Influencing Factors for Developing Managerial Behaviours That Encourage a Work-Family Culture in the University Context

This article develops and tests a theoretical model to find out which factors influence the behaviour of supervisors in terms of promoting a work-family culture. This model explains to what extent the factors studied are relevant to encourage deans to promote this type of culture at Spanish universities. The hypotheses were tested using linear regression analysis. Data were obtained through a questionnaire to deans. The results yield five key factors: (1) the personal work-family conflict of managers; (2) the transformational leadership style of managers; (3) the identification with subordinates in need of work-family cares; (4) the perceived institutional support; and (5) the perceived support from other supervisors in the centre. The findings have practical implications for human resources management (HRM) practices. Human resources management practices such as (a) providing deans and other supervisors with training about the importance of work-family programs; (b) promoting deans’ training in order to develop transformational leadership skills; or (c) increasing institutional support can be useful when implementing a work-family culture in Spanish universities.


Introduction
The increased presence of women in the labour market, as well as the growing number of one-parent families and couples in which both members work, are just some of the factors that have contributed to the increase in the interaction between work and family roles.
When one role hinders or is incompatible with the other, spillover theory proposes that the work-family interaction may lead to different forms of conflict: time-based conflict, strain-based conflict and behaviour-based conflict [1].Research evidence suggests that the resulting work-family conflict has deleterious consequences not only for performance of employees but also for their well-being [2].Consequently, the adoption of family-friendly programs (FFPs) to assist employees to better manage their work and family responsibilities has been one of the most interesting topics for researchers on the human resource management field during the last years.
Empirical studies on the benefits of FFPs do not always concur in their conclusions [3].Whereas some associate these programs with different beneficial behaviours and outcomes for organisations [4], other studies find no relationship between these factors and even show evidence of a negative relationship in terms of attracting and retaining employees, reducing stress and improving productivity [5].As a consequence, conclusions on the relationship between the availability of FFPs and work attitudes and behaviours are ambiguous [6].
It seems that the availability of FFPs alone does not guarantee their implementation nor the creation of a work-family culture.First, some employees are reluctant to use existing work-family programs, even when they need do so, due to negative perceptions of the family policy of companies [7].Second, many employees perceive that by using FFPs they are communicating that they are neither interested in career advancement nor committed to the firm [8].
As a consequence, some scholars state the availability of FFPs is not effective enough to address employee's concerns regarding work and family issues unless they are accompanied by an informal organizational support such as a work-family culture [9].Some studies show that the implementation of a work-family culture increases the effectiveness of FFPs [10,11].
Research on the association between both FFPs availability and work-family culture with employee outcomes reveal that the influence of work-family culture is stronger than the impact of FFP availability [12,13].
Work-family culture issues have become a relevant topic for managers and researchers (see, e.g., [12] or [14]) existing an urgent need to gain additional understanding about this phenomenon [15].In particular, some researchers have highlighted the need to develop family-supportive supervisors [16,17].
Supervisors play a key role in order to effectively implement FFPs and as change agents to develop a family-supportive culture [14].They often: "(a) have final approval as to whether employee can use a program; (b) lead in the creation of norms supporting the use of policies; (c) influence whether employee are cross-trained to back each other during absences; and (d) affect whether polices are well publicized" ( [7], p. 10).
However, one shortcoming of the research is that there is a lack of knowledge on factors and conditions that lead supervisors to exhibit family supportive behaviours [16,18].Although some researchers have developed conceptual models to analyse the influencing factors that encourage supervisors to develop behaviours to promote a work-family culture [14] they have not been tested empirically yet.
Moreover, most of the studies analysing work-family balance focus on the business sector in the English-speaking world and Northern Europe countries.Since cultural differences between employees from developing and developed countries could play an important role in understanding the factors that lead supervisors to exhibit family supportive behaviours [19], some authors have remarked on the need for studies on work-family issues in other cultural contexts [6].In particular, researchers have also noted the scarcity of studies concerning work-family culture issues in Southern European and Latin culture countries remarking some differences between cultural values of Hispanic and non-Hispanic employees and managers [20,21]  1 .
In order to fill this gap, this work aims to improve the knowledge of the factors that encourage supervisors to develop behaviours to promote a work-family culture.
The studied factors are (1) transformational leadership style; (2) work-family conflict perceived by the supervisor; (3) perception of benefits associated with the implementation of work-family balance programmes; (4) social identification; (5) institutional support received from the governing board; and (6) support from other managers.These factors are tested in an empirical study carried out in the Spanish public universities context.
Deans play a key role in the effective implementation of work-family programs and as change agents for work-family culture.They can promote the use of work-family programs, eliminating barriers to their use.As leaders, they can also transmit existing work-family policies to other members of the university intentionally or unintentionally.
It would, therefore, be useful to gain a better understanding of which factors lead the supervisors of university faculties to display behaviours that encourage of work-family culture in the faculties.
This study aims to provide knowledge on work-family literature.In addition, it could help universities to successfully implement work equality programs.The obtained findings will help universities to design better human resource management (HRM) practices for promoting family supportive manager behaviours and, as a consequence, to encourage the development of work-family cultures.
To fulfill this objective, the paper is structured into seven sections.Section 2 describes the role of deans in Spanish universities with regard to supporting a work-family culture.Section 3 introduces the research framework and hypotheses.The scope of the study and the measures are included in the Section 4. Section 5 shows the results of the study.Section 6 includes discussion.Conclusions are included in Section 7.

Deans' Behaviours to Promote a Work-Family Culture in the Spanish University Context
As an answer to recent legal and social changes, Spanish universities have progressively adopted models and management systems commonly used by companies.The analysis of the strategy of Spanish universities shows that some universities have clearly integrated corporate social responsibility principles into their strategic plans. 1 For instance, [19] point out that Anglo and Western European countries are individualistic whereas Latin American Countries are collectivist.
As a consequence, many government boards of Spanish universities have developed different programs in order to facilitate the work-family balance of universities' employees.However, it has been proven that the adoption of this type of measures alone is not enough for their effective implementation.One of the factors that favours their implementation is the existence of a work-family culture.
Work-family scholars have remarked that intermediate managers can be powerful change agents for promoting family supportive organizational cultures [16,18].Intermediate managers are responsible for informing employees about existing work-family programmes, seeing the rules that are observed and who grants or rejects employee requests.
As intermediate managers, deans can play a key role in developing a work-family culture in their faculties.In Spanish public universities, deans are university professors elected by all the members of a faculty for a four years period.They are the executive head of the faculty board, being responsible for the financial, academic, personnel and administrative affairs of the faculty.Some of the common specific duties of a dean are (a) conducting the meetings of the faculty board; (b) the coordination of departments; (c) participating in the selection and promotion procedures for the scientific staff of the faculty; (d) serving on faculty committees; (e) elaborating the faculty budget; or (f) attending to meetings with the government board of the university.Organizing and planning research activities are not among the specific duties of a dean in Spanish public universities.
Although promoting work-family balance is not a specific responsibility for deans, they have formal attributions to promote the use of some work-family programs developed by the Government Boards of their universities (e.g., scheduling meetings at a time that allows participants with family needs to attend; disseminating the available programs in the university; avoiding discrimination by family issues in personnel selection processes; promoting training programs focusing on work-life balance issues; ensuring the effective implementation of some university regulations regarding work-family issues…).Due to their status and power as decision-makers, they could communicate existing work-family policies to other members of the university, including scientific and administrative staff.They also could pay attention to family issues in public acts, meetings or informal messages, being consistent with work-family issues in their own behaviour.Nevertheless, the role of deans as promoters of balance between work and family is often limited to the faculty level by institutional factors as e.g., university regulations with regard to selection and promotion processes.

Work-Family Culture and Supervisors' Behaviours for Promoting a Work-Family Culture
Beauvais Thompson and Lyness ([12], p. 394) define work-family culture as the "shared assumptions, beliefs, and values regarding the extent to which an organization supports and values the integration of employees' work and family lives".They suggest that work-family culture is comprised of three main components: "managerial support to family issues", "time demands" and "career consequences" for employees.
The concept of "managerial support to family issues" refers to the extent to which employees perceive that their supervisors are sensitive to employees' work-family demands.Supervisors who support these type of demands are sympathetic to the wishes of their subordinates in terms of achieving a workfamily balance and believe that solving their subordinates' work-family issues is part of their role.
The "career consequences" construct refers to the degree to which employees perceive that the use of the work-life programmes implemented in their companies may damage their careers.The "time demands" component refers to the extent to which employees perceive that there are unwritten rules that value spending long hours in the workplace and prioritise work over family.
Supervisor behaviours that support the family role provide instrumental support "that leads to employee perceptions of emotional support from their supervisors" [14,16].The family supportive supervisor behaviour consists of four types of behaviours: instrumental support, emotional support, role model behaviours and recognition of the strategic importance of work-family issues [16].Instrumental support involves the provision of tangible goods and services [22] as, e.g., scheduling flexible work.Emotional support relates to the provision of caring, esteem, empathy, love and trust [22].It means that feelings of employees are considered, and they feel comfortable communicating their feelings about work-family issues to their supervisors.
Role model behaviours involves that supervisors demonstrate how to integrate family issues in the job, showing to employees what behaviours are acceptable with regard to balancing work and family issues.Creative work-family management involves actions at an organizational level such as redesigning work to reduce work-family conflict, enhancing organizational outcomes at the same time [14].

Transformational Leadership Style and Supervisors' Behaviours for Promoting a Work-Family Culture
Transformational leadership is defined in terms of the effects a leader has on his/her followers.Followers feel admiration, loyalty, trust and respect towards their leader [23].Transformational leaders are characterised by being able to induce positive social changes and transformations by encouraging employees' well-being and boosting social capital.
Transformational leaders were conceptualized by [24] considering seven dimensions later reduced to four dimensions: idealized influence, intellectual stimulation, individualized consideration and inspirational motivation.
Idealized influence is defined as the capability of leaders to become admired, respected and trusted.Intellectual stimulation highlights the leaders' ability to influence their subordinates so that they can think in a creative and innovative manner.Individualized consideration considers the leaders' capacity to provide personal attention to all the members of their team, taking into account their differences.Inspirational motivation refers to the leaders' ability to provide their subordinates with a stimulating, convincing inspiring, motivating, and attractive vision of the future [24].
Some studies confirm that transformational leadership style has a positive and considerable impact on: employees' motivation and commitment [25], job satisfaction [26], employee performance [27], and engagement [28].
Several studies also confirm that transformational leadership style increases the efficiency of work-family balance programmes.Transformational leaders, as open-minded people with new and creative ideas, will always be willing to develop and implement innovative measures aimed at facilitating a work-family balance [29].For this reason, their behaviours may result in positive consequences for the effective implementation of work-family balance programmes [30].
Taking all these considerations into account, we therefore formulate the following hypothesis: Hypothesis 1: A transformational leadership style by supervisors is positively related to supervisors' behaviours that encourage a family-friendly culture.

Perceived Work-Family Conflict and Supervisors' Behaviours for Promoting a Work-Family Culture
Life experiences influence one's beliefs and attitudes towards objects, actions and events [31].The attitude-intention-behaviour relation model states that behaviours are caused by an individual's appraisal of a situation and a subsequent emotional response [32].
Work-family conflict results from family and work responsibilities that make demands and compete with each other for limited personal resources [33].According to the attitude-intention-behaviour relation model, people who have experienced a work-family conflict may develop altruistic behaviours [34] and, as a result, be more inclined to encourage work-family culture than those people who have not been subject to that kind of experience.
We can therefore assume that the work-family conflict perceived by the faculty supervisors has a positive influence on their behaviours in terms of encouraging a work-family culture.
Hypothesis 2: The work-family conflict perceived by supervisors is positively related to supervisors' behaviours that encourage work-family culture.

Perceived Benefits of Work-Family Programmes and Supervisors' Behaviours for Promoting a Work-Family Culture
Organisational benefits can play an important role in eliciting desired behaviours on the part of employees and managers.According to [35], the implementation of work-family balance programmes can generate different benefits, namely: (a) improving the organisation's corporate image; (b) improving the work climate; (c) retaining and attracting talent; (d) reducing absenteeism rates; and (e) increasing productivity.Furthermore, these benefits are reinforced as much as they complement each other [36].For instance, improving the company's image may positively influence the attraction of talent and a better work climate may increase employees' commitment and reduce absenteeism rates.
Some studies point out that a supervisors' perception of the benefits of work-family balance programmes is a critical for their implementation [36,37].As a result, we consider that the perception of the benefits related to work-family balance programmes may have a decisive bearing on the behaviour of supervisors when it comes to encouraging the adoption of a family-friendly culture.Therefore, we formulate the following hypothesis: Hypothesis 3: The level of benefits associated with the implementation of work-family balance programmes perceived by supervisors is positively related to the development of supervisors' behaviours aimed at promoting a work-family culture.

Social Identification and Supervisors' Behaviours for Promoting a Work-Family Culture
Social identification reflects the degree to which people define themselves in terms of their membership in a social category or a social group [38].According to self-categorization theory, people who strongly identify with a social category or group are more likely to provide social support to other members of the group.
Self-categorization theory states that people who strongly identify with a social category tend to exaggerate attitudinal and behavioural similarity between themselves and other members of the group, what is called in-group homogeneity [39].Higher levels of in-group homogeneity can boost an individual's emphatic concern for another person's problem.In addition, social support literature suggests that empathy is also a key motivator of individuals helping to alleviate another person's psychological distress stemming from a personal problem [40], including family-work problems.
It can, therefore, be assumed that supervisors define their attitude towards employees in need depending on their perceived group membership.Consequently, they may be more concerned regarding employees' work-family balance.
Hypothesis 4: Supervisors' social identification with employees demanding a work-family-balance is positively related to supervisors' behaviours that promote a work-family culture.

Institutional and Other Supervisors' Support and Supervisor' Behaviours for Promoting a Work-Family Culture
Some researchers state that people will be more inclined to perform a certain behaviour if they feel that important referent individuals are likely to approve such behaviour [41].Social support theories note that it is relevant to differentiate between the support received from the institution and the support specifically provided by the immediate supervisor [10].

Institutional Support
Organisational support theory [42] states that the ability of organisations to meet organisational needs regarding personnel welfare often increases employees' motivation.Perceived organisational support (POS) measures the degree to which employees believe that their organisations are concerned about their well-being.
Wang and O' Reilly [43] point out that POS contributes helps fulfil the emotional needs of employees, reinforcing identity feelings of workers.Wayne, Grzywacz, Carlson and Kacmar [44] note that a higher POS by employees produces positive feelings towards their organisation, motivating employees to meet organisational goals.
Some studies also show that institutional support is a key factor in promoting a work-family culture [14].Therefore, in light of these considerations, we formulate the following hypothesis: Hypothesis 5: Supervisors' perception of strong support to the family from the institution is positively related to supervisors' behaviours that promote a work-family culture.

Support from Other Supervisors
One of the findings in the social cognitive literature is that individuals learn by observing other people's behaviours and subsequently, initiate and sustain similar patterns of behaviour themselves [45].Organisations' culture can be transmitted to employees through modelling by leaders.Role models influence ethical conduct and pro-social behaviour [46].
In this line, some studies demonstrate that employees' perception of support from their organisation and supervisors results in behaviours related to leadership based on support and altruistic behaviours with their co-workers [42].Altruistic behaviours could include the development of a work-family culture.Accordingly, supervisors could learn the attitudes and family-friendly behaviours of other supervisors in their faculties.
Therefore, based on these considerations, we formulate the following hypothesis: Hypothesis 6: Supervisors' perception of a strong support to the family from other supervisors is positively related to supervisors' behaviours that promote a work-family culture.

Participants
For the purpose of this study, the population is made up of 65 Spanish faculties.Data were collected between June 2012 and September 2012 using a questionnaire, sent out faculty supervisors (deans).Fifty-five answers were received, representing 84.6% of the population.Women made up 30.9 (17 answers) of the sample whereas 69.1% were men (38 answers).
The questionnaire was structured into different sections, including 48 Likert-type scale questions (from 1 = strongly disagree to 5 = strongly agree) related to deans' behaviours that encourage a work-family culture, the perception of the benefits associated with implementing work-family balance programmes, the work-family conflict experienced by deans, social identification, transformational leadership style, support from the governing board and the support and involvement of other supervisors in the centre (such as Heads of Department, Heads of Section, etc.).Negatively worded items were recorded.A section was also included requesting other data such as the gender of the deans.
The questionnaire was reviewed by three deans before being sent out, in order to ensure that it was well structured, unambiguous and clear.

Measures
A seven-item Likert-type scale adapted from [12] was used to measure deans' behaviours that encourage a work-family culture.The statements were adapted to fit the specific case of the study.The items include behaviours to (1) support employees' family responsibilities (e.g., "I told to other colleagues to be sensitive to employees' family needs and personal concerns"); (2) to eliminate negative career consequences from using work family programs (e.g., "I actively tried to promote employees even when they had family responsibilities"); and (3) achieve a better management of employees working time (e.g., "I avoided scheduling meetings so that employees with family responsibilities get in conflict").
An exploratory principal components factor analysis to the entire set of items was applied to identify the dimensions underlying the items, finding three factors with eigenvalues greater than unity.Factor 1 appertain supervisors' behaviours to support employees' family or personal responsibilities.Factor 2 includes those supervisors' behaviours to eliminate negative career consequences from using work family programs.Factor 3 appertains to supervisors' behaviours to achieve 'better management of employees working time.Together they account for 68.56 of the variance.The constructs' internal consistency is reported in Table 1.Fifteen items adapted from [47]'s Likert-type scale in its short version were used to measure transformational leadership of deans.Exploratory principal components factor analysis was applied to every item and three factors turned out to have values higher than one: (1) charismatic/inspirational; (2) intellectual stimulation; and (3) individual consideration.They account for 67.30% of the variance.
The five item Likert-type scale developed by [48] was used to measure the work-family conflict perceived by deans.Results from exploratory principal components factor analysis produced one factor, accounting for 70.46% of the variance.
Eight items extracted from the Likert-type scale used by [35] were used to measure the perception of the benefits associated with the implementation of work-family balance measures.The items selected refer to the benefits related to the image and competitiveness of the university, research groups and centres, the improvement of the work climate in general, staff's performance and the smooth running of the institution.The statements were adapted to fit the specific case of universities.
When carrying out an exploratory principal components factor analysis, three factors turned out to have values higher than one: (1) perception of the benefits related to internal operation, working environment and performance; (2) perception of the benefits related to image; and (3) perception of the benefits related to competitiveness.The total variance explained is 85.00%.
Following [31], social identification variable was measured using a three items Likert-type scale extracted from the six item scale developed by [49].We adapted the statements for the specific case of universities.The entire set of items was summarized in one factor when carrying out exploratory principal components factor analysis, accounting for 55.51% of the variance.
The Likert-type scale developed by [50] was used with the aim of analysing to what extent deans take an active part in implementing the work-family balance programmes and policies established in the university's strategic plan.Results from exploratory principal components factor analysis produced one factor, accounting for 64.42% of the variance.
The short version of the scale developed by [50] was used to measure deans' perception of institutional support.When carrying out exploratory principal components factor analysis, the entire set of items was summarized in one factor, accounting for 79.04% of the variance.
The content of each item and the scales included in the study was analysed by reviewing the literature.As scales were designed on the basis of this review and, as previously validated scales were suggested, it was considered that each item had the necessary content validity.
Before testing the hypotheses, we evaluated the reliability of the scales, by carrying out Cronbach's alpha test (Table 1).

Results
We carried out three multiple regression analyses with the aim of testing the formulated hypotheses.In the respective analyses, the following three constructs were used as dependent variables: deans' behaviours to support to employees' family or personal responsibilities; deans' behaviours to eliminate negative career consequences from using work-family programs, and deans' behaviours for a better management of employees' working time.
The independent variables introduced in all regression analyses were as follows: gender, transformational leadership, perceived work-family conflict, perception of the benefits related to internal operation, working environment and performance, perception of the benefits related to image, perception of the benefits related to competitiveness, social identification, perception of institutional support, and perception of support provided by other supervisors.
The correlations between these variables can be seen in Table 2. Before testing the hypotheses, we checked the requirements for normality, independence of error terms and collinearity.To evaluate normality, we used the Shapiro-Wilks test.In all cases, the value obtained in the Shapiro-Wilks test was higher than 0.05, which allowed us to conclude that variables followed a normal distribution.
The variance inflation factor (VIF) of the independent variables introduced in the regressions revealed no collinearity problems between variables, with all the factors below 5.More specifically, the highest VIF value obtained stood at 1.628.
In order to analyse independence between residuals, the Durbin-Watson statistic was used.The value obtained stood at 1.910 for the dean or Principal's support model, 1.937 for the career consequences model and 1.896 for the working time model.The statistical software used was SPSS 18. Bivariate correlations between variables are shown in Table 2.  Model 1 (Table 3) confirms that the support of deans for the development of a work-family culture depends on four variables, namely: (1) work-family conflict (B = 0.276, p < 0.05); (2) the style of leadership "intellectual stimulation" (B = 0.366, p < 0.05) and "individual consideration" (B = 0.233, p < 0.05); (3) institutional support (B = 0.293, p < 0.05); and (4) support received by other supervisors (B = 0.264, p < 0.05).The observed explained variance by the model is the 55.6%.
These results confirm the existence of a positive and significant relationship between two of the dimensions of transformational leadership (TL2 and TL3) and the behaviours that promote the development of a work-family culture by deans.Hypothesis 1 is, therefore, confirmed, in this case, in terms of the "intellectual stimulation" and "individual consideration" dimensions of transformational leadership.
Hypothesis 2 is also supported when measuring behaviours to encourage a work-family culture in terms of the deans' behaviours to support family issues construct.This means that the deans who have personal and very close experience of the work-family conflict are more likely to support a work-family culture in their faculties or schools.Furthermore, the positive and significant relationship found between both institutional support and support provided by other supervisors and deans' behaviours to encourage a work-family culture-when measured in terms of the support provided by deans-also sustains Hypotheses 5 and 6.
The Results for model 3 show that the support provided by deans for the development of a work-family culture in terms of working time depends on the development of a leadership style oriented towards "charismatic/inspirational" (B = 0.301, p < 0.05) and individual consideration (B = 0.242, p < 0.10) and the social identification (B = 0.242, p < 0.10).Therefore, Hypotheses 1 and 4 are confirmed when considering the working-time model of the work-family culture.
Finally, none of the proposed hypotheses was supported in the career consequences model.The results show that dependent variables of the model are not significant to explain deans' behaviours in order to promote a work-family culture in which family responsibilities are not an obstacle to career progress.All hypotheses regarding the perceived benefits have also been rejected.

Discussion
Analysing results from a global point of view our research confirms the existence of a relation between work-family conflict and deans' behaviours that encourage a work-family culture.Results show that the perception of work-family conflict by deans positively affects their behaviours to encourage work-family culture.
Results also confirm that deans are more likely to develop behaviours for promoting a work-family culture when they perceive that their university and other academic managers in the same centre are sensitive to the problems caused by the interaction between work and family.This finding is consistent with earlier research from social exchange theory that found that when employees recognize they are supported by their organisation they are more motivated to reciprocate through increased organisational citizenship behaviour [51].
Regarding support issues, some studies have shown that work-family specific support could be a more psychologically and functionally used resource to manage work-family stressors than general institutional support.Different types of support from different sources may have specific impacts on differential types of conflict [52].As such, subordinate perceptions regarding supervisor support in the university could yield interesting findings for a deeper knowledge of this factor.
Deans with transformational leadership skills are also more likely to develop behaviours promoting family-friendly cultures.This result is consistent with those that highlight the importance of transformational leadership style on promoting the development of a family-friendly culture in a business context [34].
Finally, the results show that those deans who identified with subordinates in need of work-family cares are more likely to promote a work-family culture.This finding is consistent with several previous studies from self-categorization and social support theories.[40] found that Chief Executive Officers (CEOs) who identify more strongly with the corporate elite provide more support to fellow CEOs.
On the other hand, all hypotheses regarding the career consequences model were rejected.Deans participate in all selection and promotion processes of scientific staff in their faculties.However, university regulations and current practices limit their influence, concentrated in hands of other members of the scientific staff.This fact could limit the explanation power of the career consequences model.
All hypotheses regarding the perceived benefits have also been rejected.According to our study, organisational benefits affecting the university and the image of the research group, competitiveness and working environment are not decisive when implementing work-family programs.
We have to consider that models for analysing influencing factors for promoting a work-family culture were, in general, developed in a business context.As public administrations, universities have specific features that can explain differences when compared with results at a business level.
Although deans and business supervisors have a similar position in the structure of their respective organisations, some questions can affect deans' behaviours as managers.
There is little relation between deans' subordinates performance and deans' incomes, promotion and position at the university.Their election depends on the vote of the members of the faculty, but there is not another external assessment of their performance.Additionally, they have academic aims above and beyond achieving profits for the institution.
In other words, there is not a strong formal pressure from the university that encourage deans to implement work-family programs in their faculties: they have formal attributions to do so but not a strong obligation.In this context, deans' personal feelings regarding work-family problems or transformational leadership skills seem to be more relevant that issues such as organisational benefits.This is an important difference regarding business supervisors.
Further research is needed to confirm this hypothesis.If it is correct, changing deans' present roles could be useful in order to promote a work-family culture in the Spanish university context.
In the short term, training programs underlying the importance of implanting work-family programs could be a first step.Secondly, institutional actions are needed to increase the influence of deans to avoid negative career consequences of their employees.
In the long term, structural changes are needed.Adding specific aims connected to results of their centres (e.g., research aims, subordinates satisfaction) to dean's academic objectives and transforming the selection processes for deans' positions in order to increase deans' participation in these processes and select valid managers to achieve these aims, are transformations required to meet the new challenges facing public universities in Spain.

Practical Implications
The obtained findings have a practical relevance for human resource managers of universities.One of the novel findings of this preliminary study is that deans' personal feelings and experiences regarding work-family problems are a relevant factor in order to implement work-family programs in their faculties, above and beyond organisational issues.
The results show that some personal feelings and experiences of deans such as social identification and work-family conflict should be taken into account for promoting a work-family culture.
For instance, being a young woman with children is not usually well looked upon when applying for a job position in some Spanish organisations.However, public universities interested in promoting a work-family culture could be interested in young women for deans' positions, since a higher exposure to work-family conflict makes them more likely to develop behaviours that encourage work-family culture.
Results also show that institutional support to work-family issues by the university is a relevant factor for encouraging deans to support family needs.This finding suggests that university boards should go beyond developing family supportive programs, it being important that university leaders act in a way that demonstrates the organization's commitment to work-family issues.Moreover, the university could provide deans with information and other resources to implement work-family programs in their faculties.This information could be useful for changing HRM practices such as recruitment and selection processes and training courses in order to promote a work-family culture.
For example, universities' human resource (HR) managers could redefine recruitment and selection processes for deans' positions in order to consider personal experiences of candidates on work-family conflict.
Secondly, universities' HR managers could design specific family supportive training programs for deans.Some evidence in the business context shows that computer-based training, behavioural self-monitoring training, and face-to-face training led supervisors to behavioural changes that promote a work-family culture [16].
Supervisor training programs could include (a) providing deans and other supervisors with needed formation about the importance of work-family programs for their universities; (b) information on the benefits of implementing a work family-culture for employees and their families' well-being; (c) information on the universities current work-family policies; or (d) deans' expected behaviours to promote a work-family culture.
Our research provides useful information for training programs design, suggesting that (1) providing deans with information on existing policies regarding work family issues in order to increase perceived institutional support; and (2) promoting deans' training in order to develop transformational leadership skills should be considered by Government Boards of Spanish public universities in order to really implement a work-family culture.

Limitations and Future Research
The results of this paper are based on the perception of one group within the university and may not be representative of the organisation as a whole.Differences in terms of salaries, job flexibility, and position in the university structures between deans and other members of the universities could explain different perceptions of work-family conflict, organisational support and supervisor support.Although using perceptions of relevant actors is a common way to approach constructs in business management research (see e.g., [5,16,34] or [53]) future research should also incorporate into the study the opinion of all the university staff, including professors and all categories of Administrative Staff.
Second, cross-sectional data makes it impossible to determine the causal relationships among the studied variables.Future research should develop longitudinal studies in order to make causal statements based on our results.A larger sample will improve significance of obtained findings.

Conclusions
This article analyses which factors determine faculty supervisors' behaviours that encourage a work-family culture.Our research provides useful information about influencing factors for supervisors promoting a work-family culture, increasing previous knowledge from other different contexts.It is important to point out that, in Spain, this subject is new.
In particular, this research is aimed at contributing to ensure the implementation of a work-family culture within the university sphere.The study focuses on the factors that affect the implementation of this type of culture from the point of view of deans, a group with the ability to encourage a family-supportive culture within their centres.
Results show that factors such as transformational leadership, perception of work-family conflict by deans, social identification, institutional support and support provided by supervisors are relevant factors for explaining the promotion of a work-life culture by this group.

Table 1 .
Reliability of scales, constructs' internal consistency and dimensions of dependent and independent variables.
SB: Supervisors behaviour that encourage a work-family culture; TL: Transformational leadership; Perception of the benefits.