Gender Equality in Sweden

Abstract

We focus in this article on contexts within which religiosity (the intensity of religious commitment) reinforces more patriarchal family values and those in which it does not, and perhaps even strengthens egalitarian family values. Using data for 1999 and 2003 from the Swedish longitudinal study, Young Adult Panel Survey, we examine the relationship between religiosity and several measures of attitudes and behaviors related to gender equality in the public and private spheres. We find for most religious denominations, greater religiosity is linked with more patriarchal views about the balance of men’s and women’s roles in the home. Among members of the Church of Sweden (formerly the Swedish State Church), however, this is not the case. No differences by religiosity were found in attitudes for gender equality in the public sphere of work.

Keywords

childcare gender role attitudes housework religiosity Sweden

Introduction

Nearly all religious systems, in both high- and low-fertility contexts, reinforce the importance of families in society and in peoples’ lives. Among individuals, then, the intensity of religious commitment normally reinforces familistic values and behaviors. In the past, familism has been linked with patriarchal gender relationships, that is, “traditional” familism (Davis & Greenstein, 2009). Hence, greater gender equality has been equated with secularization and individualism (Lesthaeghe, 1995). However, it seems likely that these relationships would only characterize the first stage of the gender revolution, in which women’s participation in the public sphere increased and their family involvement diminished, with no compensating increase in men’s contribution to the family (F. Goldscheider, 2012). The second half of the gender revolution, however, focuses on men’s participation in the life of the home, potentially strengthening the family. In the context of the second half of the gender revolution, then, the critical question becomes: Will the greater emphasis on families linked with religiosity and religious involvement break this link between religiosity and patriarchal familism for societies in the second half of the gender revolution? And perhaps lead to more rather than less egalitarian families?

In this research, we specify contexts within which religiosity (the intensity of religious commitment) might reinforce more patriarchal familism and contexts within which it might not, and perhaps even strengthen egalitarian familism. We indicate these contexts using religious community membership (religious denominational affiliation). We will focus on the relationships between religiosity and gender role attitudes and behaviors among young adults in Sweden. We examine attitudes focused on gender roles in both the public sphere of work and the private sphere of the family, as well as those on the gender balance of work and family, and also examine actual gendered behavior in terms of housework and childcare. Overall, we expect that religiosity will decrease egalitarianism among young adults in all but the dominant religious denomination in Sweden—what for centuries (until 2000) was the Church of Sweden. Among members of the Swedish church, we hypothesize that religiosity will have less effect, and perhaps even be linked with increased egalitarianism. Hence, among denominations where family values are traditional/patriarchal, that is, contexts outside the Church of Sweden that reinforce gender hierarchy, those who are more religious should have more “traditional” gender attitudes and behavior than the less religious. In contrast, in contexts/denominations where family values emphasize greater male involvement in the home, as in Sweden, this should be less the case. The more religious members of the Church of Sweden might even be more egalitarian in their gender roles and attitudes than the less religious. Thus, we argue that the relationship between religion/religiosity and gender equality is shaped by specific contexts. Whereas once the growth of egalitarianism was conceived to be “antifamily” by encouraging women to leave their domestic roles to seek equality with men in the public sphere, it may well be that egalitarianism becomes more “profamily” by encouraging men’s involvement in the family.

Background

Religion is one of the core (and perhaps one of the few remaining) institutions that support families, at least in countries with no family policies, such as the United States. Hence, it is important to incorporate considerations of religion, not only religious affiliation but also religious involvement and particularly religiosity, into the study of family change (Ammons & Edgell, 2007). Families have been changing rapidly across the industrialized (and even the industrializing) worlds (Cherlin, 2012), and few studies of religion’s role in this process have been undertaken (Ammons & Edgell, 2007).

Beginning with the rapid increase in the labor force participation of women, and particularly of married mothers (Bianchi & Spain, 1986), families have been under pressure, as women have withdrawn portions of their time, energy, and focus from their families by taking on paid employment to provide them with more financial resources. They have thereby challenged the “male breadwinner, wife caregiver” family configuration. Following this major structural change in the lives of women and in the relationships between men and women, other challenges have arisen to the traditional, stable, nuclear family of the 1950s (Goode, 1963), most involving decreases in relationship commitment, including the growth in divorce and in cohabitation. Young adults have responded to these challenges by delaying forming families and making a commitment to family roles as partners and parents, a process that began even before the current economic disruptions to their lives.

The few studies that have attempted to link religion to these new family-related behaviors have generally found that in many contexts, the more religious have resisted these changes, particularly in the United States. Pearce and Thornton (2007) found that young adults who rated religion as more important in their lives were more supportive of the breadwinner/caregiver configuration for men and women. Davis and Greenstein (2009), in their extensive review of the predictors of gender ideology, found religiosity to be normally linked with greater gender role traditionalism. However, the picture is not totally clear; one study found no relationship between religiosity and gender role traditionalism (Davis, 2007), while another (Ammons & Edgell, 2007) found that greater religious involvement was linked with men (but not women) giving greater priority to family over work in their lives.

Nevertheless, the link between religion and a wide range of traditional family patterns has been well documented. There is an extensive literature showing the positive relationship between religion and fertility, both historically and in contemporary Western societies (see Adsera, 2007; Derosas & van Poppel, 2006; Lehrer, 2009; McQuillan, 2004). This result appeared in the earliest studies of fertility in the United States (see various reviews in, among others, Frejka & Westoff, 2008; C. Goldscheider, 2006; Mosher, Williams, & Johnson, 1992), which all find that greater religiosity, however measured, increases fertility, and, assuming maternal childcare, reinforces patriarchal familism. Similarly, greater religiosity has been associated with greater union stability (Lehrer, 2009) and with disapproval of divorce, cohabitation, and premarital sex (Pearce & Thornton, 2007).

In Europe, a major concern over the growth in the numbers of immigrants from predominately Moslem countries has focused on the gender traditionalism they have often brought with them along with their Moslem faith (e.g., Diehl, Koenig, & Ruckdeschel, 2009), a concern that also emerges in research on the United States (Read, 2003). This difference, however, may be short-lived. Studies of immigrants, particularly of Moslems in a number of destination countries, have shown that immigration to Western countries has increased the labor force participation of women and hence, shaped more equal relationships between men and women in households (see the review in Donato, Gabaccia, Holdaway, Manalansan, & Pessar, 2006).

Our analysis of the potentially variable relationship between religion and gender roles affirms the importance of examining variation at the level of community. In this case, we examine religious institutions, which we use to distinguish core Swedish communities versus other communities resident in Sweden. We move beyond the characteristics of individuals to understand them as embedded not only in family but also in community contexts.

A focus on family and household units is the most direct way of approximating the links between the individual and community (see Bernhardt, Goldscheider, Goldscheider, & Bjerén, 2007). Furthermore, our fundamental argument is that context matters; that religiosity will take on different meanings in different contexts (even as it is likely to reinforce familism, broadly defined). Research has shown that, for example, Moslems have had much less difficulty in both maintaining high levels of religiosity and integrating socioeconomically in contexts where the resident population is religiously heterogeneous (like the United States and Canada) than in Europe, which has been much more religiously homogeneous (Alba, Raboteau, & DeWind, 2009; Foner & Alba, 2008).

These studies, both of religiosity generally and Islam in particular, however, primarily reflect the context of the first half of the gender revolution, the half in which women move into the public sphere. In the context of the second half of the gender revolution, in which men increasingly come to share the private sphere with women, in contrast, a few recent studies have found that greater religiosity (usually measured crudely by church attendance) increases the involvement of fathers in family activities in the United States. These activities include increased male involvement in one-on-one activities, family dinners, and youth activities (Wilcox, 2002), participation as coparents in their children’s early education programs (Roggman, Boyce, & Cook, 2002), and making profamily trade-offs with work (Ammons & Edgell, 2007). These patterns in the second half of the gender revolution suggest that by reinforcing men’s roles in families, religiosity might be contributing to less patriarchal, more egalitarian families.

This suggests that greater religiosity might have very different relationships with gender roles, whether measured as attitudes or behavior, depending on gender equality in which of the two major spheres—women’s activities in the public sphere or men’s in the private sphere—is being considered. Scales constructed based on attitudes toward whether women should work (and perhaps risk harming their children) clearly focus on first half issues; those constructed using attitudes toward whether men should give priority to their jobs or their families raise the core issues of the second half, often with very different relationships with family-related attitudes and behavior (F. Goldscheider, Oláh, & Puur, 2010).

Sweden provides a useful context for such a study, a country in the forefront of the ongoing gender revolution. At least among the Swedish-origin population, religiosity might increase gender egalitarianism in the context of the second half of the gender revolution, given that Swedish policy has been to strongly reinforce men’s family roles (Duvander, 2012).

Furthermore, as a country of recent immigration, Sweden has experienced religious heterogeneity for the first time, providing enough cases of individuals for whom religiosity would decrease egalitarian gender roles, such as those from countries of greater traditionalism in terms of gender roles, such as Moslems from Turkey. As have many other European countries, Sweden experienced rapid increases in immigration over the last decades of the 20th century. The new immigration is not only larger in volume than in earlier periods but also originates from more culturally distant areas, such as the Middle East, Asia, and Africa, increasing the diversity of Sweden’s population. While the total Swedish-born population increased by less than 5% between 1970 and 1999, the foreign-born population increased by 82%, so that about one in nine persons living in Sweden at the end of the 20th century was foreign born (Bernhardt et al., 2007). Specifically, our question in this context is: How does religiosity and its connection to familism operate in a general context of the egalitarianism that characterizes Sweden, and do these relationships differ between those engaged in strongly Swedish cultural contexts such as the Church of Sweden, and those in less strongly Swedish contexts, such as other religious institutions (or none)?

We will examine several aspects of attitudes about gender role egalitarianism, including those about gender roles at home, at work, and how to balance the two, as well as the actual sharing of housework and childcare. We will test the following hypotheses, separating our expectations of the impact of religiosity in both egalitarian and traditional contexts:

Hypothesis 1: Greater religiosity will reduce respondents’ likelihood of expressing egalitarian attitudes about work–family balance for most religious groups.

Hypothesis 2: However, this should be much less the case among the Swedish majority group, those who are members of the Church of Sweden, for whom higher levels of religiosity may even increase egalitarian attitudes, by increasing attitudes that support men’s involvement in the home.

Hypothesis 3: Finally, although we do not test this directly, we expect that the relationships expected in Hypotheses 1 and 2 should be stronger for attitudes and behaviors focused on the home than for those focused on the public sphere.

Data and Method

Data

This analysis uses the Swedish Young Adult Panel Survey. This data set has been designed to enable studies of the complex interrelationships between attitudes and demographic behavior (see www.suda.su.se/yaps for more information). The database is a unique combination of population register and survey data, with the Survey Unit of Statistics Sweden in charge of the field work. Designed from the beginning to be longitudinal, the first two waves of data collection were carried out in 1999 and 2003, which provide the measures for this analysis. A third wave was completed in 2009, but it contained fewer cases and less heterogeneity, because those of Turkish origin were not resurveyed. Hence, we analyze the links between religiosity in the context of religious communities and gender role attitudes collected in 1999, as well as domestic sharing behavior (housework and childcare) in 2003.

Measures

We have five dependent variables, three focusing on gender role attitudes and two on behavior. Two of the attitudes variables are indices based on several questions (attitudes toward gender roles in the private and public spheres), which result from a factor analysis of six questions. We have analyzed responses of agreement/disagreement recorded on the 5-point Likert-type scale (where 1 corresponds to don’t agree at all, 5 to agree completely, and 6 to don’t know, which was recoded to a value of 3) with the following statements:

A society where men and women are equal is a good society.

Men can do as well as women in caring jobs.

Women can do as well as men in technical jobs.

It is as important for a woman as for a man to support herself.

The woman should take the main responsibility for housework.

The man should take the main responsibility for support.

Our measure of attitudes toward gender balance in the private sphere is based on items e and f, which together make clear that housework is women’s responsibility while men do not have time for it. Given the very high level of gender equality responses for all these items, only those reporting strong disagreement for both items were considered as “egalitarian” (coded as 1), while the other responses were considered as nonegalitarian (coded as 0). Among women, 71% described themselves as egalitarian on this measure; the level was somewhat lower for men (53%). The distributions on this measure and the variables described below appear in Table 1.

Table 1.

Means and Percentages for Independent and Dependent Variables, by Sex.

	All	Men	Women
Variables	$\bar{X}$ or %	$\bar{X}$ or %	$\bar{X}$ or %
Egalitarian combined sphere attitude (2003)	77.2	75.3	78.7
Egalitarian public sphere attitude (2003)	57.4	50.4	62.8
Egalitarian private sphere attitude (2003)	63.2	53.3	71.0
Egalitarian share of housework (2003)	57.6	61.6	54.7
Egalitarian share of childcare (2003)	74.0	78.0	66.2
Member, Swedish State Church (1999)	65.6	63.5	67.3
Importance of religion (2003)	0.232	0.191	0.264
Little or no importance	83.0	86.3	80.4
Rather important	10.9	8.3	12.9
Very important	6.1	5.4	6.7
Female (2003)	56.30	NA	NA
Civil status (2003)
Not in a coresidential union	29.8	33.1	27.3
Cohabiting	43.3	42.7	43.8
Married	26.8	24.2	28.9
Parent (2003)	46.2	40.6	50.6
Age (2003)	29.7	29.8	29.6
Educational level (2003): comp. = 1, postgrad = 5	3.6	3.5	3.6
Grew up in a two-biological parent family (2003)	74.4	74.2	74.5
Number of cases	2,052	897	1,155

Similarly, our measure of attitudes toward gender balance in the public sphere is based on strong agreement on Items a to d. Those who strongly agreed with all four items were considered as “egalitarian” (coded as 1), with the rest of responses considered as “nonegalitarian” (coded as 0). Interestingly, the level of egalitarian attitudes on this indicator was somewhat less than for private sphere attitudes, perhaps reflecting the very strong support by the Swedish government for equal sharing at home (and the high level of gender segregation in employment).

Our measure of attitudes toward work–family balance is based on a question about “the ideal family situation for a family with preschool children.” It has four response alternatives: (1) only the man works and the woman takes the main responsibility for home and children; (2) both work, but the woman works part-time and takes the main responsibility for home and children; (3) both parents work roughly the same hours and share the responsibility for home and children equally; and (4) the woman works more than the man, who takes the main responsibility for home and children. For this analysis, the first two were combined and labeled “traditional” gender roles, while the few in the fourth category were combined with those in the third category and labeled “egalitarian.”

In addition to these three attitude variables, we also examine two measures of actual sharing of housework and childcare in 2003, 4 years after the attitudinal information was collected. Our measures of actual sharing derive from the answers to two separate questions about how the respondents shared housework and childcare with their partners, each with three responses: (1) I do more, (2) We share equally, and (3) My partner does more. For the few cases in which the male partner did the majority of housework/childcare, we grouped these couples with those reporting that they shared these chores equally. In our analytic sample, 58% reported that they shared housework equally, while somewhat more (74%) reported that they shared childcare equally. This appears to be a common pattern in Europe (Esping-Andersen, 2009). This allows us to test the hypothesis that the ways religiosity and denominational affiliation shape gender roles affect not only the attitudinal and normative but also affect the actual sharing of both housework and childcare. For actual housework and childcare, men report higher levels of equal sharing than women, unlike the case for attitudes.

For our key independent variable, our study will explore the question of religiosity directly rather than infer it from religious affiliation. First, respondents were asked at the original survey (1999) about whether they belonged to a religious denomination and, if so, which: Church of Sweden, “free” church, Catholic, Orthodox, Moslem, Jewish, and “other.” (We note that the Lutheran church, which for centuries was supported by the Swedish state and is referred to in Swedish as “Svenska kyrkan,” was deinstitutionalized in 2000; we will refer to it as the “Church of Sweden.”) Nearly two thirds of the respondents reported belonging to the Church of Sweden. More than half of the remainder reported either no affiliation or refused to answer; most of the rest reported belonging either to the “Free Church” (an evangelistic congregation), the Catholic (Western) or Orthodox (Eastern) church or to mosques. Very small numbers listed Jewish or other affiliations.

In addition, they were asked to specify the importance of religion in their lives, with three levels: very important (6%), quite important (11%), and of little or no importance (83%). As we expect that these terms may have different meanings within denomination, and possibly by ethnic origin, we began by examining subjective religiosity by denomination and ethnic origin, in order to create a parsimonious variable that is likely to capture most of the variation in family-related attitudes and behavior.

Based on respondents’ answers to the questions on the importance of religion in their lives, their religious group membership, and their ethnic origin, we originally created six categories. We considered the numbers of available cases, theoretically appropriate combinations, and respondents’ views on the ideal family configuration. We combined those for whom religion was “somewhat important” with those for whom it had little or no importance in their lives, because for every religious membership group with more than a handful of cases, the “somewhat important” group more strongly resembled those who felt religion to have little or no importance. We also combined members of the Free Church and the Catholic Church, which we call “other Western Christian churches,” those with membership in other traditional religious groups (Moslems, Orthodox Christians, and “other”), distinguishing by whether they were of Turkish origin (most of the Moslems) or of other origins. We combined those respondents with either no religious affiliation or missing religion (whether of Turkish or non-Turkish origins), together with other small groups.

Our multivariate analyses also include crucial controls likely to be related both to religiosity and to our outcome measures of family-related values and behaviors, the inclusion of which reduces, although certainly does not eliminate, concerns that the relationships we observe are spurious. These are the gender of the respondent (whose religiosity is being measured); the extent of commitment to their partner relationship (if any), distinguishing those who were married, cohabiting, or living outside a coresidential relationship in 1999; and whether they already have a child. We also control age (distinguishing the three cohorts who were aged 22, 26, and 30 years in 1999), educational level, and whether they grew up in a two-biological parent family.

Method

We examine the differing relationships by denominational contexts between religiosity and a series of outcome indicators of gender role attitudes and behavior, using cross-tabulations and multivariate regression. We examine their relationships both with egalitarian attitudes (expressed in 1999 and 2003) and with actual sharing of housework and childcare reported in 2003. For the multivariate analyses, as we have constructed the attitudinal and actual sharing variables as dichotomies (egalitarian, nonegalitarian), we use multivariate logistic regression.

Results

Descriptive Analysis of Work–Family Balance, 1999

As an overall portrait of the relationship among religiosity, church affiliation, and egalitarian attitudes, we show, using as an example the question on the ideal balance of paid and domestic work when children are young, how these groupings differ in the percentage expressing egalitarian attitudes in 1999 by denomination and religiosity (Table 2).

Table 2.

Religiosity in Context and Attitudes Toward Work–Family Balance^a.

Denomination and religiosity	Percentage egalitarian
Total	75.0
Swedish State Church, religion very important	81.5
Swedish State Church, religion less important	76.7
Other Western Christian church,^b religion very important	52.5
Other Western Christian church,^b religion less important	71.8
Turkish-origin religious groups^c	48.0
Other^c	76.0

1999 work–family balance = shared equally or he does more in the home.

Free church, Catholic church.

Moslem, Orthodox, other, no or missing religiosity.

Overall, the most striking pattern is that young Swedes of whatever denomination and level of religiosity are very likely to express egalitarian attitudes about the ideal balance between men and women in the responsibility for earning and home care. Fully three quarters of the young adults express egalitarian attitudes on this question (slightly less than their level on this question in 2003), and every group has close to half or more who do so. Considerably more of these young adults expressed egalitarian attitudes among members of the Church of Sweden and those who expressed some “other” religious affiliation (most of whom said “no affiliation” or refused to answer) than was the case for members of “Other Western Christian” churches or Turkish-origin traditional religious groups.

Furthermore, members of the Church of Sweden and those belonging to other Western Christian churches differed in the effect of having expressed great importance for religion. Among members of the Church of Sweden, those for whom religion is very important are five percentage points more likely to express egalitarian attitudes than those for whom religion is less important (82% egalitarian for those who feel religion is very important compared with 77% for those who find it less important). In contrast, members of other Western Christian churches were less likely to express egalitarian attitudes if they felt religion to be very important in their lives than otherwise similar members of other Western Christian churches who felt religion to be less than very important. In this case, however, the difference was 20 percentage points (72% of those for whom religion was not very important were egalitarian compared with 52% of those for whom it was very important), suggesting that even these small differences in the opposite direction among members of the Church of Sweden are likely to be consequential.

These findings are consistent with our first hypothesis that among those most likely to be concerned with egalitarian homes (members of the Church of Sweden), it is the most religious who are more egalitarian, while among others, religious attitudes are linked with less egalitarianism. However, we do not yet control for other characteristics of these respondents that might influence egalitarian attitudes, and that are likely to distinguish these groups (e.g., the higher education of members of the Church of Sweden relative to others). Furthermore, we need to examine more dimensions of egalitarian attitudes, as well as egalitarian home-based behaviors, both in terms of housework and childcare.

Multivariate Analysis of Domestic Sharing

We analyzed domestic sharing in a multivariate model using the 2003 versions of questions on gender-related attitudes and behaviors. For attitudes, these focused on gender roles in the public and private spheres, together with appropriate ways for combining them, based on indices for all respondents in 2003; for behavior, these focused on actually sharing housework in 2003 for the subset of respondents who were in a coresidential relationship and actually sharing childcare for the even smaller subset who were parents in a coresidential relationship. We simplified the groups, dividing those who were members of the Church of Sweden and all others, and included in the analysis the measure of the importance of religion and the full set of controls. For each analysis, we include, in addition to these variables, an interaction between the importance of religion and membership in the Church of Sweden, in order to test our hypothesis. These results are presented in Table 3.

Table 3.

Logistic Regression of Religiosity and Religious Affiliation on Three Measures of Attitudes Toward Gender Roles and Two Measures of Actual Behavior (Odds Ratios).

Variables	Public sphere attitude 2003	Private sphere attitude 2003	Combined spheres attitude 2003	Report share housework 2003	Report share childcare 2003
Importance of religion	0.80*	0.53***	0.54***	0.80	0.95
Swedish Church member	0.93	1.03	1.02	0.96	1.12
Interaction: Importance of religion × Swedish Church member	1.09	1.63*	1.61*	1.17^†	0.84
Female	1.66***	2.26***	1.33*	0.80*	0.69*
Civil status (cohabiting = ref.)
Not in a coresidential union	0.84	0.82(*)	1.15	NA	NA
Married	0.98	1.03	0.96	1.05	1.29(*)
Age	0.96	1.16*	1.04	0.96	0.99
Educational level	1.24***	1.40***	1.29***	1.20***	0.99
Grew up in a two–biological parent family	1.01	0.78*	0.92	1.00	0.87
Child in household	0.88	0.67**	0.65***	0.46***	NA
Constant	0.50**	0.31***	1.25	1.24	1.77
Number of cases	1,997	2,006	1,998	1,410	828

(*)

p < .10. *p < .05. **p < .01. ***p < .001, significantly different from 0 (reference category). ^†Significantly different (p < .05) for married only.

We begin with the measure of attitudes toward gender equality in the public sphere of work, which was the focus of most gender role research in the early decades of the gender revolution (when advertisements for jobs routinely indicated whether the job was for men or for women). Not surprisingly, consistent with research from this era, those who describe religion as very important to them are significantly less likely to endorse gender equality (odds ratio [OR] = 0.8). Denominational membership has no significant relationship (OR = 0.93), nor is there a significant interaction between the relationships by denominational membership (OR = 1.09).

This pattern changes completely, however, when results focus on considerations of men’s role in the private sphere. The results are quite similar for the measure of attitudes toward gender equality in the private sphere (column 2) and that in column 3 focusing on the ideal gender balance in the public and private spheres (“combined sphere attitudes”). In each case, we see that those reporting that religion is highly important to them are significantly less likely to express egalitarian attitudes (OR = 0.53, OR = 0.54). Membership in the Church of Sweden continues to have no relationship with this or any of our other gender roles measures. However, members of the Church of Sweden for whom religion is extremely important do not show the same pattern as that for the other respondents, with a significant interaction coefficient that nearly totally offsets the negative effect of religiosity (OR = 1.63, OR = 1.61).

Religiosity has a much weaker relationship with actual sharing of housework and childcare, however, than it does with attitudes. This might reflect in part the smaller number of respondents who answered the question, as those who were not living in a coresidential relationship were not asked the question on sharing housework, and even more had not become parents and hence were not asked the question on sharing child care. It also likely reflects that there are many other influences on one’s behavior than one’s attitudes (including the feelings of one’s partner). Furthermore, there is no significant interaction between religiosity and membership in the Church of Sweden for actual housework sharing, or for actual childcare sharing. Interestingly, however, there was a significant interaction between membership in the Church of Sweden and religiosity in the relationship between religiosity and sharing housework among married respondents. This seems likely to reflect the greater attitudinal similarity among the married than among those cohabiting, as most studies show the married to be more traditional on gender relationships than cohabiting couples.

Turning to the effects of the controls, most results were as expected. Women hold considerably more egalitarian attitudes vis-à-vis their roles in the public sphere of work, the private sphere of the family, as well as the combined sphere (ideal situation of a couple with a small child). However, when it comes to reporting actual sharing of housework and childcare, women are less likely to report that their households are egalitarian than men are, a commonly found result that suggests that men are more satisfied (and women less satisfied) by their efforts at taking on these responsibilities.

The other consistent findings are those for education and for whether a child is present in the household. The more educated report significantly more egalitarian attitudes for each of the three attitude measures, as well as greater sharing of housework (but not childcare). Those with a child consistently report less egalitarian attitudes, significantly for those on the combined sphere and the private sphere, and are much less likely to share housework. Civil status, meaning whether the respondent is in a coresidential union and if so, whether married or cohabiting, has little relationship with these attitudes and behaviors. There is no impact of childhood family structure; the only effect of age is to weakly increase attitudes toward sharing domestic tasks (but not actual sharing).

Concluding Discussion

The data we have analyzed provide two results that have significance for research on religiosity and family outcomes. First and foremost is the general indication that religiosity continues to have important influence on issues of gender role attitudes and family values even in the secular national context of Sweden. This finding appears even after controlling for the effects of education and related factors. Religion remains a salient feature of attitudes toward the balance of work and family among men and women. Studies of familism, family values, and gender issues, therefore, need to consider the continuing power of religious affiliation and religiosity.

Second and of striking importance are the ways in which contexts shape the relationship between religiosity and gender role attitudes and behavior. We have shown that this relationship is not uniform, but is shaped in part by denominational context. Among those who are not members of the Church of Sweden, those expressing greater religiosity are less likely to hold egalitarian attitudes on measures that include a consideration of men’s engagement in the tasks of the home; among members of the Church of Sweden, religiosity has little or no relationship with gender role attitudes. Hence, in the first stage of the gender revolution in which women’s participation in the public sphere increased and family dependence diminished, the familism associated with religiosity meant greater emphasis on traditional, gendered relationships, reinforcing the importance of women’s role in the home. However, with the gender revolution increasingly focusing on the family, encouraging men to participate in the life of the home, religion’s emphasis on strengthening families appears to no longer promote patriarchal families, and might even lead to more rather than less egalitarian families. Hence, among many young adult Swedes, those of immigrant origins or who are not members of the Church of Sweden, religiosity is linked with greater but traditional familism. Among the members of the Church of Sweden, however, the most religious are as egalitarian in their familism as the less religious.

Our results are limited by the relatively small number of cases available for the analysis, and by the very high levels of egalitarian attitudes and behavior among Swedish young adults. As well, it is difficult to make causal conclusions, given that unobserved heterogeneity might account for the relationships we observe. Members of the Church of Sweden who express feelings that religion is very important to them might differ in other ways that shape their egalitarian attitudes, although by controlling for education and family context, we reduce some of this concern.

Nevertheless, the analysis has great strengths, in pointing to new directions for the analysis of religiosity and gender issues in the context of the second half of the gender revolution. For too long, religion has been linked with traditional, patriarchal family behaviors, and this research suggests that this link is not a necessary one. These findings are likely to also apply to other countries that are well into the gender revolution, particularly to selected segments, such as the more educated.

These findings need to be further explored using some additional familism measures and incorporating more complex indicators of religiosity. It is nevertheless clear that there are no universal relationships between religiosity and familism. Even in Sweden with its overall secular culture and egalitarian policies, religiosity is a factor not to be dismissed in understanding familism in terms of gender role attitudes and gender. The impact of institutional context on both religiosity and familism in the emergent forms of the gender revolution needs to be explored further.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by the Swedish Research Council for Working Life and Social Research (Grant No. 2008-0489).

References

Adsera

(2007). Reply to the note by Newman “Is fertility indeed related to religiosity?” Population Studies, 61, 225-230.

Alba

R. A.

Raboteau

DeWind

(Eds.). (2009). Immigrants and religion in America: Comparative and historical perspectives. New York: New York University Press.

Ammons

Edgell

(2007). Religious influences on work-family trade-offs. Journal of Family Issues, 28, 794-826.

Bernhardt

Goldscheider

Bjéren

(Eds.). (2007). Immigration, gender, and family transitions to adulthood in Sweden. Lanham, MD: University Press of America.

Bianchi

Spain

(1986). American women in transition. New York, NY: Russell Sage Foundation.

Cherlin

(2012). Goode’s world revolution and family patterns: A reconsideration at fifty years. Population and Development Review, 38, 577-607.

Davis

(2007). Gender ideology construction from adolescence to young adulthood. Social Science Research, 36, 1021-1041.

Davis

Greenstein

(2009). Gender ideology: Components, predictors, and consequences. Annual Review of Sociology, 35, 87-105.

Derosas

van Poppel

(Eds.). (2006). Religion and the decline of fertility in the Western world. Dordrecht, Netherlands: Springer.

10.

Diehl

Koenig

Ruckdeschel

(2009). Religiosity and gender equality: Comparing natives and Muslim migrants in Germany. Ethnic and Racial Studies, 32, 278-301.

11.

Donato

Gabaccia

Holdaway

Manalansan

Pessar

(2006). A glass half full? Gender in migration studies. International Migration Review, 40, 3-26.

12.

Duvander

(2012). What are the effects of reforms promoting fathers’ parental leave use? Journal of European Social Policy, 22, 319-330.

13.

Esping-Andersen

(2009). The incomplete revolution: Adapting to women’s new roles. Cambridge, England: Polity Press.

14.

Foner

Alba

(2008). Immigrant religion in the U.S. and Western Europe: Bridge or barrier to inclusion? International Migration Review, 42, 360-392.

15.

Frejka

Westoff

(2008). Religion, religiousness and fertility in the US and in Europe. European Journal of Population, 24, 5-31.

16.

Goldscheider

(2006). Religion, family, and fertility: What do we know historically and comparatively? In Derosas

van Poppel

(Eds.), Religion and the decline of fertility in the Western world (pp. 41-57). Dordrecht, Netherlands: Springer.

17.

Goldscheider

(2012, June 13). The gender revolution and the second demographic transition: Understanding recent family trends in industrialized societies. Keynote address at the 2012 European Population Conference, Stockholm, Sweden.

18.

Goldscheider

Oláh

Puur

(2010). Reconciling studies of men’s gender attitudes and fertility: Response to Westoff & Higgins. Demographic Research, 20(8), 188-197.

19.

Goode

(1963). World revolution and family patterns. New York, NY: Free Press.

20.

Lehrer

(2009). Religion, economics and demography: The effects of religion of education, work and the family. London, England: Routledge.

21.

Lesthaeghe

(1995). The second demographic transition in western countries: An interpretation. In Mason

K. O.

Jensen

A.-M.

(Eds.), Gender and family change in industrialized countries (pp. 17-62). Oxford, England: Oxford University Press.

22.

McQuillan

(2004). When does religion influence fertility? Population and Development Review, 30, 25-56.

23.

Mosher

Williams

Johnson

(1992). Religion and fertility in the United States: New patterns. Demography, 29, 199-214.

24.

Pearce

Thornton

(2007). Religious identity and family ideologies in the transition to adulthood. Journal of Marriage and Family, 69, 1227-1243.

25.

Read

J. G.

(2003). The sources of gender role attitudes among Christian and Muslim Arab-American women. Sociology of Religion, 64, 207-222.

26.

Roggman

Boyce

Cook

(2002). Getting dads involved: Predictors of father involvement in Early Head Start and with their children. Infant Mental Health Journal, 23, 62-78.

27.

Wilcox

(2002). Religion, convention, and paternal involvement. Journal of Marriage and Family, 64, 780-792.