Employee attitudes towards employer-sponsored child care: evidence from France.

Barel, Yvan ; Fremeaux, Sandrine ; Michelson, Grant 等

Introduction

The task of balancing work and personal life is arguably becoming more challenging in many countries as a result of the changing nature of work precipitated by factors including workplace reorganizations, technological develop-ments, and an expansion of spaces and times pertaining to where and when paid work can occur (Lewis et al. 2007). Such challenges are increasingly evident when placed in the context of increasing number of women in labour markets, the rise of dual career couples, higher divorce statistics, the prevalence of single-parent families, and an ageing population. These broader societal trends oblige many employees to seek new solutions for combining work with their personal and family lives. In addition, business firms can also develop proposals intended to help their employees better balance their work and personal lives (Davis & Kalleberg 2006), even if some employers might view them as disruptive (den Dulk & de Ruijter 2008). These employer-based proposals, which to date have generally come under the term of 'family-friendly' practices, may be grouped according to three categories: flexible work policies, paid holidays or vacations, and assistance with child care and family health care.

It is with the last category of practices, specifically the provision of employer-sponsored child care services, that this paper seeks to engage. There might be a number of factors which motivate employers to promote such services including strengthening the image and communication of corporate values, along with the desire to mitigate absenteeism, encourage employee involvement and loyalty, and facilitate future recruitment. Nonetheless, the real impact of company and human resource initiatives in this area is less than evident. Some studies have observed a reduction in turnover and turnover intentions (Milkovich & Gomez 1976, Grover & Crooker 1995) or an increase in satisfaction (Kossek & Nichol 1992, Ezra & Deckman 1996), whereas other researchers did not observe these effects (Miller 1984, Goff et al. 1990). What is even less well known is how employees regard the provision of child care arrangements by their organizations. Because of the ostensible benefits they provide in helping to reduce work-family conflicts, it might be generally assumed that child care is universally desirable and welcomed from an employee perspective. However, this claim has not always been evaluated in empirical terms (for an exception, see Haar & Spell 2004). Therefore, the major contribution of the paper is to investigate the views of employees towards the provision of employer-sponsored child care arrangements.

This study poses the following two major questions. First, is there a relationship between work-family conflict and a positive view of employer-driven child care initiatives? A related question is: what role do organizational and personal situations play in explaining this conflict and do the sources of work-family conflict affect employees' attitudes? Second, what other variables, independent of the work-family conflict, help to inform employees' views of employer-sponsored child care?

In order to better understand the reactions of employees during the development of a work-life balance programme, a survey was conducted at a large shopping centre (involving many different small businesses) in France that had introduced a proposal for the joint financing of a child care centre. The national setting is one of established state legitimacy and intervention in work regulations, including state infrastructure and support for child care and the 35 hour working week. However, such public interventions do not necessarily mean that work-life conflicts are eliminated as the traditional domestic division of labour in France remains strong (Windebank 2001, Crompton & Lyonette 2006) and the situation of fewer work hours is not always improved for those who might work 'unsocial or flexible hours of work in exchange of a reduction of their working time' (Fagnani & Letablier 2004: 551). Thus, it is precisely in such locations as shopping centres where work-family conflicts and balance issues are more likely to be exacerbated because of the nature of operations with their extended days and hours of business.

The research findings show positive but not overwhelming support for the child care proposal. Indeed, there were employees who did not value such arrangements even when their personal situations may have suggested this demand. From this, it is argued that the economic and social repercussions of work-family conflict and work-life balance programmes are quite complex and need to take into account the diversity of organizational and personal situations.

Theoretical Framework & Research Hypotheses

To date, many studies (e.g. Duxbury et al. 1994, Saltzstein et al. 2001, McManus et al. 2002) have confirmed a relationship between a high work-family conflict and a positive reception towards child care proposals. This means that employees who face considerable difficulties in balancing their work and personal lives would view such a proposal in favourable terms. Evidence from previous research further suggests that employees who feel overwhelmed and unable to find the time for their different activities will appreciate any 'kind' and supportive gestures from their supervisor (Anderson et al. 2002, McManus et al. 2002). Thus, a positive response by supervisors or managers towards supporting greater work-life balance among their workforces will be welcomed by employees. The link between a strong work-family conflict and a positive view of the employer-sponsored child care proposal will be explored in the following hypothesis.

Hypothesis 1: The stronger is the work-family conflict, the more positive is the perception of a work-life balance proposal.

To elucidate the role of the work-family conflict concept, our study will measure the effect of the independent variables of the work-family conflict on the attitudes towards an employer-sponsored child care proposal. Frone et al. (1992) define two categories of independent variables: those related to individual and family requirements, and those related to work and organizational requirements. The logic underlying their model is that family requirements can cause a spill-over of the family domain onto the work domain, and thus dissatisfaction at work, whereas work requirements can impinge on family time, and cause dissatisfaction in the family. Building on the findings of Frone et al. (1992), the current study categorizes the independent variables of the work-family conflict by juxtaposing individual and family constraints alongside work and organizational constraints. We explore whether each variable increases the work-family conflict and whether each variable increases employees' positive attitudes towards the child care programme.

Individual & Family Constraints

Past research has shown that work-family conflict is linked to a number of factors including gender, parental status, number and age of children, and family status (Greenhaus & Beutell 1985, McManus et al. 2002). These variables are examined below.

Because women devote on average more time than men to the domestic and educational tasks in the home (Thompson & Walker 1989, Major 1993), it is hardly surprising that they are more likely to complain of lack of time than are men (Greenhaus et al. 1987, Kossek 1990). If work-family programmes strongly affect productivity when women constitute a higher percentage of the work force (Konrad & Mangel 2000), it is underst-andable that women will express a greater work-family conflict and report a particula-rly positive view towards work-life balance proposals.

Many studies also reveal that having children and the number of children increases the work-family conflict (e.g. Kirchmeyer 1995). Moreover, the age and diversity of ages of the children influences the perception of personal constraints (Frone & Yardley 1996). Having children whose ages require different child care methods will increase the work-family conflict considerably (Duxbury & Higgins 2001). It is therefore probable that workers with young children experience the greatest work-family conflict and are particularly receptive to any work-life balance programme. Similarly, single parents reportedly experience a greater work-family conflict than those living with a partner in that they have less help and possibly fewer financial resources (McManus et al. 2002). This might suggest that they will be more favourable towards the balance proposal. This leads to the following hypotheses.

Hypothesis 2a: Women employees, employees with young children or employees who are single parents experience the strongest work-family conflict.

Hypothesis 2b: Women employees, employees with young children or employees who are single parents have a more positive perception of the work-life balance proposal.

Work & Organizational Constraints

Previous studies have revealed that work-family conflict is linked to time allocated to work, consideration of the workers' personal constraints by the supervisor or manager, and predictability of work schedules (Greenhaus & Beutell 1985, Goff et al. 1990, Kossek 1990).

Kossek (1990), for example, asserted that dedicating considerable time to work can increase the difficulties of balancing one's work life and personal life. Other scholars in contrast, notably Evans & Bartolome (1980), demonstrated that perceptions of employees regarding work-family conflict are not connected with the actual number of hours worked. Thus, analyses based on the number of hours worked alone are insufficient. It must be supplemented by assessments of the perceived time allocated to work. Workers who feel that they spend too much time at work may be experiencing the greatest work-family conflict. As a result, they might have a more positive view of any work-life balance proposal.

The findings of Anderson et al. (2002) and McManus et al. (2002) remind us of the importance of support by the direct supervisor or manager. This support entails creation of a climate that favours dialogue, variable work schedules, and tolerance of some personal phone calls (Thomas & Ganster 1995), along with a capacity to listen to employees and offer flexibility to workers in crisis situations (Goff et al. 1990, Warren & Johnson 1995). Research has further shown that support by the immediate supervisor in the form of empathy and flexibility or assistance attenuates the work-family conflict (Thomas & Ganster 1995, Warren & Johnson 1995). We may then infer that the absence of respectful practices in an organization will increase work-family conflict and the positive views towards work-life balance programmes.

Finally, the ease of co-ordination of work hours, along with workers' and their families' use of time, can also reduce work-family conflict (Greenhaus & Beutell 1985, Anderson et al. 2002). We may then ask whether the unpredictability of work schedules will increase the difficulties of balancing an employee's work life and personal life. By implication, this might suggest that employees with unpredictable work schedules are more likely to hold a positive view towards work-life balance programmes than those employees with more predictable work schedules. This discussion leads to the following hypotheses.

Hypothesis 3a: The perception of spending too much time at work, a lack of consideration of personal constraints by the immediate supervisor or manager, and unpredictable work schedules increases the work-family conflict.

Hypothesis 3b: The perception of spending too much time at work, a lack of consideration of personal constraints by the immediate supervisor or manager, and unpredictable work schedules increases the positive perception of the work-life balance proposal.

Other Determinants

Employees' positive views towards a work-life balance proposal do not emerge exclusively from the independent variables of the work-family conflict. The possibility of participating in such a programme and their opinion of the 'ideal' organizational response may also contribute to positive employee attitudes.

Even if employees were not experiencing a strong work-family conflict at the time of the study, they might anticipate benefiting from a work-life balance programme in the short or medium term future. Rather than the direct personal and work-related constraints that employees experienced at the time of the study, it is the possibility of eventually benefiting from the programme that could also explain a positive view towards the employer-sponsored proposal. Just as it is those employees who would benefit immediately from this measure tend to perceive the programme most positively (see Grover 1991), employees who plan to benefit in the future from the child care centre would also be expected to report the most positive response. From this, we derive the following hypothesis.

Hypothesis 4: Employees' positive perception towards a work-life balance programme is positively related to the possibility that they will ultimately benefit from the measure. Employee attitudes are not influenced solely by personal interests. They can also be conditioned by their understanding of the 'ideal' organizational response to the work-family conflict. The successive studies by Kanter (1977), Orthner & Pittman (1986), and Kirchmeyer (1995) propose a classification of organizational responses to work-life conflict that range from 'separation', to 'respect' and 'integration'. An organization applies the principle of 'separation' when it ignores the personal life of its employees. The principle of 'respect' described by Hall & Richter (1998) assumes that a boundary is maintained between one's work life and private life. The employer does not assume the personal responsibilities of the employees on their behalf, but rather allows them to assume their own responsibilities by considering their personal constraints and by allowing flexible work hours, for example. The 'integration' response is when a firm tries to help its employees meet the multiple requirements of work life and private life. The proposed joint financing of a child care centre reflects an integration perspective.

Researchers have examined the efficacy of each of these organizational responses or models (e.g. Hall & Richter 1988, Kirchmeyer 1995). However, we know of no study that has attempted to evaluate the perceptions of employees concerning the 'ideal' organizational response. Using the same typology, our study aims to identify the influence of employee endorsement of the models of 'respect' and 'integration' on their views towards work-life balance programmes. This leads to our final hypothesis.

Hypothesis 5: Employees' positive perception of a work-life balance programme is positively related to employees' support for the models of respect and integration.

Figure 1 summarizes the variables that could explain a favourable response by employees towards the proposed work-life balance programme sponsored by employers.

Method

The study explored the attitudes of employees towards an employer-sponsored work-life programme. The employees were all based in a large shopping centre in Nantes, a city in the west of France with a total population of about 800,000 persons (wider metropolitan area). The shopping centre contained a supermarket and 49 stores, where the centre management was considering the possibility of jointly financing a child care facility with the businesses in the centre. In exchange for partial financing, each participating business could obtain a priority assignment of places in the child care centre for one or more of their employees' children. The proposal was expected to be well received in the shopping centre for at least three reasons :

[FIGURE 1 OMITTED]

1. The context of child care places: If the first organization-provided child care centres were set up in France in the early 1980s, only a few French private firms had taken the initiative to create or jointly finance a childcare facility. In fact, it is very difficult to obtain a place in a state-funded child-care centre because the number of openings is limited but yet such centres remain the preferred means by parents for taking care of their children. Giving the children of employees' priority access can, as a consequence, be seen as considerable assistance even if the fee proposed was not necessarily advantageous.

2. The wide span of hours of the stores: Shop employees' were expected to be available for work from early morning through to late in the evening, including Saturdays and public holidays. There is also pressure for Sunday opening to be authorized (see Baret et al. 1999). This situation increases the need for child care services to be made available beyond the times established by state or public child care providers.

3. The proportion of female employees: The vast majority of employees at the shopping centre were women and this group generally experiences more nonwork-related constraints than do male employees.

The survey questionnaire was administered to the 650 employees of the shopping centre in 2003. During the questionnaire distribution phase, we met all managers of the 49 stores to make them aware of our study. Of those distributed, 311 questionnaires were returned. Of this figure, 300 questionnaires were usable (response rate of 46%). Of the respondents, 78% were female employees; 72% had a full-time, permanent employment contract; 69% were under the age of 36 years; and 74% lived with a partner (married or de facto). In addition, 49% of respondents had children and these employees had an average number of 1.21 children.

Respondents were told that the questionnaire was intended to help researchers understand employees' perception of the work-family balance. It included four sections that covered their work situation, personal situation, perceptions of the work life-personal life balance, and the work-life balance assistance measures.

Employees without children were also encouraged to complete the questionnaire. Although precautions were taken, it is possible that the 311 employees who completed the questionnaire were those who felt particularly concerned by the child care proposal. It is likewise possible that the perception of employees were more positive at the point of the development of the project than at the time of its subsequent implementation. It is often at the implementation stage that employees leave the symbolic sphere of influence and become aware of the concrete difficulties of its application. As the study objective was not to identify the proportion of individuals interested in the proposed child care centre per se, but rather to understand the determinants of the positive views of employees, possible over-representation of employees who were in favour would not bias the validation of the research hypotheses.

In order to discover the existence of statistical relationships between variables, either the Chi2 test for nominal variables or the Fischer test when at least one of the variables was an ordinal variable (scaled), was employed. The p value is the probability the real distribution of the results is random (and thus not due to the influence of the causal variable tested). The relationship is statistically significant when p is less than 0.05 and very significant when it is less than 0.01.

Measures

The measure of work-family conflict does not result from the lack of time and the individual states of fatigue and irritability appearing in the traditional measurement scales of inter-role conflict (Greenhaus & Beutell 1985, Frone et al. 1992). Rather, it is derived from the level of satisfaction in the search for a work life-personal life balance. The formulation of the question is similar to that used in classic studies of the concept of satisfaction in life (Campbell et al. 1976, Rode 2004): 'How satisfied are you with the balance you have achieved between your work life and personal life?' The employees' level of satisfaction was measured on a 4-point scale ('very dissatisfied', 'somewhat dissatisfied', 'somewhat satisfied', and 'very satisfied'), similar to those applied by Ezra and Deckman (1996). The index varied between 1 (strong work-family conflict) and 4 (weak work-family conflict) with a theoretical average of 2.5 corresponding to a neutral opinion.

In order to prompt employees to specifically appraise their possible difficulties of balancing work and non-work, the question related to satisfaction in the quest for the work life-personal life balance was preceded by two questions: 'How satisfied are you with your work life?' and 'How satisfied are you with your personal life?'

Positive perceptions towards the proposed child care facility in the shopping centre were measured using two criteria: perception of normality and employee preferences. First, employees had to reply either 'yes' or 'no' to the question of whether they found it 'normal' that places in a child care centre jointly financed by different organizations be assigned to employees. Second, to create a hierarchy of employee preferences, respondents were asked to specify the services that would benefit them the most. They selected a maximum of three choices from the following services: assignment of a given number of places in a child care centre co-financed by employers; occasional child care assistance service (for example, people that could visit the home if necessary); legal and financial consulting services; messenger services (delivery); laundry services (dry cleaning) and services for seniors. The two criteria used to determine the perception of the proposal are complementary. Analysis of normality is based on employees' judgments of the theoretical 'normality' of the proposal, whereas the analysis of preferences highlights their individual and practical interest in the child care centre.

Satisfaction in the quest for a work life-personal life balance and the perception of the balance assistance proposal were measured by the following personal and family variables: gender (male/female), family situation (single or couple/ married or de facto), the number of children (numerically open variable) and age of children (under three years, between four and six years, over seven years). The following work-related and organizational variables were also evaluated: average number of hours per week spent at work (numerically open variable), judgment of work time (too much time at work; not enough time; just the right amount of time); consideration of personal constraints by the immediate supervisor or manager (never, sometimes, often, always) and predictability of work schedules (never, sometimes, often, always).

To measure the extent that employees believe they could benefit from the programme, two criteria were used. The first question inquired if the employees who were most positive toward the child care proposal had particular socio-demographic characteristics. Second, employees had to answer 'yes' or 'no' to whether they thought they would be interested in placing a dependent in a child care centre over the short or medium term.

Consistent with the terminology used by Kirchmeyer (1995), we considered that employees who endorsed the 'separation' model were those that answered in favour of the item: 'In your opinion, under ideal circumstances should the manager take into account work aspects only (each worker resolves his or her own personal constraints)?' Employees who adhered to the 'respect' model would reply positively to the following item: 'In your opinion, under ideal circumstances should a manager take into account the demands of employees' personal life when planning work schedules (shifts, meetings, paid leave, etc.)?' Finally, employees who endorsed the 'integration' model were deemed to support the item: 'In your opinion, in ideal circumstances should a manager offer services to facilitate balancing the personal life of the employees (child care, legal services, etc.)?' Although employees could be favourable to all three organizational responses, attention will be especially directed towards those employees who supported the models of 'respect' and 'integration'. This stands in contrast to employees who indicated an exclusive interest in the separation model.

Results & Analysis

Of the 300 respondents, we found that 88% considered it 'normal' that places in an employer-sponsored child care proposal be given to employees (rather than the direct public). Moreover, the service preferences expressed by employees indicate that the main expectations concern assistance services for daily child care or for occasional support to care for sick children (Table 1). Along with legal and financial services, these constituted over three-quarters (77%) of employee preferences. The lack of interest in messenger and laundry services can be explained by the fact that the shopping centre already offered practical solutions in both areas.

Written comments on the survey questionnaire confirmed these results despite the fact that a few employees demonstrated reticence with regard to the child care proposal. Some employees expressed feelings of injustice, seeing in the work-family balance proposal a source of discrimination. Among those employees over 50 years of age, several made critical comments: 'In our time, all that didn't exist. Today, everything is done for young people'. If the project is seen as a discriminatory measure, it is because it is not intended to satisfy the needs of all the employees. Only a few employees could obtain priority for their children in the child care centre. Further, one employee remarked that 'not everybody needs child care on the work premises, whereas everybody needs [more] money'.

Table 2 shows the findings do not support hypothesis 1, which posited that the stronger the work-family conflict, the more positively the child care proposal is viewed. On the contrary, workers who adopted a positive view of the child care facility were on average more satisfied in their quest for a work life-personal life balance.

Influence of Individual & Family Constraints

Consistent with previous studies (e.g. Goff et al. 1990, Konrad & Mangel 2000), socio-demographic variables such as gender, parental status, number and age of children, and family status influenced the work-family conflict.

(a) Gender: Women were significantly less satisfied (mean = 2.58) than men (mean = 2.90) with the balance between their work life and personal life (p < 0.05).

(b) Number and age of children: The number of children appears to be very significant in accounting for work-family conflict, because 'very dissatisfied' or 'somewhat dissatisfied' employees have on average 1.51 children, compared with an average of 0.92 children for those that are 'somewhat satisfied' or 'very satisfied' (p < 0.01). Having children whose ages require different child care modes is another explanatory variable of the work-family conflict. Employees who have at least one child under 3 years of age and at least one child over 4 years of age have a level of satisfaction of 2.50, compared with 2.61 for employees who have only one child or several children but these children are all over 4 years of age (p < 0.05).

(c) Family status: Women without children, whether single or not, do not differ significantly in their opinions (p = 0.58). The level of satisfaction is 2.57 in the case of single women and 2.68 for women who are married or live with a partner. Single employees with at least one child clearly express a relatively greater difficulty in balancing their work life and personal life (mean = 2.20) than women who live with a partner (mean = 2.81) (p < 0.01).

Hypothesis 2a, which posited that the work-family conflict is linked to individual and family variables, was therefore supported. The following personal variables--gender, number and age of children and single parent status--all influence the work-family conflict. We will now evaluate whether these variables also shape the way employees perceive the proposed child care centre.

Hypothesis 2b, whereby a positive perception of the work-life balance programme is linked to individual and family variables, also received support: women employees, parents with young children (under 3 years of age), and employees who were single parents all had a more positive perception of the proposal (p < 0.05).

Influence of Work & Organizational Constraints

The number of hours worked per week did not appear to be an important criterion among the surveyed employees (p = 0.53). Employees who were 'somewhat satisfied' or 'very satisfied' with the work-life balance worked about the same number of hours (36.08 hours) per week as those who were 'somewhat dissatisfied' or 'very dissatisfied' with their work-life balance (36.54 hours per week).

It was therefore not the number of hours worked that significantly affected employee satisfaction as Kossek (1990) has previously noted, but rather the personal judgment of the time spent at work (p < 0.01). Employees who believed they spent 'too much time at work' are least satisfied with the balance between their work and private lives (mean = 2.13). However, those who considered that they spent 'not enough time' are hardly more satisfied (mean = 2.25). Satisfaction is significantly greater among employees who reported they spend 'just the right amount of time' at work (mean = 2.82).

The claim that consideration by the direct supervisor or manager of the employees' personal constraints reduces the work-family conflict was supported (p < 0.01). The level of employee satisfaction was only 2.0 when the manager 'never' considers personal constraints. It increases to 2.73 if the manager 'sometimes' considers the employees' personal constraints, to 2.76 when it is 'often' the case and 2.78 when personal constraints are 'always' considered.

The assertion that unpredictability of work schedules increases the work-family conflict was also confirmed (p < 0.01). Employees who felt that their schedules were 'never' or 'sometimes' predictable were less satisfied with the work life-personal life balance (mean = 2.43) than employees who believed their schedules are 'often' or 'always' predictable (mean = 2.76).

Hypothesis 3a, which stated that work-family conflict is related to the following work and organizational variables--judgment of too much time spent at work, absence of consideration of personal constraints by the immediate supervisor, and unpredictable work schedules--was therefore supported. The feeling of being overwhelmed by time spent at work had no significant effect on perceptions of the child care project: 88% of employees said that they spent 'too much time at work' and 89% of those who believed they spent 'just the right amount of time' found the child care joint financing proposal to be 'normal'.

If the lack of consideration of personal constraints increases the work-family conflict, it does not necessarily contribute to a positive attitude towards the child care facility. In fact, the inverse phenomenon was observed. The less employees believed that their managers considered their personal constraints, the less 'normal' they considered the assignment of child care places (Table 3).

Although it increases work-family conflict, work schedule unpredictability also does not significantly contribute to a positive perception of the proposed child care centre (Table 4).

Therefore, hypothesis 3b, which stated that the perception of the work-life balance programme is more positive among employees facing organizational sources of work-family conflict, was not supported. On the contrary, consideration of employees' personal constraints by supervisors and managers increased the positive attitudes towards the proposed child care centre.

Influence of Other Determinants

Two results illustrate the influence of employees' personal plans. First, employees who were reportedly in favour of the child care proposal have particular characteristics: 69% of the employees without children versus 63% of employees who were parents considered the assignment of a given number of places in a child care centre funded by businesses in the shopping centre as one of the services that would help them the most. Under-representation of employees who were parents may be related to the fact that they have already chosen a child care system (for children under 3 years of age) or they no longer need this service (for children over 3 years of age). Employees under 35 years of age who did not yet have children were considerably more likely than the other workers to intend to have a child, and were therefore more interested in the child care centre. Moreover, employees who expressed their intention to enrol their child in a child care centre over the short and medium term also viewed the proposal much more favourably (Table 5).

Hypothesis 4, whereby the positive perception of the work-life balance programme is linked to the possibility that employees will at some later stage benefit, was supported.

Turning to the 'ideal' organizational response to the work-family conflict, at first glance the data indicates a significant preference for the 'respect' model. In other words, the responses were as follows (in ascending order): model of separation (22%); model of integration (60%); and the model of respect (93%). Employees favourable to the logic of 'integration' were also strongly in support of the model of 'respect': all those who supported the 'integration' model attached strong importance to the principle of respect. On the other hand, only 65% of employees who favoured the 'respect' model also subscribed to the integration model. It appears, then, that the model of integration was seen at a supplementary level, and somewhat difficult to obtain, relative to the respect model. The various combinations (and responses) are depicted in Table 6. The 'respect' model alone (or in combination with the 'integration' model) attracted about three-quarters of all responses (74%).

Different visions of the role of the organization led employees to perceive the employer-funded child care centre in different ways (p = 0.01). Specifically, 93% of proponents of the 'integration' model and 92% of employees that endorsed the 'respect' model regarded the allocation of places in a child care centre financed by business to be 'normal', compared with only 69% of employees who embraced the 'separation' model. Thus, hypothesis 5, whereby the positive perception of the work-life balance programme is linked to employee support for the models of 'respect' and 'integration', was supported. The results of the statistical analyses are summarized in Figure 2.

Written comments on survey questionnaires were consistent with the statistical results on two major points. First, employees did not necessarily see how the proposed child care centre would reduce the work-family conflicts that they faced. Thus, the link between the centre and the practical difficulties of reconciling work and personal life was not always evident. For example, employees who had experienced a strong work-family conflict because of the number and diversity in the ages of their children, and who did not plan to have other children, did not necessarily see the utility of such a measure. Second, some employ-ees did not see the value of a programme rooted in an integration logic when they experienced insufficient practices of respect. These initial steps of respect seem to be a necessary prerequisite for the implementation and success of any measure aimed at supporting the 'integration' model.

[FIGURE 2 OMITTED]

Certain employees felt that the child care proposal was superfluous and incoherent, particularly when considered alongside the daily pressure and rigid organization of their work schedules: 'Overtime work isn't paid. Workers are dismissed. But management is going to talk about setting up a child care centre!'. The lack of consideration of personal constraints was seen as a clear obstacle to the work-life balance programme. The most frequently expressed employee frustrations related to meetings that were scheduled by supervisors for late in the day, the absence of flexible work hours and the impossibility of working fewer hours, or to take vacation days on Wednesdays (a day on which many schools in France are closed). There seemed to be a gap between the employer-sponsored child care proposal and the wider context of work methods in the shopping centre which were judged to be too rigid.

Even though the vast majority of the surveyed employees perceived the child care centre as 'normal', a number among them considered as a greater priority the recognition of more autonomy for employees in the organization of their working time. The highly variable and often unpredictable structure of work schedules remains a problem in this industry sector (Baret et al. 1999).

Conclusion

The study has found that if a range of personal constraints can influence work-family conflict, they do not guarantee a positive attitude among employees towards work-life balance programmes (research question 1). Employees' vision of a proposed assignment of child care places in a French shopping centre funded by employers depends not only on individual and family constraints but also on their own personal plans, their perception of the work organization and their concept of the role of the employing organization (research question 2).

It would appear that the success of any employer-driven proposal intended to balance one's work and personal life should meet the following conditions. First, the work-life balance programme must be tailored to the real needs of the employees and the organization. A programme must be designed to directly address the aspirations of the largest number of its employees. Second, the work-life balance programme must be consistent with the organization of work that defends the fundamental rights of employees and takes into account their personal constraints as much as possible. Businesses would only seem to benefit from offering priority places in a child care centre to their employees as part of an integration response if they first emphasize respect, and take into account the day-to-day personal circumstances of their employees.

In other words, the 'respect' model is perceived as a consensual solution, whereas the model of 'integration' is considered as somewhat more difficult to achieve. This conclusion is justified because implementing a work-life balance proposal as part of an integration response may exacerbate feelings of inequity and discrimination among some employees who will not benefit from the measure. In future empirical studies, it would be worthwhile to pursue the investigation using the same methodology on already established child care centres (provided by the employer) and on less well targeted work-family balance programmes such as services for the elderly, for example. We also see potential in doing comparative research both in terms of reactions towards employer-funded and government-funded child care centres, as well as studies that extend across different national settings (Zeytinoglu et al. 2010).

Since work-life balance programmes can sometimes evoke diverse reactions for reasons that are not always easy to identify, it is important that human resource and other managers not assume that measures that purportedly favour an improved work-life balance will have a uniquely positive impact--both economically and socially. Only once a 'respect' approach is applied--especially by an employee's immediate supervisor--is it then opportune to advance further towards adopting the 'integration' model. Until this occurs, it would certainly be prudent for employers to limit communication of any new initiatives that are still tentative as to do otherwise might create expectations that will prove difficult to satisfy.

References

Anderson, S., Coffey B. & Byerly, R. (2002), "Formal Organizational Initiatives and Informal Workplace Practices: Links to Work-Family Conflict and Job-Related Outcomes", Journal of Management, 28(6): 787-810.

Baret, C., Gadrey, J. & Gallouj, C. (1999), "France, Germany, Great Britain: The Organization of Working Time in Large Retail Food Stores", European Journal of Industrial Relations, 5(1): 27-48.

Campbell, A., Converse, P. & Rodgers, W. (1976), The Quality of American Life: Perceptions, Evaluations and Satisfaction. New York: Sage.

Crompton, R. & Lyonette, C. (2006), "Work-Life 'Balance' in Europe", Acta Sociologica, 49(4): 379-93.

Davis, A. & Kalleberg, A. (2006), "Family-Friendly Organizations? Work and Family Programs in the 1990s", Work and Occupations, 33(2): 191-223.

Den Dulk, L. & de Ruijter, J. (2008), "Managing Work-Life Policies: Disruption Versus Dependency Arguments. Explaining Managerial Attitudes Towards Employee Utilization of Work-Life Policies", International Journal of Human Resource Management, 19(7): 1222-36.

Duxbury, L. & Higgins, C. (2001), Work-Life Balance in the New Millennium: Where are We? Where Do we Need to Go? Survey financed by Health Canada.

Duxbury, L., Higgins, C. & Lee, C. (1994), "Work-Family Conflict: A Comparison by Gen der, Family Type and Perceived Control", Journal of Family Issues, 15: 449-66.

Evans, P. & Bartolome, F. (1980), "The Relationship Between Professional and Private Life", in C. Derr (Ed.), Work, Family and Career (281-99), New York: Praeger.

Ezra, M. & Deckman, M. (1996), "Balancing Work and Family Responsibilities: Flextime and Child Care in the Federal Government", Public Administration Review, 56(2): 174-79.

Fagnani, J. & Letablier, M.-T. (2004), "Work and Family Life Balance: The Impact of the 35Hour Laws in France", Work, Employment and Society, 18(3): 551-72.

Frone, M., Russell, M. & Cooper, M. (1992), "Antecedents and Outcomes of Work-Family Conflict: Testing a Model of the Work-Family Interface", Journal of Applied Psychology, 77(1): 65-78.

Frone, M. & Yardley, J. (1996), "Workplace Family-Supportive Programmes: Predictors of Employed Parents' Importance Ratings", Journal of Occupational and Organizational Psychology, 69(4): 351-66.

Goff, S., Mount, M. & Jamison, R. (1990), "Employer Supported Child Care, Work/Family Conflict, and Absenteeism: A Field Study", Personnel Psychology, 43(4): 793-809.

Greenhaus, J., Bedeian, A. & Mossholder, K. (1987), "Work Experiences and Job Performance and Feelings of Personal and Family Well-Being", Journal of Vocational Behavior, 31: 200-15.

Greenhaus, J. & Beutell, N. (1985), "Sources of Conflict Between Work and Family Roles", Academy of Management Review, 10(1): 7688.

Grover, S. (1991), "Predicting the Perceived Fairness of Parental Leave Policies", Journal of Applied Psychology, 76(2): 247-55.

Grover, S. & Crooker, K. (1995), "Who Appreciates Family-Responsive Human Resource Policies? The Impact of Family-Friendly Policies on the Organizational Attachment of Parents and Non-Parents", Personnel Psychology, 48(2): 271-88.

Haar, J. & Spell, C. (2004), "Programme Knowledge and Value of Work-Family Practices and Organizational Commitment", International Journal of Human Resource Management, 15(6): 1040-55.

Hall, D. & Richter, J. (1998), "Balancing Work Life and Home Life: What can Organizations do to Help?", Academy of Management Executive, 2(3): 213-23.

Kanter, R. (1977), Work and Family in the United States: A Critical Review and Agenda for Research and Policy, New York: Russell Sage.

Kirchmeyer, C. (1995), "Managing the Work-Nonwork Boundary: An Assessment of Organizational Responses", Human Relations, 48(5): 515-36.

Konrad, A. & Mangel, R. (2000), "The Impact of Work-Life Programs on Firm Productivity", Strategic Management Journal, 21(12): 1225-37.

Kossek, E. (1990), "Diversity in Child Care Assistance Needs: Employee Problems, Preferences, and Work-Related Outcomes", Personnel Psychology, 43(4): 769-91.

Kossek, E. & Nichol, V. (1992), "The Effects of On-Site Child Care on Employee Attitudes and Performance", Personnel Psychology, 45(3): 485-509.

Lewis, S., Gambles, R. & Rapoport, R. (2007), "The Constraints of a 'Work-Life Balance' Approach: An International Perspective", International Journal of Human Resource Management, 18(3): 360-73.

Major, B. (1993), "Gender, Entitlement and the Distribution of Family Labor", Journal of Social Issues, 49: 141-59.

McManus, K., Korabik, K., Rosin, H. & Kelloway, K. (2002), "Employed Mothers and the Work-Family Interface: Does Family Structure Matter?", Human Relations, 55(11): 1295-1311.

Milkovich, G. & Gomez, L. (1976), "Day Care and Selected Employee Work Behaviors", Academy of Management Journal, 19(1): 111-15.

Miller, T. (1984), "The Effects of Employer-Supported Child Care on Employee Absenteeism, Turnover, Productivity, Recruitment, or Job Satisfaction: What is Claimed and What is Known", Personnel Psychology, 37(2): 277-89.

Orthner, D. & Pittman, J. (1986), "Family Contributions to Work Commitment", Journal of Marriage and the Family, 48: 573-81.

Rode, J. (2004), "Job Satisfaction and Life Satisfaction Revisited: A Longitudinal Test of an Integrated Model", Human Relations, 57(9): 1205-30.

Saltzstein, A., Ting, Y. & Saltzstein, G. (2001), "Work-Family Balance and Job Satisfaction: The Impact of Family-Friendly Policies on Attitudes of Federal Government Employees", Public Administration Review, 61(4): 452-67.

Thomas, L. & Ganster, D. (1995), "Impact of Family-Supportive Work Variables on Work-Family Conflict and Strain: A Control Perspective", Journal of Applied Psychology, 80(1): 6-15.

Thompson, L. & Walker, A. (1989), "Gender in Families: Women and Men in Marriage, Work and Parenthood", Journal of Marriage and the Family, 51: 845-71.

Warren, J. & Johnson, P. (1995), "The Impact of Workplace Support on Work-Family Role Strain", Family Relations, 44: 163-69.

Windebank. J. (2001), "Dual-Earner Couples in Britain and France: Gender Divisions of Domestic Labour and Parenting Work in Different Welfare States", Work, Employment and Society, 15(2): 269-90.

Zeytinoglu, I., Cooke, G. & Mann, S. (2010), "Employer Offered Family Support Programs, Gender and Voluntary and Involuntary Part-Time Work", Relations Industrielles, 65(2): 177-95.

Yvan Barel is Associate Professor, University of Nantes, France. Email: yvan.barel@univ-nantes.fr. Sandrine Fremeaux is Associate Professor of Law, Audencia Nantes School of Management, France. Email: sfremeaux@audencia.com. Grant Michelson is Professor of HRM and Director of Research, Audencia Nantes School of Management, France. Email: gmichelson@audencia.com

Table 1: Preferences in Child Care Assist

Preferences                 Frequency (%)

Assignment of a number of              30
places in a child care
centre financed by
employers

Occasional child care                  28
services (e.g., people
that could visit the home
if necessary)

Legal and financial                    19
consulting services

Messenger services                     11
(delivery)

Laundry services (dry                   8
cleaning)

Services to help seniors                4

TOTAL                                 100

Table 2: Impact of Work-family Conflict on Perception of the Child
Care Proposal

It is normal for places     Degree of satisfaction in
in a child care centre      the quest for work life--
financed by organizations   personal life balance
be assigned to employees

           Yes                        2.68
           No                         2.44
          TOTAL                       2.65

(p < 0.05)

Table 3: Impact of Consideration of Personal Constraints on
Perception of the Child care Proposal

It is normal for places     Degree of satisfaction in
in a child care centre      the questfor work life--
financed by organizations   personal life balance
be assigned to employees

           YES                        2.58
           No                         1.89

(p < 0.01)

Table 4: Impact of Predictability of Work Schedules on Perception
of the Child Care Proposal

It is normal for places     Degree of satisfaction in
in a child care centre      the quest for work life--
financed by organizations     personal life balance
be assigned to employees

           Yes                        2.86
           No                         2.67
          TOTAL                       2.84

(p < 0.21)

Table 5: Impact of the Possibility of Benefiting on
Attitudes towards the Child care Proposal

                            The assignment of places in a
                            child care centre financed
                            by business is normal

                                    Yes   No    TOTAL

Respondent would be         Yes     95%   5%    100%
interested in enrolling a
child in a child care       No      79%   21%   100%
centre over the short or

medium term                 TOTAL   88%   12%   100%

(p < 0.01)

Table 6: Combinations of Organizational Responses

              Separation   Integration      Respect model   Number

              yes model    model no or NR   no or NR         24

Single        no or NR     yes              no or NR          0
combination   no or NR     no or NR         yes              68

Double        no or NR     yes              yes             152
combination   yes          no or NR         yes              24
              yes          yes              no or NR          0

Triple        yes          yes              yes              16
combination   no or NR     no or NR         no or NR         16
                           TOTAL                            300

              Frequency

                8%

Single          0%
combination    23%

Double         51%
combination     8%
                0%

Triple          5%
combination     5%
              100%

NR = No Response