Religious Identity and Family Practices in a Post-Communist Society: The Case of Division of Labor in Childcare and Housework

Abstract

This paper aims to analyze the way religious identification and practices influence family practices in the division of labor in childcare and housework in contemporary Lithuania. The analysis is based on a quantitative survey (n = 3000) representing the last Soviet generation born between 1970 and 1985. The sample was distributed across five groups according to religious identification and practices—devout religionists, somewhat devout religionists, traditional religionists, cultural religionists and secularists. Statistical data analysis showed devout religionists and secularists were applying equal childcare and housework division practices. Meanwhile, the other three groups were practicing more traditional types of childcare and housework division practice where the main role is played by women. The results also show that religious identity is not relevant in explaining the way couples share housework duties. The results show that religious identification may lead to diverse family practices regarding childcare and housework divisions: reflexive and practiced (non)religious identification leads to more egalitarian family practices.

1. Introduction

Religion and family are both intertwined and interdependent social institutions (Edgell 2009). The role of religious values and their impact in contemporary societies continue to be the focus of researchers as both institutions go through the changes affected by modernization processes (Ammerman and Roof 1995; Ammons and Edgell 2007). The most recent research into the way religion and family institutions interact focuses on gendering practices and their impact on childcare and housework divisions (Goldscheider et al. 2014; May and Reynolds 2018; Perales and Bouma 2019; Frenkel and Wasserman 2020; Gull and Geist 2020).

Societies that lived under the Soviet regime experienced its effects on both religion and family social institutions. Changes in the field of religion during the Soviet period and after the collapse of the Soviet Union have been widely researched (Ramet 1992, 2010, 2014; Pelkmans 2009; Ramet 2010; Pollack et al. 2012; Simons and Westerlund 2015). Similarly, the changes in family patterns in respective societies have been researched, focusing on gendering practices (Maslauskaite and Steinbach 2020). As of yet, the interaction between the two social institutions in post-communist societies has received little scholarly attention. This paper will contribute to academic discussion on the role of religion and its values in everyday life particularly focusing on family practices in the divisions of childcare and housework within the post-communist society of Lithuania.

In contemporary societies, religion is individualized, privatized and public (Casanova 1994); similar tendencies might be observed in post-communist societies (Pollack et al. 2012). In many countries of post-communist societies, a backlash against Western values might be observed. In the wake of national populism on both sides of the Atlantic, Brubaker observed the undergoing shift from nationalism to civilizationalism, and Christianism, the secularized “Christianity-as-culture”, emerged in the public (Mouritsen 2006, 77 quoted in Brubaker 2017). While Brubaker focused on the public sphere, developments in societies of Central and Eastern Europe (CEE) might be considered as the contexts where the latter shift is also evident in the private sphere, particularly considering the anti-gender movements, challenging the rights of LGBT people and promoting so-called traditional family values (Stoeckl 2020; Schwörer and Romero-Vidal 2020). Existing social tensions have historical roots in Soviet experimental modernization, which had an effect on both religious and family realities. The communist state-imposed secularism and gender equality are continuously challenged even after the thirty years of introduction of liberal democracy and neoliberal market economy. In this paper, we aim to analyze the interaction of religious values and family practices of the division of labor in childcare and housework among representatives of the last Soviet generation in the post-communist society of Lithuania.

The term “generation” is used in this work to refer to sociologist Karl Mannheim (1952, pp. 276–320), who defined it as a group of people who share a common historical place in the social process. This part of the population has a distinct set of potential experiences, which predisposes it to a particular mode of cognition and experience, as well as a particular form of historically appropriate action. Every generation is distinct, although it shares much with its forefathers, as Wade Clark Roof (2009, p. 632) put it; generational identity, values, and outlook are all shaped via processes of cultural exchange and negotiation. Similarly, according to Roof, religion serves as a source of continuity across generations, as evidenced by custom and tradition, yet the cultural flexibility and diversity of new forms, both personal and social, are also present (Roof 2009, p. 632).

In the decades following 1990, members of the last Soviet generation, those born between 1970 and 1984, engaged in active family life (partnership formation, childbirth and child raising, and marital life) in the backdrop of growing neoliberal capitalism and social reforms. The last Soviet generation experienced social, economic and political transformations—the fall of the Soviet Union and the restoration of the state of Lithuania in the 1990s. In this generation's religious worldviews, the gradual shift from a dominant worldview of scientific atheism in the late 1980s to the return of religion to the public sphere, the re-emergence of Roman Catholicism as the national Church, and increasing tendencies of religious individualism were among the changes (Ališauskienė 2021).

In this paper, we present a theoretical discussion on religious identities and religious values of the last Soviet generation, followed by the analysis of family values, particularly focusing on gender roles and gender equality in the spheres of childcare and housework division. The empirical part of this paper focuses on the statistical analysis of the quantitative data (n = 3000) gathered in the Families and Inequalities Survey which investigated the life chances and choices, patterns of family lives and social mobility of the last Soviet generation and included questions on religious values.

2. Religious Identity and Religious Values in the Context of Social Transformations

Identity is the characteristics that an individual attributes to themselves, distinguished by social and self-conscious identities. Gender, sexual orientation, citizenship, religion and ethnicity become the source of social identity. Social identity shows how an individual sees another and what qualities are attributed to him or her. Individuals may have different identities at the same time and connect and reconstruct them. Religious identity might be used as a social marker to mobilize against others who are seen as a threat. Theories of religious identity argue that these mobilizations are strategies and can change over time and as the context changes. In addition, these mobilizations are also rational efforts to gain power for others (Banton 1983; Jenkins 2014; Day and Lee 2014).

Generally, in the social sciences, it is agreed that identity is socially constructed through the interaction of the individual with the social environment. Anthony Giddens (1990) coined the term “reflexive self-identity”, referring to the ongoing process of self-defining in the light of new information, and social practices following this process are constantly examined and reformed. Reflexivity brings individualization, according to Beck (2002), constant reconstructions and the need to actively engage with one’s everyday life, make decisions and take risks for it. Thus, modernity challenges the way social identity was constructed during primary and secondary socialization by observing and comparing oneself with others, drawing boundaries between oneself and others and connecting with a social group. Religious identity also faces challenges in modernity. Traditionally, religious identification commonly referred to a family inheritance of belonging to a religious community. Modernity brings the choice of religious identity with consequent challenges and risks. Religious identification may not necessarily lead to expressing one’s religious beliefs and following its practices: criticism of one’s religion can also be observed.

The majority of the Lithuanian population identify with Roman Catholicism (Ambrozaitienė et al. 2013, p. 152). The diversification of beliefs and practices among those who identify as Roman Catholics in Lithuania was observed within social research (Laumenskaitė 2015; Kuznecovienė et al. 2016; Žiliukaitė et al. 2016). Various typologies of religious identity dependent upon the social and historical context are discussed within social research (Hervieu-Léger 1998; Barker 2006; Inglis 2007). At least three types of Roman Catholics might be distinguished in Lithuania according to their religious practices. Active Catholics would make up about 13 percent of the population, attending Mass every Sunday or more frequently and going to confession several times a year. Formal Catholics would make up about 29 percent, attending Mass several times a year and during the major holidays and going to confession at least once a year. A third type would be nominal Catholics, who make up about 42 percent of the population, who attend Mass once a year or less often and do not go to confession (Laumenskaitė 2015, p. 186).

The identification with Roman Catholicism in Lithuania, as well as in other Catholic-majority countries, shows the interaction of religious identity with national identity and discloses the role the Roman Catholic Church played, and continues to play, in the history of the discussed country (Grzymala-Busse 2015). The Roman Catholic Church in Lithuania has played a crucial role in the process of nation building since the late 19th century, throughout the periods of resistance against the Russian empire and Soviet occupation (Streikus 2012). Thus, being Lithuanian is commonly equated to being Roman Catholic. While the majority of the Lithuanian population identify as Roman Catholics, every fifth representative of the population would identify as belonging to other Christian denominations, minority religions and non-believers (including atheists and agnostics) (Ališauskienė 2020).

The identification of most Lithuanians with Roman Catholicism is ambiguous for the above issues related to national identity and leaves questions about patterns of practiced religion on a daily basis. A similar problem of religious identification and its ambiguous relationship with religious practices has been creatively approached by the Pew Research Center, which in 2018 proposed a new typology for analysis of religiosity in the United States (Pew Research Center 2018). Three types and seven subtypes were distinguished. The type of Highly Religious included three subtypes of Sunday Stalwarts, God-and-Country Believers and Diversely Devout. The type of Somewhat Religious included two subtypes of Relaxed Religious and Spiritually Awake. The third type of Non-Religious, commonly referred to as Nones, included two subtypes of Religion Resisters and Solidly Secular. According to the authors of this typology, it allows one to find connections and differences between individuals with the same religious affiliation (Pew Research Center 2018).

Taking into consideration the above-discussed questions of religious identification, we decided in the analysis of the last Soviet generation’s religious values embedded in family practices to use a new typology, which would allow us to analyze the way religious or non-religious practices influence family practices among those who identify as Roman Catholics and those who are non-believers. Thus, in the empirical analysis, we used the following types: (1) devout religionist, identifies as Roman Catholic and participates in religious rituals at least two to three times a month (or more frequently); (2) somewhat devout religionist, identifies as Roman Catholic and participates in religious rituals at least once a month; (3) traditional religionist, identifies as Roman Catholic and participates in religious rituals several times a year; (4) cultural religionist, identifies as Roman Catholic and did not participate in religious rituals in the past 12 months; and (5) secularist, does not belong to any religious community and does not participate in religious rituals.

Results of the analysis of the dynamics of religious values in Lithuania since the 1990s, based on the European Values Study (EVS), show that, in the period of 1999–2008, there were no visible changes in the patterns of religiosity of the Lithuanian population, and the median Lithuanian was a lukewarm Catholic for whom religion and God were more important than irrelevant in life, who found peace or a certain source of strength in religion, attended church during major religious holidays and did not consistently profess traditional Christian doctrine but trusted the church, especially in its experience of responding to people’s spiritual needs and family problems (Žiliukaitė et al. 2016, p. 147).

The qualitative analysis of the religious values of the last Soviet generation showed that religious socialization during the Soviet period had diverse patterns and the role of religion in their everyday life had existing trends of religious individualism, critical engagement with institutional religiosity and conformity in religious identification with being a Roman Catholic. The research data also showed that education had effects on the role religion played in the lives of the last Soviet generation; less-educated informants placed more emphasis on the importance of religious values in their lives, while more educated ones critically engaged with religion’s role in their lives, those of previous generations and Lithuanian public life (Ališauskienė 2021).

3. Family Practices and Religious Identity

Gender division of labor in the family is substantially documented in the research literature, particularly in relation to the North American or Western and Northern European contexts. Overall, women’s time spent on household tasks decreased and men’s moderately increased over the last half of the century, but men still contribute substantially less (Sullivan et al. 2018). The time spent on childcare increased for mothers substantially, but also there is a small increase for fathers (Altintas and Sullivan 2017). These general long-term trends toward more gender egalitarianism in families are theorized as the second half of the gender revolution (Goldscheider et al. 2015). The shift is discussed as universal and irreversible, even though it might be delayed or stalled depending on the country context (Sullivan et al. 2018). Yet, some note that growing inequalities and social stratification of the family life might result in gender egalitarianism becoming the privilege of the higher social classes (Cherlin 2016).

In Central and Eastern European (CEE) countries, there is less systematic evidence on the long-term trends and factors contributing to the gender division in families. In the early 2000s, CEE countries manifested less gender equality in sharing household labor as compared with Northern and Western European countries (Aassve et al. 2014). Attitudinal surveys also show that CEE countries made progress in relation to gender equality, though they are still lagging (Buber-Ennser and Panova 2014; Wike et al. 2019). Lithuania clusters with the countries with more pronounced attitudes of gender traditionalism (Wike et al. 2019). Gender division of labor made some progress toward egalitarianism from the early 1990s; however, it has stagnated since the end of the first decade of the 2000s (Maslauskaitė 2022).

Several theoretical arguments explain the sharing of housework and childcare at the couple level. The relative resource perspective implies that partners with better employment and higher incomes will contribute less to household tasks (Davis and Greenstein 2020). The second perspective suggests that women are economically dependent and perform more, but non-normative roles lead to gender-deviance neutralization (Sullivan 2011). For example, high-earning women will still perform more household tasks in order to resolve the conflict between social expectations and their professional status. The third perspective shifts attention to gender ideology and implies that sharing of the tasks is determined by the way the partners perceive gender roles in family and society (Davis and Greenstein 2020).

Whether religiosity has an impact on family practices has been a subject of social research, although its scope is still limited in comparison with research into the way religiosity influences attitudes toward family (Read 2003; Goldscheider et al. 2014). Social research has been centered around gender differences as well as religious differences that influence the choice of family practices. Wilcox (2004) defined the phenomenon of soft patriarchy providing data about Evangelical Protestant fathers engaged in their family lives, who have incorporated various egalitarian practices in their everyday lives. Ammons and Edgell (2007) found that religious involvement and religious culture shaped trade-offs regarding family––work strategies particularly those of men. Pearce and Thornton (2007) found that families were socializing their children into their family ideologies that later shaped their family behaviors. The social research showed a positive relationship between religious fathers and their engagement in family practices. May and Reynolds (2018) found that conservative Protestant women reported less work-to-family conflict and less family-to-work conflict than their peers in other religious groups, even after controlling for religious service attendance, specific job features and sociodemographic characteristics. Meanwhile, according to May and Reynolds (2018), Catholic/Orthodox men reported less family-to-work conflict than conservative Protestant men.

Petts (2018) found that fathers who take leave and attend religious services were more likely to be involved with their children than fathers who take leave but do not attend religious services. Gull and Geist (2020) found that increased religious participation was associated with men’s greater participation in some housework tasks and time spent on housework, though the findings show great variation by task, religious tradition, level of religious attendance and across cultural zones. The research found that increased religious participation at the individual and cultural zone levels was associated with greater participation in some housework tasks and time spent on housework. It also indicated two potential paths leading to men’s increased housework participation: a non-religious, egalitarian one and a religious, family-centered one (Gull and Geist 2020).

4. Family Values, Practices and Gender Equality from State-Imposed to Lived Reality

In the Soviet Union, professed gender equality was a key component of government policy aimed at redefining male and female roles in both the public and private realms. Women’s emancipation was at the heart of this strategy, which aimed to make women equal to men. As a result, women were given paid jobs and political representation. Attempts by the government to reinforce the family by emphasizing women’s household responsibilities were attempted at the same period. Worker-mothers with responsibilities for employment, reproduction, and home obligations were described as women’s roles. Men served as “leaders, managers, soldiers, and workers”, and the Soviet Union functioned as a “universal patriarch” for men and women alike (Ashwin 2000, p. 1). Women became both paid employees and household laborers as a result of the Soviet state’s gender equality policy, working two daily shifts, one at work and one at home. Despite state-sponsored claims of women’s equality, women did not experience equality during the Soviet era.

Following the demise of the Soviet Union, neoliberal economic and political agendas called for the state to withdraw from citizens’ private lives, putting women at a disadvantage once more. The aftermath of the Soviet Union’s demise has been labeled “a post-socialist patriarchal renaissance” by some researchers. (Azhgikhina 2006 cit. in Kay 2007) These government-imposed obstacles for women have shaped women’s subservient status in present Lithuania.

Lithuanian society’s attitudes toward family and women remain patriarchal, although the changes within family life have been affected by cultural and social factors (Maslauskaitė 2010). Among cultural and social factors that influenced attitudes toward women’s rights in Lithuania, one might mention the Soviet gender equality policy that was implemented top to bottom in the country. During the time period between 1990 and1999, some changes occurred; however, this did not change the tendency toward greater implementation of gender equality in public life. Nevertheless, traditional gender roles within the family continue to be maintained in Lithuania (Kraniauskas 2009; Žiliukaitė et al. 2016).

Gender equality in Lithuania has been falling in the fields of labor, money, knowledge, power, and work–life balance, according to the Gender Equality Indexes calculated by the European Institute for Gender Equality (EIGE) from 2005 to 2017. Gender equality in Lithuania has reduced the greatest compared to other European Union countries, according to the 2019 Gender Equality Index (European Institute for Gender Equality 2020). These changes in the social constructions of gender roles in Lithuania reveal that the so-called gender equality imposed by Soviet authorities was strongly resisted by Lithuanian society, and skepticism regarding many aspects of gender equality ideals remains.

The research into the dynamics of family changes shows that generation change leads Lithuanian society from familialism—that family is the precious central unit of social order and that family should be governed by religious moral imperatives (Ammons and Edgell 2007)—toward individualism. Taking into consideration the above-discussed changes in the family values and the continuous resistance toward gender equality, we wanted to test the following hypotheses with empirical data from the last Soviet generation sample:

• Religiosity influences family practices in the division of housework and childcare: the more religious the individuals are, the more traditional and non-egalitarian family roles they sustain.
• Non-believers or secularists practice more egalitarian family practices manifesting in more equal childcare and housework division.

5. Data and Methods

Our analysis was based on the Families and Inequalities Survey (www.kartosirseimos.lt) dataset collected in 2019 in Lithuania. The population of the survey encompassed cohorts born between 1970 and 1984; the sample size was 3000 respondents. Thus, our data covered individuals who, at the time of the interview, were 35–49 years of age. The representative sample was obtained by using a stratified sampling method. Face-to-face interviews were carried out with respondents in their homes by using a standardized questionnaire. The survey recorded a wide range of themes related to the respondents’ parental home, lifestyle, partnership and fertility histories, parenting, household, well-being and employment conditions. Our analytical sample was restricted to the respondents living as a couple and having in the household at least one child younger than 14 years of age because only these respondents answered the questions on the childcare division in the family. Consequently, the working sample consisted of 1268 respondents.

5.1. Dependent Variable

The dependent variables in this study were the composite measures on childcare and housework. Childcare was measured on a six-item scale, which was presented to the respondents with children younger than 14 years and living in the same household. It included single items: (1) “dressing the children or seeing that they are properly dressed”; (2) “putting children in bed or taking care that they went to bed”; (4) “staying at home when the children are ill”; (5) “playing with the children and/or taking part in leisure activities with them”; (6) “helping children with homework”. The response categories were: 1 = “always me”; 2 = “usually me”; 3 = “me and my partner equally”; 4 = “usually my partner”; 5 = “always my partner”; 6 = “always or usually someone else”; 7 = “children themselves”.

Responses were recorded for each item taking into account the gender of the respondent, leading to a value of: 1, if the mother handles the childcare; 0, if mother and father share the task, it is done by children or someone else; and −1, if the father is responsible for the task. A composite summary index variable was created with values ranging from −5 to 5. Thus, −5 indicates that the father is 100 percent responsible for all childcare tasks, and +5 shows that mother 100 percent accomplishes the tasks. Thus, higher values indicate the more traditional gender division of childcare. Categories −5 to −1 comprised only 6.1 percent of the variance, and they were recorded as one category (−1).

Our second dependent variable was the division of housework. It was measured on a six-item scale. The items were: (1) = “preparing daily meals”; (2) = “washing up”; (3) = “grocery shopping”; (4) = “cleaning the house”; (5) = “doing laundry, ironing”; (6) = “paying household bills”. The response categories were identical to those in the childcare scale except for “children do it themselves”. The same procedure was applied, and a summary index variable was composed ranging from −6 to 6. As the values 1 to 6 comprised only 4.8 percent of the variance, they were recorded as one category (1). Again, higher values of the variable indicate that women are more responsible for performing housework.

Both variables were standardized for regression analysis.

5.2. Independent Variable

The main independent variable was religiosity. It is a composite variable created, which integrates self-identification with religion and frequency of participation in religious rituals. The variable on self-identification was restricted only to the Roman Catholics and those not belonging to any religious community. The selection of the Roman Catholics was motivated in the Lithuanian context as discussed previously and in order to secure more homogeneity related to religious tradition and doctrine. Eighty-three percent of all respondents in our sample declared to be Roman Catholics. Around 10 percent declared that they did not belong to any religious community. The second indicator considered was the frequency of the attendance of religious rituals (Mass) in the past 12 months. The main independent variable consisted of the following categories: (1) devout religionist, Catholic who participates in religious rituals at least 2–3 times a month (or more frequently); (2) somewhat devout religionist, Catholic who participates in religious rituals at least once a month; (3) traditional religionist, Catholic who participates in religious rituals several times a year; (4) cultural religionist, Catholic who did not participate in religious rituals in the past 12 months; (5) secularist, who does not belong to any religious community and does not participate in religious rituals.

Descriptive information of the main independent variable is presented in

Table 1. Descriptive sample statistics, percent and means (standard errors).

	Column Percent	Means (SD)
Childcare division (−1 to 5)		2.6 (1.9)
Housework division (−1 to 6)		3.4 (2.0)
Religiosity
Devout religionist	11.4
Somewhat devout religionist	11.4
Traditional religionist	60.4
Cultural religionist	8.5
Secularist	8.3
Relative incomes
R earns more than P	32.8
R earns less than P	45.8
R and P equally	21.4
R education
University	33.4
Semi-tertiary	29.6
Secondary or lower	36.9
Household income
1st quartile (lowest)	16.0
2nd quartile	23.2
3rd quartile	29.0
4th quartile (highest)	31.9
Women	59.2
Age of the youngest child living at home		10.3 (5.8)
Gender of children
At least one girl	52.8
No girls	47.2
Gender attitudes (traditional–egalitarian)		2.42 (0.53)
Living area
Urban	66.0
Rural	34.0
Number	1268

Source: Families and Inequalities Survey, 2019.

.

5.3. Control Variables

Several variables were included in the analysis as controls following the results reported in the previous research. First, we included socio-economic indicators such as relative incomes, education of the respondent and household incomes. Relative income is a subjective measure with the assessment of personal earnings in relation to a partner’s. The variable has several categories: (1) “I earn more than he/she”; (2) “He/she earns more”; (3) “We earn more or less equally”. Dummy variable was recorded for the regression analysis. Education was measured on the ISCED 2011 scale; three broad educational groups were recoded: university (1), semi-tertiary (2) and secondary or lower education (3). The semi-tertiary group included education in establishments which existed prior to 1991 and afterward were converted into higher schools (which existed until 2000) and later into colleges. Duration of studies was three to four years. The last educational category included the level corresponding to ISCED 0 to 4. We also considered household incomes, which were measured as the self-reported average household net income per month. We imputed the accidentally ascribed values to the cases where information on the incomes was not provided. As in other national surveys, the non-response to the income questions was high (around 30 percent of the total sample). For the regression analysis, we categorized the income variable splitting it into income quartiles.

Based on the existing theory and empirical evidence, we also included the indicator on the gender values. Index variable was composed out of the three items: (1) “whose responsibility is it to take care of children and home?”; (2) “whose responsibility is it to earn money for the family?”; (3) “who can take better care of small children?”. The response categories ranged from 1—“solely women”, 2—“possibly women”, 3—“doesn‘t depend on the gender”, 4—“possibly men”, 5—“solely men”. A summary index variable was created and summary value divided by 3.

In addition, we controlled for the family structure variables and socio-demographic characteristics. The analysis included the indicators of the age of the youngest child and gender composition of the children with two categories: (1) there is at least one girl younger than 14 years living in the family; (2) there are no girls in the family. Gender, age and the place of residence (urban or rural) of the respondent were also considered.

All descriptive statistics are provided in Table 1.

6. Results

Descriptive Analysis

Figure 1 presents the childcare and housework division in the couples by respondent’s religiosity. To recap, the higher values of both indices show that more work is performed by women in the family. We begin with the comparative mean analysis of the childcare and housework division by the religious groups (one-way ANOVA).

One can observe that childcare is less a solo burden of women within devout religionists and secularist groups; thus, the most religious and non-religious share a similar model of childcare task division, with women performing less and having a lower burden (Figure 1). Conversely, somewhat devout religionist, traditional religionist and cultural religionist groups manifest the childcare division where women have a larger burden and more solo responsibilities. Multiple comparison tests prove the statistically significant differences between the devout religionist group and other religious groups and no difference between devout religionists and secularists A similar trend is observed for the housework division in the families.

Thus, contrary to our expectations, the descriptive analysis shows non-linear relationships between religiosity and actual gender division of labor in the families. Those who are most religious—devout religionists—and non-religious—secularists—organize childcare and housework in a way that includes more task sharing and a lower burden for a woman, while among the somewhat devout religionists and traditional religionists, we observe the most unequal organization of household work.

In the next step of the analysis, we employed OLS regression with the main independent variable of religiosity and stepwise-added other control variables relevant for the study.

Table 2. Multivariate regression results for childcare task division (egalitarian–traditional), unstandardized b and standard errors.

	Model 1		Model 2		Model 3		Model 4
	b	(SE)	b	(SE)	b	(SE)	b	(SE)
Religiosity
Devout religionist
Somewhat devout religionist	1.08 ***	0.22	0.96 ***	0.22	0.80 ***	0.20	0.76 ***	0.20
Traditional religionist	0.63 ***	0.15	0.60 ***	0.16	0.51 ***	0.14	0.48 ***	0.14
Cultural religionist	0.74 ***	0.24	0.71 ***	0.24	0.59 **	0.22	0.54 **	0.23
Secularist	0.1	0.25	0.12	0.25	0.37	0.23	0.32	0.24
Relative incomes
R and P equally (ref.)
R earns more than P			0.50 ***	0.15	0.51 ***	0.15	0.055 **	0.18
R earns less than P			0.84 ***	0.14	0.40 ***	0.14	−0.02	0.35
R education
University (ref.)
Semi-tertiary			0.06	0.14	0.00	0.13	0.05	0.13
Secondary or lower			0.08	0.14	0.06	0.13	−0.01	0.13
Household incomes
4th quartile (highest) (ref.)
3rd quartile			−0.12	0.17	−0.004	0.18	0.00	0.12
2nd quartile			−0.07	0.15	0,04	0.14	0.06	0.13
1st quartile (lowest)			0.13	0.14	0.26	0.187	0.30	0.18
Women					0.62 ***	0.15	029	0.26
R’s age					0.01	0.01	0.01	0.01
Age of the youngest child					−0.03 **	0.01	−0.03	0.01 **
At least one girl in the family					0.15	0.10	0.15	0.10
Gender values (traditional–egalitarian)					−0.57 ***	0.09	−0.55 ***	0.09
Housework division					0.33 ***	0.02	0.33 ***	0.02
Urban					0.28 **	0.11	0.28 **	0.11
Women*earns more than P (ref.)
Women*earns less than P							0.7 *	0.42
Women*earns same as P							0.29	0.32
R2	0.02		0.06		0.24		0.25
Number	1043

Source: Families and Inequalities Survey, 2019. Note: *** p < 0.01, ** p < 0.05, * p < 0.1.

shows the multivariate regression results for the dependent childcare task division. Model 1 presents the baseline with only the main independent variable, i.e., religiosity. We see that, in comparison to the devout religionists, all other groups of religiosity except the secularists have statistically significant positive associations, meaning that the women’s burden in childcare is larger (higher positive values). In addition, the association for the non-religious is statistically not significant, and thus, compared with the most religious type of devout religionists, their childcare task division is shared similarly.

In the next step, we added the variables related to the couple’s relative incomes, respondent’s education and average household net incomes (Model 2). All associations for religiosity remain in place. In addition, we see that relative incomes are statistically significant in predicting childcare division, while the education and household incomes are not. In comparison with the couples with the equal income contribution of the partners, all other couples have less equal childcare division, with women performing more childcare tasks.

Model 3 includes additional control variables related to the respondent’s socio-demographics, age of the youngest child and gender composition of children. It also controls for the gender values and housework division of labor. Adding all the control variables significantly improves the goodness of fit of the model (R² = 0.24). In Model 3, coefficients related to childcare division and religiosity remain statistically significant even after including the additional controls, yet their values decreased, meaning that added control variables explain part of the variation. Compared with men, women report more workload related to childcare. Age of respondent is not aa significant determinant. As it could be expected, the older the youngest child in the family is, the more childcare tasks are shared or done by other family members than the mother alone. Having at least one girl in the family does not affect childcare division between the parents. We also observe that having more egalitarian gender attitudes leads to more sharing of the childcare tasks. In addition, housework division in the couple is positively related to childcare, meaning that a more traditional division of housework is associated with more childcare performed by the mother. Interestingly, we also found that urban respondents report more childcare work performed by the mother than rural.

Model 4, in addition to all the above mentioned controls, also includes the interaction effects between gender and relative incomes in the couple. We see that if a woman earns less than the partner, she has more solo burden of childcare compared with the couples where women earn more than the partner. Interestingly, for couples with income parity, the association is not statistically significant, meaning that childcare task division is similar in couples where women earn more or the same as the partner if all else is controlled for.

Table 3. Multivariate regression results for housework division (egalitarian–traditional), unstandardized b and standard errors.

	Model 1		Model 2		Model 3		Model 4
	b	(SE)	b	(SE)	b	(SE)	b	(SE)
Religiosity
Devout religionist
Somewhat devout religionist	0.40 *	0.24	0.30	0.24	−0.05	0.22	−0.11	0.22
Traditional religionist	0.30 *	0.17	0.22	0.17	0.01	0.16	−0.04	0.16
Cultural religionist	0.30	0.26	0.19	0.26	−0.03	0.24	−0.07	0.25
Secularist	−0.14	0.26	−0.21	0.28	−0.12	0.25	−0.07	0.26
Relative incomes
R and P equally (ref.)
R earns more than P			0.39 **	0.16	0.52 ***	0.17	0.43 *	0.20
R earns less than P			0.44 ***	0.15	−0.09	0.15	−1.35 ***	0.38
R education
University (ref.)
Semi-tertiary			0.49 ***	0.15	0.46 ***	0.14	0.45 ***	0.13
Secondary or lower			0.73 ***	0.14	0.70 ***	0.14	0.68 ***	0.14
Household incomes
4th quartile (highest) (ref.)
3rd quartile			−0.24	0.15	−0.19	0.13	−0.17	0.13
2nd quartile			-0.10	0.16	−0.09	1.52	−0.05	0.15
1st quartile			−0.27	0.21	−0.40 **	0.19	−0.32 *	0.19
Women					0.62 ***	0.17	0.14	0.28
R’s age					0.01	0.01	0.01	0.01
Age of the youngest child					0.04 ***	0.01	0.04 ***	0.01
At least one girl in the family					−0.24 **	0.11	−0.23 **	0.01
Gender values					−0.10	0.10	−0.05	0.10
Childcare division					0.4 ***	0.03	0.39 ***	0.03
Urban					−0.11	0.12	−0.11	0.12
Women*earns more than P (ref.)
Women*earns less than P							1.79 ***	0.42
Women*earns same as P							0.25	0.34
R2	0.06		0.04		0.21		0.22
Number	1043

Source: Families and Inequalities Survey, 2019. Note: *** p < 0.01, ** p < 0.05, * p < 0.1.

presents the OLS regression results for the housework division in the couples. We followed the same analytical logic as in the case of the childcare task index and model stepwise-adding the control variables. The baseline model (Model 1) controls only for religiosity, and we observe that the association between housework division and various groups of religiosity is statistically significant only for somewhat devout religionists and traditional religionists, meaning that these groups report more traditional housework division if compared with the devout religionists. However, the association between religiosity and housework task sharing turns out to be statistically insignificant in the following models (Models 2 to 4), meaning that religiosity does not explain how couples share housework.

Yet, we see that relative incomes and education are statistically significant predictors of the housework division in the couple (Model 2). Couples where one of the partners earns more have a more traditional gender division of housework compared with the couples with income parity. Those lower educated also have more traditional housework division (women perform more) compared with those university educated. The economic well-being of the family measured as the average household income does not contribute to the explanation of housework task division between partners.

Model 3 shows that all above-discussed effects (relative incomes, education) remain in place. However, we see that the economic well-being of the family is somewhat linked to the way the housework tasks are shared. Families with the lowest incomes have less traditional division compared with the highest-income group. Women report more unequal housework division. Interestingly, mothers’ solo burden is less pronounced in families with at least one girl compared with families where there are only boys. Age of the youngest child is positively associated;; thus, the older the youngest child is, the more housework chores are performed by mothers. The way couples organize childcare is interrelated with housework division, meaning that more traditional childcare is linked to the more traditional housework task distribution. Interestingly, we do not find any effect of the gender values. The place of residence also does not contribute to the model.

In the last step (Model 4),), we added the interaction terms between gender and relative incomes. We see that if a woman earns less than a partner, she also has the highest housework burden as compared with couples where women earn more than the partner. Housework division in couples where women earn the same as the partner is not different from couples where women earn more than the partner.

7. Discussion and Conclusions

This paper aimed to contribute to the discussions about the way religiosity impacts family practices by analyzing empirical data on the last Soviet generation (birth cohort 1970–1984) in contemporary Lithuania. We focused on Catholics, which are the dominant religious community in Lithuania, and on the division of labor in childcare and housework in families.

Existing literature suggests that developed societies are moving toward greater gender egalitarianism in the private sphere, though the pace of the transition depends on the context (Sullivan et al. 2018). Post-communist countries are lagging behind on gender equality attitudes and behavioral aspects of gender egalitarianism in families (Aassve et al. 2014; Wike et al. 2019). Analysis of previous research on the relationship between religion and divisions of labor in housework implied that differences emerge among North American and European societies as well as among societies dominated by different religious cultures such as Protestant, Catholic or Orthodox (Voicu et al. 2009). The latter two tend to create less-favorable cultures for gender equality in Europe. The research has shown an increasing involvement in family practices by more religious fathers providing a more family-centered path and a more egalitarian path by non-religious fathers (Gull and Geist 2020).

We expected that the intensity of religious practices would show differences in the family practices of this generation that has been living most of its life after the sociopolitical transformations and so experienced freedom of religious practice.

Our first hypothesis was that religiosity, i.e., Roman Catholic religious identification and frequency of religious practices, would influence family practices in childcare and housework divisions in such a way that more religious individuals would sustain more traditional and non-egalitarian family roles. First, we found that religiosity is linked to the way couples share childcare but not housework Second, the association between religiosity and childcare division is non-linear as it was assumed. The most religious group (devout religionists) manifest more equal division of childcare compared to other religious groups (somewhat devout religionists traditional and cultural religionists). Thus, among the religious, the more traditional childcare division is observed for those identifying as Roman Catholics but following religious practices nominally, i.e., attending religious services occasionally. Thus, our research, at least in relation to childcare, supports previous research findings (Gull and Geist 2020) stating that increased religious participation leads to greater gender equality in the private sphere.

Our second hypothesis that non-believers or secularists practice more egalitarian family practices was also confirmed partially. In relation to the sharing of childcare, secularists are more egalitarian compared to somewhat devout religionists, traditional religionists, and cultural religionists. Yet, they are similar to the most religious (devout religionists). However, in relation to housework, the secular identity does not contribute to the way tasks are shared between the partners.

The results of our analysis also allow questioning the level of impact religiosity/non-religiosity has on the actual division of housework. Devout religionists and secularists might be considered as social groups illustrating reflexive (non)religious identities. In our case, devout religionists and secularists represent social groups that have chosen their (non)religious identities reflectively, and this singles them out from the older generations. The personal choice to practice or not the religion might signal higher general reflectivity of the person, and this could also encourage critical assessment and re-evaluation of the normative patterns of gender division of childcare, roles of motherhood and fatherhood. We might theorize that these groups could be also more open to the cultural ideas of intense parenthood and new fatherhood, which implies a more active role of the father in childrearing and the trespassing of the economic role of the father.

On the other hand, religious identity does not contribute to the way couples share housework tasks. This realm of family relationships is socially organized depending on the partners’ earning parity and educational resources, and religiousness and assumed reflectivity are not at play. We found strong support for the argument of relative resources, meaning that women with lower earnings than the partner contribute more to housework compared to families where women earn more or about the same as the partner. Education lower than university also encourages less equal gender division of housework Interestingly, we found that “doing gender” in the family is transmitted between generations. Mothers have a lower burden of housework in families with daughter(s) compared to families with boys. This indicates that families remain the sites of intergenerational transmission of traditional gender arrangements.