Work-family conflict and facilitation among teachers in Israel and Switzerland

In this paper, we examine the attitudes toward integrating work and family in a sample of 247 teachers in Switzerland and Israel. More particularly, we focus on the national context’s role in mediating the relations between professional and private spheres. The data were collected by a questionnaire implemented and administered in the two countries. The analysis reveals differences between Israeli and Swiss teachers regarding the importance of attribution to life roles and their attitudes toward conflict and facilitation. Findings suggest new insights into the consideration of cultural elements in shaping the teachers’ attitudes toward the integration of family and work.

lives, namely, professional and private. As the difficulties in finding a balance between the work activity and the home sphere are one of the central issues for teachers, and often a cause of dropouts from the career (McIlveen et al., 2019;Van Droogenbroeck et al., 2014), we intend to analyze the attitudes which can have an impact on this issue, as well as the facilitating elements that can mitigate the conflict. We also focus on the cultural aspect of the issue since it is an essential factor in shaping the professional-private life interface. Accordingly, we chose to locate our study in two countries (Israel and Switzerland) with similar predicted levels of women and men imbalance between work and family, based on national policies (Cullati, 2014;Stier et al., 2012).
The paper is organized as follows: in the "Teachers' balance of work and family" section, we present a concise review of the literature on the conflict between work and family and the facilitation attitudes among teachers; in the "Context of the study, aim, and research questions"section, we introduce some contextual aspects of our study, its aim, and the research questions; the method and the procedures are described in the "Method" section; afterward, we present the results of the analysis (the "Results" section), as well as a discussion of the main findings (the "Discussion" section); finally, we conclude the paper by reflecting on challenges, as well as some limitations and strengths of our study (the "Conclusions" section).

Teachers' balance of work and family
The interface concerning work and family often appears in the research literature as a "conflict in which the role pressures from the work and family domains are mutually incompatible in some respect […] participation in the work (family) role is made more difficult by virtue of participation in the family (work) role" (Greenhaus & Beutell, 1985, p. 77). Although the general term of conflict between work and family already indicates the individuals' ideas that professional and private responsibilities can interfere with each other, Frone (2003) suggests referring to this issue by using a more specific distinction between work-to-family conflict (WFC) and family-to-work conflict (FWC). The first occurs when work-related demands interfere with home responsibilities (for example, when work assignments occupy nonworking hours and limit the person's time and space to attend family affairs). The second arises when family responsibilities impede work activities (for instance, when caring for a family member limits their ability to address work assignments).
The research literature indicates that FWC is less common than WFC (Bellavia & Frone, 2005) and that both are connected to different difficulties concerning professional and personal lives (Whiston & Cinamon, 2015). FWC is associated with various negative outcomes such as dissatisfaction, poor functioning at work, and burnout (Cinamon & Rich, 2010;Slan-Jerusalim & Chen, 2009), while aspects concerning family life, like marital dissatisfaction and unsupportive spousal relations, have been recognized as being influenced or mediated by WFC (Grzywacz & Marks, 2000). Both WFC and FWC are associated with poor psychological well-being, such as depression, anxiety, and mood disorders (Franche et al., 2006). Barnett and Hyde (2001) indicate that besides conflictual interfaces between work and family, one should also acknowledge a facilitation relationship. It concerns the enrichment and positive spill-over emphasizing the beneficial impacts of integrating work and family in people's lives (Greenhaus & Powell, 2006;Iskra-Golec et al., 2016).
Similar to conflict, the facilitation relationship can also be divided into two different forms: work-to-family facilitation (WFF) and family-to-work facilitation (FWF). The first one relates to a positive influence the work domain has on family lives (for example, when professional skills are useful in family engagements). The latter refers to a beneficial effect the family domain has on professional lives (for instance, when practices held in family lives are helpful in the working place). WFF is correlated to general life satisfaction, physical and mental health, and is a better predictor of job satisfaction than family satisfaction (Chen et al., 2014;McNall et al., 2010). FWF is influenced by a stable family identity, the nature of the relationships among family members, and the family-based skills implemented at work (Ben-Uri, 2016;Whiston & Cinamon, 2015).
Focusing on teachers, work-family balance is usually mentioned in research regarding the reasons for choosing the profession (Struyven et al., 2013). Nevertheless, the conflict between the teaching profession and the home sphere is often described as one of the central stress factors in working activities (Goyal & Arora, 2012;Palmer et al., 2012). According to Alvarado and Bretones (2018) and Demirel and Erdamar (2016), a teacher unable to spare enough time for his/her family due to a heavy workload and obliged to take it home (lesson planning, preparing, or grading exams) might lead to WFC. Teachers having to push work aside due to childcare and household tasks might lead to FWC. Furthermore, the nature of the professional skills and the emotional inputs teachers express during working hours are very similar to those they imply as parents . Rich (2005a, 2005b) point out that WFC among teachers is prevalent with respect to other occupations. In another work, Cinamon et al. (2007) investigated the relationship between work and family generic and occupation-specific stressors, as well as support variables among high school teachers. The authors assess WFC effects on burnout and vigor by finding a relation to distinctive teaching characteristics (e.g., investment in student behavior and parent-teacher relations). Both WFC and FWC are predicting burnout, whereas only FWC predicts vigor. Cinamon and Rich (2010) studied female married teachers' interrelations between conflict and facilitation regarding spousal, managerial, and collegial social support (serving as antecedents), and professional vigor and burnout. The results reveal a complex relationship between conflict and facilitation and different association patterns to the work and family domains. Only managerial support predicts conflict and facilitation relations. In a similar vein, Drago (2001) points out the organizational factors as essential elements in understanding WFC among teachers. He found a positive correlation between school characteristics such as the supportive atmosphere and the length of time parents (teachers) spend with their children. Carlotto and Camara (2018), who examined sociodemographic, labor, and psychosocial factors concerning work-family interactions among female-teachers in Brazil, indicated that overload, weekly workload, performance of another occupational activity, social support, and self-efficacy variables are predictors of the work-family interaction dimensions. For instance, WFC increases as overload and weekly working hours increase, and social support at work decreases, in addition to developing another occupational activity. An interesting perspective of work-family balance among female teachers was recently described by Edwards and Oteng (2019). Their study focused on female teachers' opinions in Ghana about the work-life balance and career progression, in the midst of a high societal expectation for them to be role models. The study underlined the female teachers' position as leaders in balancing their career goals with societal demands and expectations, which often compete or interfere with their private lives.
Previous research has been limited by a dichotomous observation of work versus family, assuming that one must exclude the other and that a high level of family centrality, for instance, will probably appear along with a low level of work centrality. Consequently, there is still a lack of research demonstrating the possible joint effects of work and family centralities on conflict levels.
In our view, the question is not exclusively related to the centrality attributed to one of the spheres. We advance an argument in favor of culture's impact in mediating the relation between work and family among teachers. Although the effect of cultural context on different aspects of teachers' work stress has been addressed in previous research (Klassen et al., 2013), little attention has been given to teachers' attitudes toward work and family relations and the role of the cultural context. In a study conducted by Moore (1995) on work and family conflict among university teachers from five different countries (Australia, Bulgaria, The Netherlands, Israel, and the USA), the cultural origin was a significant factor for examining the intensity of the conflict between work and family. More particularly, Bulgarian and Israeli teachers reported a stronger conflict than participants from the other countries. They indicated a greater contribution of the home atmosphere, rather than by a workload, to create a conflict. Stier et al. (2012) claim that various features of advanced industrial societies and economically less developed societies produce differences in the balance between work and family among teachers in societies. Nevertheless, it is not clear in which group of countries one would find stronger similarities or differences in the perception of WFC and FWC, as patterns can vary considerably and be assumed as a private or political matter according to the countries' cultures and welfare organizations.
For the present study, we decided to focus on two countries (Israel and Switzerland) that, despite different policies, share a societal interest in balancing professional careers and private lives. In particular, Switzerland is considered one of the best countries according to the reduced work-family imbalance for employed men and women (OECD, 2017). In contrast, in Israel, this imbalance is much higher (Stier et al., 2012). For this reason, we intend to analyze more in-depth the conflict and the facilitation between work and family among teachers in the two countries. To specify the peculiarities of these contexts in terms of the teaching profession, we present a short overview of both education systems, the goal and the questions addressed by our work in the next section.

Education systems in Israel and Switzerland
The Israeli education system is governed by national laws implemented by the Ministry of Education. Public education is compulsory and free for children aged between 3 and 18. The national education system consists of 5 levels: pre-primary, primary education (grades 1-6), secondary (grades 7-9), post-secondary (10-12 grades), and higher education. There are separate schools for the Jewish and Arab communities. However, some Arab pupils attend Jewish schools and several bilingual schools for both populations. As for 2016, there were about 170,000 employed teachers with a prediction of a shortage of around 7700 teachers in the next few years (ICBS, 2017). Israeli teachers' salary is 13% lower than other workers with similar training years in the country. This proportion is close to the OECD countries' average (Wiessblie, 2013). Teacher education in Israel is provided by (1) studies in teacher education colleges-which provide BA degrees (3-4 years of training) and MA degrees (additional 1-2 years of training) and offer specialization in a specific age group and discipline, and (2) studies in education schools in universities and programs for career retraining in colleges-designated for students who are studying to acquire BA or MA degrees. The studies provide a teaching certificate after 1 or 2 years of training.
Swiss Confederation consists of 26 cantons with considerable autonomy in the domain of public education that is free and compulsory for at least 9years. Due to the cantonal organization, the entry age and the number of years of compulsory education can vary. However, an effort to adopt standard practices in the different cantons has been made. In most of the cantons, compulsory education starts at age 4 and has a duration of 11 years. Neuchâtel is the Swiss canton in which the study has been mainly performed. Its educational system consists of 11 years of mandatory education organized around three levels. "Cycle 1" concerns the years 1-4, "cycle 2" concerns the years 5-8, and "cycle 3" includes the years 9-11.
Teacher education is provided at the university level, which delivers BA (for primary education) and MA degrees (for secondary and special needs education). Salaries of teachers are not far from the median salary in Switzerland, although they depend not only on the teacher's experience and his/her teaching degree but also on cantonal rules. In 2016, 93,500 teachers were employed in 1-11 grades and 26,000 teachers in post-compulsory grades (for students up to 18 years old).

Aim and research questions
Our goal is to analyze the interface between teachers' work and family (in terms of conflict and facilitation) in the two selected countries.
To reach our objective and to compare the factors mentioned above in Israel and Switzerland, we address the following research questions (RQ): RQ1: Would there be any differences between Israeli and Swiss teachers regarding the levels of importance attributed to work and family roles? RQ2: Would there be any differences between Israeli and Swiss teachers regarding the work-family (or vice versa) conflict levels? RQ3: Would there be any differences between Israeli and Swiss teachers regarding the levels of work-family (or vice versa) facilitation attitudes?

Method
A team of researchers set the study in the frame of an international collaboration established years ago. All the actions conducted in each country were discussed, agreed upon, and implemented in a similar way. The ethical research principles of the implied institutions were guaranteed. Adaptations related to specific contextual aspects were shared and discussed by the international team.

Sample
A total of 155 teachers of all disciplines and degrees of education at the Beit Berl College (Israel) were involved in the study. In Switzerland, 92 teachers of all disciplines and degrees of education (primary and secondary school teachers) at the Haute Ecole Pédagogique BEJUNE (Switzerland) were participating. In both countries, a sample of 247 teachers was constituted, and women represented more than 70% of the sample (see Table 1). However, some differences in the demographics of the two samples are observed. Teachers were questioned about their ages according to four groups: up to 25 years old; 26 to 30 years old; 31 to 40 years old; and older than 41. It turns out that 7.7% of the teachers in Israel and 18.5% of the teachers in Switzerland are up to 25 years old. Similarly, 29.6% of teachers in Israel are below 30 years old, whereas 52.2% of teachers in Switzerland are below 30 years old. This structural difference in the two samples is probably due to the compulsory (for women) and more extended military service in Israel, so it does not appear to be a sampling bias. Moreover, in Israel, 59.4% of teachers have two or more children, whereas, in Switzerland, only 26.1% of teachers have two or more children. More particularly, a third of the Israeli participants have three children or more, whereas less than 10% of the Swiss sample has three children or more.
The researchers contacted all the participants through emails presenting the objective of the study and the conditions to participate. The teachers were enrolled in the study voluntarily, and they did not receive any financial reimbursement. In line with the ethical framework guiding the research, the teachers were assured that their anonymity would be maintained at all stages of the study. The data were treated in the strictest confidence and used only by the researchers.

Instruments
A questionnaire measuring the teachers' perceptions regarding the relationship between work and family was administered in the two countries between November 2016 and February 2017. The instrument was built in the form of a four parts questionnaire.
The first part included respondents' personal information (age, gender, marital status). The second part was adapted from the Life Roles Salience Scale (LRSS), measuring the attribution of importance to life roles (Amatea et al., 1986). Initially, it included 40 items (5 items for each of the 8 dimensions of investigation). For the current research, we decided to consider 6 dimensions. The 6 selected latent variables give values for occupation role reward, occupational role commitment, parental role reward, parental role commitment, marital role reward, and marital role commitment. Each dimension is defined on the basis of 5 items, and for each teacher the value is defined as the average value of the five items involved in the definition. This section included items like Having a career that is interesting and exciting to me is my most important life goal; I expect to be very involved in the day-to-day matters of rearing children of my own; I expect to work hard to build a good relationship with my  Campbell et al., 1994). The third part was based on the conflict and facilitation between work and family perceptions' questionnaire (Wayne et al., 2004). It included 16 items, namely, 4 items for each of the dimensions we investigated (WFC, FWC, WFF, FWF). We refer to this model as the WF/FW/C/F 4-factor model. Both questionnaires are based on a 5-level Likert scale and are widely used in literature.
The fourth part included qualitative issues that are not pertinent to the present study. The original questionnaires are in English. We used Hebrew versions of them for the Israeli sample and French versions for the Swiss sample. Concerning Israel, the attribution of importance to the life roles part was adapted by Cinamon and Rich (2002) and Shick (2008). The conflict and facilitation attitudes sections were adapted by Cinamon and Rich (2004). Concerning the Swiss instruments, we adapted and translated the questionnaire of Wayne et al. (2004). The LRSS was adapted from the French version of Lachance and Tétreau (1999). The Cronbach alpha values for both Hebrew and French questionnaires are resumed in Table 2. Although it is much more usual to require Cronbach alpha values exceeding 0.7, according to Loewenthal (1996), an alpha coefficient of 0.6 could be acceptable. We got most of the values above 0.7, but some of them are slightly below 0.7. This result might be due to a new and never tested before translation into French.
Confirmatory factor analysis shows that, for both countries, the comparative fit index is better for the WF/FW/C/F 4-factor model than for the LRSS 6-factor model. Comparative fit indexes (CFI) and Tucker-Lewis indexes (TLI) are globally above or not far from 0.9, a value that is indicatively considered as for a good level of configural, metric, or scalar invariance (see Table 3). Moreover, when a multigroup analysis is performed, measurements of invariance show a value of ΔCFI and ΔTLI <0.01 between configural and metric models. Values of the root mean square error of approximation (RMSEA) are always below 0.1 and more often below 0.08. The differences of RMSEA in the nested group comparisons are always below 0.01. For the subsequent analyses, we may assume a measurement invariance allowing to confidently proceed with some limitations that will be discussed further.

Analytical approach
In the present paper, we are concerned with the first three parts of the questionnaires we administered. The data were treated and analyzed through SPSS. After performing usual reliability tests and a confirmatory factor analysis intended to check the measurement invariance, we did statistical tests to compare the teachers' answers in Switzerland and Israel. We performed several t tests for independent samples to relate the teachers' importance of attribution of life roles and work-family relations, including conflicts and facilitations. We also compared (through ANOVA) certain variables among the four age groups of teachers.

Results
As already mentioned, the dependent variables were measured by the 5-level scale items for which the score 1 means "not agree at all" and score 5 "completely agree." Tables 4 and 5 show the mean values and the differences (Δ) with their statistical significance according to a two-sample independent t test. For each two-sample independent t test, tables show t statistics and provide a measure of the effect size in terms of Cohen's d statistics.

A comparison of the importance of life roles within the countries
Each average measure of the two samples for each of the analyzed dimensions gives a value mostly above 3.0 on a 1 to 5 scale, indicating that these factors are designated as important in the teachers' lives. Regarding the comparison between Swiss and Israeli teachers, we found that Israeli participants attribute significantly higher importance to life roles than Swiss teachers (see Table 4). This result concerns the items related to the occupation reward or commitment factors, marital commitment factors, and parental reward or commitment factors. In both countries, occupation role reward values are higher than occupation role commitment values. This result reflects maybe a generally good integration of work and personal issues. Marital role commitment values are higher than marital role reward values denoting the importance of commitments in both societies' marital roles. We also note that parental role reward value is higher than parental role commitment value in Switzerland, but lower than parental role commitment value in Israel. It is worth noting that an ANOVA shows no significant differences in the estimated means of any of these values among the four different age groups of teachers in both countries.

A comparison of conflict issues in the relation between work and family
The analysis of the global sample shows that WFC level is generally higher than FWC. WFC mean values in Switzerland and Israel are all above 3.0 on a scale from 1 to 5, indicating real conflict issues in the interaction of work and family life without significant differences between the two countries. On the other hand, for the FWC, the mean values are 2.98 for Israel and 2.52 for Switzerland. A t test (see Table 5) reveals that Israeli participants have significantly higher levels of conflict issues in the interaction between work and family than in Switzerland, with a medium effect size.

A comparison of facilitation issues in the relation between work and family
In general, facilitation attitudes are largely recognized by respondents in both countries. Average values of WFF and FWF in Israel or Switzerland are always above 3.3 on a 1 to 5 scale. However, facilitation attitudes appear to be significantly more important in Israel than in Switzerland, for both WFF and FWF, with a medium effect size. Facilitation values in Table 5 appear to be generally higher than conflict values. It is not clear whether teachers' lives are effectively more made of facilitations than conflicts, or whether teachers resiliently try to give less importance to problems and give more importance to the facilitations related to their role.
When we look at these results in terms of demographic variables, WFF, WFC, and FWC do not show significant differences across various teachers' age-based groups. However, in Israel and Switzerland, FWF mean scores across four age-based groups show some differences (see Table 6). As showed by ANOVA, these differences are significant in Israel (p=.022), but not enough in Switzerland (p=.127), probably because of a smaller size sample in this country. In both national contexts, teachers who are over 40 years are related to slightly lower values of FWF, reflecting the higher difficulty that senior teachers have in integrating work and family lives.

Discussion
The extent of studies focusing on the national context's role in mediating the relations between work and family among teachers is still limited. In the present paper, we investigated the teachers' attitudes toward the teaching activity and the relationship between professional and private lives in Israel and Switzerland.
In both countries, we observed that occupation role reward values are higher than occupation role commitment values, as a possible sign of good integration of work and personal issues. Marital role commitment values are higher than marital role reward values denoting the importance of commitments in both societies' marital roles. We also note that parental role reward value is higher than parental role commitment value in Switzerland but lower than parental role commitment value in Israel. This result is probably due to a difference in family structures in the two countries. Regarding the levels of the importance of attribution to life roles, it is interesting to note that the scores are much stronger in Israel than in Switzerland: although there is no apparent reason to justify this difference, a possible explanation refers to the family structure's discrepancies and its traditional role in the two societies. This cultural difference and the effect on the teachers' conception may be explained by previous research pointing out the tendency of people living in developed societies to choose the teaching profession because of intrinsic and altruistic motivations (Bastik, 2000;Kyriacou et al., 2003;Watt et al., 2012). By contrast, in underdeveloped and developing societies, the choice is mostly based on extrinsic motivation, such as favorable employment conditions (Çermik et al., 2010;Park & Byun, 2015;Yuce et al., 2013). Even though Israel is no longer a developing country, it might be that some cultural patterns, like the centrality of the family in one's life, are closer to the features of traditional societies rather than to the cultural characteristics of Western developed societies. This possibility is supported by Klassen et al. (2011), showing how Canadian teachers made significantly fewer references to the family as a source of motivation for a teaching profession in comparison to Omanis. This interpretation might be helpful to understand the additional findings of the present study. We found differences in the teachers' attitudes toward conflict between Switzerland and Israel, especially regarding FWC, which often appears. Distinctions in the marital and parental status of the two samples and the higher levels of the importance of attribution to almost all life roles in the Israeli sample may explain the higher level of conflicts as an attempt to meet the high requirements of the work and family spheres. This interpretation fits previous studies pointing out that work and family centralities may provide a better understanding of how individuals cope with the conflicts they experience in everyday lives (Carlson & Kacmar, 2000).
The conflict's depth depends on the measure of centrality and salience the individual attributes to each component. Nevertheless, high levels of attribution to parental and marital roles in the Israeli sample seem to do not result in lower FWC levels. A possible explanation might be the higher relevance of attribution to the occupational role advanced by the Israeli participants with respect to the Swiss respondents. This possibility is in line with the claim that a focus on the moderating effects of work or family centrality narrows the understanding of a more complex picture of the issue (Bagger & Li, 2012).
Concerning the work and family facilitation attitudes, our findings differ from previous evidence regarding the fact that the facilitation attitudes did not mediate the conflict levels (Whiston & Cinamon, 2015). Along with higher levels of conflict attitudes, teachers in Israel also show higher attitudes of WFF and FWF than Swiss teachers. This result might also be explained by un-exclusive perspectives because one apparently can perceive at the same time difficulty and a feeling of enrichment (Barnett & Hyde, 2001;Ben-Uri, 2016).

Conclusions
The present study suggests that understanding the interface between work and family may have practical significance, mainly because the family is a more permeable domain than the work (Depolo & Bruni, 2015). We are aware that job satisfaction is not only related to outcomes such as engagement but it also affects other people interacting within the school context. For instance, as the teachers' greater achievement is encouraged through the job satisfaction experience (Caprara et al., 2006;Sargent & Hannum, 2005), this aspect should be considered for further investigations. As the teachers' work can extend beyond the job site and can require extra efforts when teachers are at home, the flexible schedule at home may encourage discretionary behaviors that are often needed. These aspects seem crucial to invite school principals and managers to consider the importance of time-based facilitation to improve teachers' social behaviors by investing more in flexible professional schedules and family-friendly programs. Similarly, societal policies can play a crucial role in balancing the relations between work and family and in framing the capacity of teachers to act and provide adequate services for different types of people.
We conclude this paper by highlighting some limitations of our study. The first weakness concerns the reduced number of institution participants that belong to this research, and the relatively modest size of the samples for such purposes. In fact, all teachers were recruited at one institution per country. In our opinion, this should be considered as a limitation in generalizing the findings because elements such as the "local" (intra-institutional) culture could contribute to determining a certain attitude. A second limitation is connected to the nature of our investigation that constrains inferences about the cross-cultural comparison. Both these limitations appear in not fully satisfactory confirmatory factor analysis results. Indeed, we are convinced that a longitudinal design should be a viable way to investigate culturalrelated elements better and promoting the value of the samples we have considered in both countries. Although economic and political matters can substantially impact the evaluation of country-related issues, a view on the teaching activity should provide insights into better understanding how people manage to balance private and professional lives.