 2001). Hence the role of management is critical to the amelioration or attenuation of work/non-work conflict.

Downloaded from http://jos.sagepub.com by Cristina Cristina on April 12, 2009

234 Journal of Sociology 43(3)

Research indicates that management in some organizations is receptive to the work/non-work needs of workers. In some cases, management may promote ‘family-friendly’ policies on equal opportunities grounds to enable men and women to make an optimal contribution in both the working and non-working spheres of their lives. Alternatively, management may seek to reduce work/non-work conflict so as to lower stress and improve worker well-being (Lewis, 1997). The ‘business case’ in support of ‘family-friendly’ policies indicates that higher organizational performance may be achieved due to lower staff turnover, reduced employee stress, attraction and retention of high-quality staff and greater staff satisfaction and commitment (Russell et al., 2000). However, much research shows that management support for policies and practices to reduce work/non-work conflict is limited to a small number of organizations. The proportion of firms adopting cutting-edge policies in this area is very small (Russell et al., 2000). The take-up of such policies tends to be higher in the public sector and within large firms (Evans, 2001). Australian research indicates that the provision of family-friendly policies tends to be restricted to highly skilled staff in managerial and professional roles (Gray and Tudball, 2002). Thus, while management behaviour plays a vital role in reducing work/non-work conflict, few organizations voluntarily adopt such policies. The prospects for improvements in this area appear poor given two recent developments. First, as Evans (2001) observes, recent trends in human resource management (HRM) are not favourable to work/nonwork balance. There has been a shift towards greater devolution of human resource functions, resulting in a weakening of centralized policies and an expansion of line management authority in decision-making on such matters. Clear policies and centralized support is essential for effective implementation of work/non-work balance. Additionally, changing organizational structures associated with growing casualization of the workforce are also a major challenge as an increasingly large proportion of the workforce falls outside the standard employment system. The growth of a long-hours culture in some organizations is also a major family-hostile development (see Campbell, 2002). Second, paralleling these negative HRM trends at the workplace level, there has been a growing deregulatory employment relations environment. The current federal government approach has been to reduce regulation and provide a supposed ‘positive climate’ within which employees and employers can negotiate family-friendly arrangements that best suit their needs (DFCS and DEWR, 2002). However, as Buchanan and Thornthwaite (2001) note, the current policy framework does not work well. Reliance on voluntary arrangement is inadequate because, in the absence of legal provisions, few employers provide satisfactory measures for employees. As a result, many workers miss out on arrangements that would allow them to

Downloaded from http://jos.sagepub.com by Cristina Cristina on April 12, 2009

Allan et al.: Influences on work/non-work conflict 235

better balance their working and non-working lives. As Charlesworth et al. have stated: ‘[l]eaving work and family initiative up to individual firms is insufficient as a basis for public policy that aims at generalizing work/family rights and benefits to all workers’ (2002: 19). While our findings point to the centrality of management in ameliorating work/non-work conflict, the prognosis for any improvements on this front looks grim. Our research highlights the importance of containing workload pressure but the problems of work intensification are widely reported (Watson et al., 2003). Our study points to the negative effects of long hours on work/non-work balance, yet there has been a marked growth in the proportions of Australians working long hours (Campbell, 2002). Our findings indicate that supportive management and increased employee control are important for reducing work/nonwork conflict. However, with the decline of union membership and influence, Australian workers are currently in a deregulated industrial relations environment where the balance of power has shifted markedly in the employers’ favour (Group of 151, 2005). The federal government’s agenda for labour market deregulation appears to be gaining momentum and we can anticipate a reduction rather than enhancement of worker entitlements. Except for workers with labour market status and bargaining power, the prospects for an improvement in work/non-work balance appear slim.

Conclusion
High-strain jobs with low employee control, low support and high workload pressure can result in lower-quality family interactions. Emotional interference reduces both the quantity and quality of family contact time because workers do not feel capable of participating in family activities. Workers are recovering from time spent at work and thus emotionally unavailable for their family. It is not just an issue of long hours – it is also an issue of the stress employees endure at work, and the emotional baggage they bring home. Our research throws some light on this issue. While we need to exercise some caution, as the results of our survey are not representative fully of Australian organizations, the research does highlight that reversing the deterioration in the work–life balance requires employees to have supportive management, the genuine capacity to take time off work, and control over the central source of the problem – long working hours and the workload they endure. Despite the critically important role of management in this process, few organizations actively pursue policies to ameliorate work/non-work conflict for their staff. In the deregulated industrial relations environment, it is likely that the work–life balance of many employees will continue to worsen.

Downloaded from http://jos.sagepub.com by Cristina Cristina on April 12, 2009

236 Journal of Sociology 43(3)

Acknowledgements
We would like to thank the participants in our study. We would also like to acknowledge the assistance and support of our colleagues, Andréa Fox, Chris Houghton, Bob Russell and Keith Townsend. This project was funded by Australian Research Council Grant C00106852 and the Queensland Department of Employment and Industrial Relations.

References
ABS (2000) Labour Force Status and other Characteristics of Families, Catalogue No. 6224.0, June. Canberra: Australian Bureau of Statistics. Allan, C. (1998) ‘The Elasticity of Endurance: Work Intensification and Workplace Flexibility in the Queensland Public Hospital System’, New Zealand Journal of Industrial Relations 23(3): 133–51. Barton, J. and S. Folkard, S. (1991) ‘The Response of Day and Night Nurses to Their Work Schedules’, Journal of Occupational Psychology 64: 207–18. Berg, P., A.L. Kallenberg and E. Appelbaum (2003) ‘Balancing Work and Family: The Role of High Commitment Environments’, Industrial Relations 42(2): 168–88. Bohle, P. and A. Tilley (1989) ‘The Impact of Nightwork on Psychological Wellbeing’, Ergonomics 32: 1089–99. Bond, J., E. Galiski and I. Swanberg (1998) The 1997 National Study of the Changing Workplace. New York: Families and Work Initiative. Buchanan, J. and L. Thornwaite (2001) Paid Work and Parenting: Charting a New Course for Australian Families. Sydney: ACIRRT. Campbell, I. (2002) ‘Extended Working Hours in Australia’, Labour and Industry 13(1): 91–110. Charles, N. and D. Brown (1981) ‘Women, Shiftwork and the Sexual Division of Labour’, Sociological Review 29: 685–704. Charlesworth, S., I. Campbell, B. Probert, J. Allan and L. Morgan (2002) ‘Balancing Work and Family Responsibilities: Policy Implementation Options’, A report for the Victorian Department of Premier and Cabinet (DPC) and Department of Innovation, Industry and Regional Development. Melbourne: CASR. Dempster, F. (2003) ‘Family and Work: Issues and Risk across the Life Course’, paper presented to 8th Australian Institute of Family Studies conference – Steps Forward for Families: Research Policy and Practice, Melbourne, February. DFCS and DEWR (Department of Family and Community Services and Department of Employment and Workplace Relations) with the assistance of the Work and Family Life Consortium (2002) OECD Review of Family Friendly Policies: The Reconciliation of Work and Family Life, Australia’s Background Report. Canberra: Commonwealth of Australia. D’Souza, R.M., L. Strazdins, L.L.-Y. Lim, D.H. Broom and B. Rodgers (2003) ‘Work and Health in a Contemporary Society: Demands, Control and Insecurity’, Journal of Epidemiology and Community Health 57: 849–54. Duxbury, L. (2003) ‘Work–Life Conflict in Canada in the New Millennium’, The Sydney Papers 15(1): 78–97. Evans, J. (2001) Firm’s Contribution to the Reconciliation between Work and Family Life, Labour Market and Social Policy Occasional Papers No. 48. Paris: OECD. Gadbois, C. (1981) ‘Women on Nightwork: Interdependence of Sleep and Off-thejob-Activities’, pp. 223–7 in A. Reinberg, N. Vieux and P. Andlauer (eds) Night and Shiftwork: Biological and Social Aspects. Oxford: Pergamon Press.

Downloaded from http://jos.sagepub.com by Cristina Cristina on April 12, 2009

Allan et al.: Influences on work/non-work conflict 237

Golden, L. and D. Figart (2000) ‘Doing Something about Long Hours’, Challenge 43(6): 15–37. Gray, M. and J. Tudball. (2002) Family-friendly Work Practices: Differences within and between Workplaces, Australian Institute of Family Studies, Research Report No. 7. Melbourne: Australian Institute of Family Studies. Greenhaus, J.H. and N.J. Beutell (1985) ‘Sources of Conflict between Work and Family Roles’, Academy of Management Review 10: 76–88. Group of 151 Australian Industrial Relations, Labour Market, and Legal Academics (2005) Research Evidence about the Effects of the ‘Work Choices’ Bill, Submission to the Inquiry into the Workplace Relations Amendment (Work Choices) Bill 2005, Senate Employment, Education and Workplace Relations Committee, Canberra, November. Gutek, B.A., L. Repetti and D.L. Silver (1988) ‘Non-work Roles and Stress at Work’, pp. 141–76, in C.L. Cooper and R. Payne (eds) Causes, Coping and Consequences of Stress at Work. Toronto: John Wiley and Sons. Jackson, S.E., S. Zedeck and E. Summers (1985) ‘Family Life Disruptions: Effects of Job-induced and Emotional Interference’, Academy of Management Journal 28(3): 574–86. Janssen, N. and F. Nijhuis (2004) ‘Associations between Positive Changes in Perceived Work Characteristics and Change in Fatigue’, Journal of Occupational and Environmental Medicine 46(8): 866–75. Kahn, R.L. and P. Byosiere (1992) ‘Stress in Organisations’, pp. 571–650 in M.D. Dunnette and L. Hough (eds) Handbook of Industrial and Organizational Psychology, 2nd edn. Mountain View, CA: Consulting Psychologists Press. Kossek, E.E. and V. Nichol (1992) ‘The Effects of Onsite Childcare on Employee Attitudes and Performance’, Personnel Psychology 48: 271–88. Leslie, L.A., E.A. Anderson and M.P. Branson (1991) ‘Responsibility for Children: The Role of Gender and Employment’, Journal of Family Issues 12(2): 197–210. Lewis, S. (1997) ‘ “Family Friendly” Employment Policies: A Route to Changing Organizational Culture or Playing About at the Margins?’, Gender, Work and Organization 4(1): 13–23. Lewis, S. (2001) ‘Restructuring Workplace Cultures: The Ultimate Work–Family Challenge?’, Women in Management Review 16(1): 21–9. Loscocco, K.A. and A.R. Roschelle (1991) ‘Influences on the Quality of Work and Nonwork Life: Two Decades in Review’, Journal of Vocational Behavior 39: 182–225. Loudoun, R.J. and P.L. Bohle (1997) ‘Work/non-work Conflict and Health in Shiftwork: Relationship with Marital Status and Social Support’, International Journal of Occupational and Environmental Health 3(suppl. 2): S71–S77. McCampbell, A.S. (1996) ‘Benefits Achieved through Alternative Work Schedules’, Human Resource Planning 19(31): 30–37. Meneghan, E. and E. Mervis (1984) ‘Coping with Occupational Problems: The Limits of Individual Efforts’, Journal of Health and Social Behavior 25: 406–23. National Centre for Epidemiology and Population Health (NCEPH) (2003) Health for Life: Report from the Pilot Study. Canberra: Australian National University. Parker, D.F. and T.A. DeCotiis (1983) ‘Organizational Determinants of Job Stress’, Organizational Behavior and Human Performance 32: 160–77. Peetz, D., K. Townsend, B. Russell, C. Houghton and C. Allan (2003) ‘Race Against Time: Extended Hours in Australia’, Australian Bulletin of Labour 29(2): 126–42. Pisarski, A., P. Bohle and V. Callan (1998) ‘Effects of Coping Strategies, Social Support and Work/non-work Conflict on Shift Worker’s Health’, Scandinavian Journal of Work, Environment and Health 24(suppl. 3): 141–6. Pocock, B. (2001) The Effects of Long Hours on Family and Community Life: A Survey of Existing Literature. Brisbane: Queensland Department of Industrial Relations.

Downloaded from http://jos.sagepub.com by Cristina Cristina on April 12, 2009

238 Journal of Sociology 43(3)

Pocock, B. (2003) The Work/Life Collision. Sydney: The Federation Press. Pocock, B. (2006) The Labour Market Ate My Babies. Sydney: The Federation Press. Pocock B. and Clarke J. (2004) Can’t Buy Me Love? Young Australians’ Views on Parental Work, Time, and Guilt and their own Consumption, Discussion Paper No. 61. Canberra: Australia Institute. Quinlan, M. (1993) ‘The Industrial Relations of Occupational Health and Safety’, pp. 140–70 in M. Quinlan (ed.) Work and Health. Melbourne: Macmillan Education Australia. Robson, M. and A. Wedderburn (1990) ‘Women’s Shiftwork and their Domestic Commitments’, pp. 137–42 in G. Costa, G. Cesana, K. Kogi and A. Wedderburn (eds) Shiftwork: Health, Sleep and Performance. Frankfurt am Main: Peter Lang. Roman, P.M. and T.C. Blum (1993) ‘Work Family Conflict and Employer Responsibility’, pp. 299–326, in R.T. Golembiewski (ed.) Handbook of Organizational Behavior. New York: Mancel Dekker Inc. Russell, G., P. Hwang and L. Hass (2000) ‘Conclusion’, in G. Russell, P. Hwang and G. Hass (eds) Organizational Change and Gender Equity: International Perspectives on Fathers and Mothers at the Workplace. Thousand Oaks, CA: SAGE. Shinn, M., M. Rosario, H. Morch and D.E. Chesnut (1984) ‘Coping with Job Stress and Burnout in the Human Services’, Journal of Personality and Social Psychology 46: 864–76. Sparks, K., C. Cooper, Y. Fried and A. Shirom (1997) ‘The Effects of Hours of Work on Health: A Meta-analytic Review’, Journal of Occupational and Organizational Psychology 70(4): 391–9. Squirchuk, R. and J. Bourke (1999) ‘Gender Equity: From Equal Opportunity in Employment to Family-friendly Policies and Beyond’, pp. 119–34, in L. Hass, P. Hwang and G. Russell (eds) Organisational Change and Gender Equity: International Perspective on Fathers and Mothers at the Workplace. Thousand Oaks, CA: SAGE. Staines, G.L. and J.H. Pleck (1984) ‘Nonstandard Work Schedules and Family Life’, Journal of Applied Psychology 69(3): 315–23. Theorell, T. and R. Karasek (2000) ‘The Demand-control-support Model and CVD’, pp. 78–83 in P.L. Schnall, K. Belkiæ, P. Landsbergis and D. Baker (eds) The Workplace and Cardiovascular Disease, Occupational Medicine: State of the Art Reviews 15(1). Philadelphia, PA: Hanley and Belfus. Topf, M. (1992) ‘Stress Effects of Personal Control over Hospital Noise’, Behavioral Medicine 18(2): 84–94. Voydanoff, P. (1988) ‘Work Role Characteristics, Family Structure Demands, and Work/family Conflict’, Journal of Marriage and the Family 54: 749–61. Watson, I., J. Buchanan, I. Campbell and C. Briggs (2003) Fragmented Futures: New Challenges in Working Life. Sydney: The Federation Press. White, N. and M. Bray (2003) ‘The Changing Role of Nursing Unit Managers: A Case of Work Intensification?’, Labour and Industry 14(2): 1–20. Willis, E. (2002) ‘Enterprise Bargaining and Work Intensification: An Atypical Case Study from the South Australian Public Hospital Sector’, New Zealand Journal of Industrial Relations 27(2): 221–32. Zetlin, D. and G. Whitehouse (1998) ‘Balancing Work and Family Commitments: Developments in Innovative Organisations’, Australian Journal of Early Childhood, 2(3): 9–13.

Downloaded from http://jos.sagepub.com by Cristina Cristina on April 12, 2009

Allan et al.: Influences on work/non-work conflict 239

Biographical note
Cameron Allan is a senior lecturer in the Department of Industrial Relations at the Griffith Business School, Griffith University. His research interests include non-standard employment, working-time, young people at work, management strategy and employment relations in the service industry. Address: Department of Industrial Relations, Griffith University, Nathan, QLD 4111, Australia. [email: c.allan@griffith.edu.au] Rebecca Loudoun is a lecturer in the Department of Industrial Relations at the Griffith Business School, Griffith University. She does research in the areas of enterprise bargaining, health and safety at work, work intensification, working time, employee participation and shift roster scheduling. Address: Department of Industrial Relations, Griffith University, Nathan, QLD 4111, Australia. [email: r.loudoun@griffith.edu.au] David Peetz is a professor in the Department of Industrial Relations at the Griffith Business School, Griffith University. His research interests include unionism, individual contracting, voting and electoral behaviour, Asian employment relations, wages policy, work intensification and labour adjustment. Address: Department of Industrial Relations, Griffith University, Nathan, QLD 4111, Australia. [email: d.peetz@griffith.edu.au]

Downloaded from http://jos.sagepub.com by Cristina Cristina on April 12, 2009

Master your semester with Scribd & The New York Times

Special offer for students: Only $4.99/month.

Master your semester with Scribd & The New York Times

Cancel anytime.