0social compass
56(2), 2009, 144-162
Malina VOICU
Religion and Gender across Europe
European societies have experienced a decrease in the social importance of religious issues. Values and attitudes towards gender roles have also changed in the last decades. In European countries, people have become more egalitarian with respect to the position of women in society. The author tries to identify the relationship between secularization and changes in gender values. As a result of secularization, the individual value system has become fragmented and religious values have lost their coordinating role. The investigation employs cross-national and longitudinal analysis of European Values Survey data (1990, 1999), most of the European countries being included in the study. The results indicate the decreasing impact of traditional religious belief on values related to gender roles during the 1990s in Europe and a common pattern of relationships between gender values and religiosity in most European societies.
Key words: religiosity • gender values • secularization • European Values Survey
L'importance .sociale de la religion a diminué dans les sociétés européennes. Les valeurs et les attitudes de la population ont aussi évolué durant les dernières décennies. Dans tous les pays européens, la population est devenue plus égalitaire en ce qui concerne la position des femmes dans la société. L'auteure essaie d'identifier la relation entre la sécularisation et le changement de valeurs des genres. Du fait de la sécularisation, le système de valeurs individuelles devient plus fragmenté et les valeurs religieuses ont perdu leur rôle de guide. L'investigation se porte sur des comparaisons transversales et longitudinales se fondant sur les données des European Values Surveys (1990, 1999), la majorité des pays européens étant inclus dans l'étude. Les résultats indiquent une diminution de l'impact de la religion sur les valeurs des hommes et des femmes dans les années 1990, ainsi que l'existence d'une trame de relations entre la religion et les valeurs des genres dans la plupart des sociétés européennes.
Mots-clés: religiosité • valeurs de genre • sécularisation • European Values Survey
Introduction
Over the last few decades, European societies have experienced a decrease in the social importance of religious issues. Even if the secularization thesis is a controversial one, empirical research data have proven that both religious
DOI: 10.1177/0037768609103350 © The Author, 2009. Reprints and permissions:
http://scp.sagepub.com http://www.sagepub.co.uk/journalsPermissions.nav
practices and religious beliefs have decreased in Europe. However, the values and attitudes towards gender roles have also changed in the same period. In all European countries, attitudes have become more egalitarian with respect to the position of women in society. This article tries to identify the relationship between secularization and changes in gender values in Europe, analyzing the impact of traditional Christian values on attitudes towards women's participation in paid employment and towards women's role as housekeeper. The investigation employs cross-national and longitudinal analysis of two waves of EVS data (1990, 1999/2000).
The article is structured in five parts. The first summarizes theoretical approaches to secularization and its impact on public and private gender values, formulating some hypotheses about the relationships between the investigated phenomena. The second part presents the data used, the measurement methods and the analytical strategy, and the third is dedicated to the results. The last sections contain the conclusions and discussion.
1. Secularization and gender values
Secularization and changing gender roles are both part of a major cultural shift in Europe and North America over the last half century. Many studies have mentioned that the high level of economic prosperity and social security have changed people's religious orientation and gender relationships (Inglehart, 1990; Inglehart, Halman and Weltzel, 2004; Lundmark, 1995). Industrialization, urbanization and modernization have secularized society, decreasing the social importance of religion. According to Inglehart and Norris (2003), the first step in changing gender roles is related to industrialization, and the second to post-industrialization. The industrialization process set women free from traditional gender relationships, while post-industrialization established equality between the sexes.
Numerous studies have emphasized the direct relationship between religious values and gender values. Sherkat (2000) and Hofstede (1980, 1991) point out the association between Judeo-Christian culture and a patriarchal orientation towards gender roles. The Christian tradition has flourished on the basis of masculine cultures, which stress the differences between men and women and attribute different gender roles. On the other hand, Inglehart (1990) indicates that religious norms have had a functional role in traditional society. They have kept the family together, where the family was the main economic unit. Other studies have demonstrated the great influence of religious beliefs and practices on attitudes towards gender roles (Sherkat and Ellison, 1999; Ghazel, 2003; Wilcox and Jelen, 1991; Peek, Lowe and Williams, 1991; Gay, Ellison and Powers, 1996; Thornton, Alwin and Comburn, 1983; Sherkat, 2000; Hertel and Hughes, 1987). They have shown that religiously affiliated people are more inclined to support unequal gender roles and to consider that women are first of all housekeepers and mothers. Some studies have identified a similar relationship at the macro level. Hofstede (1980, 1991) has classified countries according to the main gender role orientation: countries with a masculine culture and countries with a feminine culture, encouraging gender equality.
Using Hofstede's classification, Verweij, Easter and Nauta (1997) observe that those countries with a feminine culture are more secularized than those with a masculine one.
Studies focused on religiosity have pointed out that, in Europe, the social significance of religion has decreased in the last centuries due to industrialization, urbanization and modernization. Secularization is the direct consequence of differentiation on the institutional level. The institutional spheres become more and more independent of one another and produce their own sets of rules and their own rationalities, different from those hitherto imposed by religion (Dobbelaere, 1981). Therefore, the churches have been eliminated from the areas in which, traditionally, they used to exert their control (Berger, 1969). The other social institutions have become autonomous (Tschannen, 1991) and transferred lay functions, such as education, healthcare and leisure, from the religious institutions (Willaime, 2001).
The impact of religion has diminished on the individual level, too. In traditional society, individual beliefs and practices depended on the norms prescribed by the church and on the practices of the local community (Halman and Draulans, 2004). Urbanization and modernization have broken community ties; the individual becomes free to adapt the religious practices to her/his own system of beliefs and lifestyle. Thus, individuals are free to build their own belief system and the result is a bricolage or patchwork (Dobbelaere and Jagodzinski, 1995a; Hervieu-Leger, 1998). These systems combine elements that come from different traditions and religions.
In the light of the fragmentation of individual value systems, religion has lost its guiding role on the individual level (Ester, Halman and de Moor, 1993). Within the individual value system each subsystem comes to function independently, developing its own internal logic. Several studies have pointed out the decreasing impact of religious values on other value orientations in the last decades (Halman, Petterson and Verweij, 1999; Halman, Petterson, 1996; Scheepers, Te Grothenhuis and van der Silk, 2002). The influence of religion on morality, politics, sexuality and family life has diminished.
In the context of secularization and the fragmentation of individual value systems, one can expect that the impact of traditional Christian religiosity on gender values will also decrease. Summarizing, one can formulate the following hypothesis:
(H1): The association between gender values and traditional Christian religious values has decreased in the last decade due to secularization.
However, the impact of secularization is not the same in all life domains. Secularization has affected mainly the values related to the public sphere, while the impact of religion on private issues remains stronger (Berger, 1969; Luckmann, 1967; Halman, Petterson and Verweij, 1999; Halman and Petterson, 1996; Thompson, 1996). Halman and Petterson (2004) have shown that the influence of religiosity on political values has decreased. In terms of private issues, Berger (1969) and Halman, Petterson and Verweij (1999) indicate the strong impact of religious values on family matters and private life.
According to Wilcox and Jelen (1991), attitudes towards gender roles have two distinct dimensions: a public dimension, which refers to women's
involvement in the labour market and politics, and a private dimension related to domestic roles. Due to the lack of empirical data, I will restrict my analysis of the public dimension to the relationship between religiosity and attitudes towards women's involvement in the labour market. In addition, I will test the effect of religious values on attitudes towards domestic roles. Based on the findings of previous research, the following hypotheses can be stated:
(H2): Religious values will have a greater influence on values related to private gender relationships, while the influence of religion on attitudes to women's participation in the formal labour market will be weaker.
(H3) The impact of religiosity on values related to women's participation in paid work will strongly diminish, while the impact on private gender values will slightly reduce over time.
However, the level of secularization differs from one country to another. Dogan (1999) shows that Western Europe is more diverse from the religious point of view than from the economic and social one. Remond (2003) indicates that there are many "Europes" from the secularization perspective and these "Europes" overlap with the main religious denominations. Therefore the particular country's effect on the impact of religious values on gender values has to be taken into account during the analysis.
Some categories of people within the same society are more influenced by secularization than others. Studies dedicated to the subject have pointed out that men are more exposed to secularization than women (Dogan, 1999; Hayes, McAllister and Studlar, 2000; Halman and Draulans, 2004). And educated people and the younger generation are less willing to accept religion inference in other life domains.
2. Data, measurement and analytical strategy
The analysis uses data from two waves of the European Values Survey, 1990 and 1999. These provide comparable data with the respect to religiosity and gender values.
The comfirmatory factor analysis shows that there are two dimensions of gender values (see the factorial structure in Figure 1 below). The first, Household Traditionalism, indicates individual attitudes towards private gender relationships, stressing the preference for the traditional role of women within the household, as mother and housekeeper. It is built from the responses to the question: "How much do you agree with each of the following statements?"
• A working mother can establish just as warm and secure a relationship with her children as a mother who does not work.
• A pre-school child is likely to suffer if his or her mother works.
• A job is alright, but what most women really want is a home and children.
• Being a housewife is just as fulfilling as working for pay.
The second indicator, Equal Labour, refers to the attitude towards women's involvement in the labour market. The latent variable explains the variation in responses to the question: "How much do you agree with each of the following statements?":
• A working mother can establish just as warm and secure a relationship with her children as a mother who does not work.
• Having a job is the best way for a woman to be an independent person.
• Both the husband and the wife should contribute to the household income.
The answer categories vary from 1 to 4, 1 meaning "strongly agree" and 4 representing "strongly disagree". The measurement model for gender values has been validated by previous analyses and it fits the data for cross-sectional and longitudinal analyses (Voicu, 2004; Voicu and Voicu, 2002). The model fits the data from both the 1990 and the 1999 EVS.
The lack of empirical data does not allow the investigation of other forms of religiosity such as popular religion, or belief in saints or in the Virgin Mary. The analysis was therefore restricted to the effect of traditional Christianity on gender values. With respect to religious values, a few indicators have been employed: Importance of God, which indicates, according to Dobbelaere and Jagodzinski (1995b), the centrality of the religious beliefs. The indicator uses the answer to the question: "How important is God in your life?" measured on a 10-point scale. Orthodoxy is a summative scale built from the positive answers to the questions: "Do you believe in God, life after death, sin, Heaven and Hell?" The scale takes values from 0 to 5 and indicates the number of traditional religious beliefs held by the individual. Confidence in Church is a summative scale built from positive answers to the questions: "Do you think your church in your country offers adequate answers to moral problems, problems of family life, people's spirituals needs and social problems?" The values vary from 0 to 4, showing the degree to which people rely on the church for solving individual and social problems. Religiosity is a latent factor, built as a factorial score, which aggregates the three indicators related to religious values: Importance of God, Orthodoxy and Confidence in church.
Other socio-demographic indicators have been used as control variables. Education was measured as the age when education was completed. Country of residence was recorded as a dummy variable indicating whether the respondent was a resident of the indicated country or not. France was used as a reference category.
The relationship between religiosity and gender values was investigated using Structural Equations Modeling available in AMOS. The path diagram, presented in Figure 1 investigates the impact of religiosity on the two gender values, Household Traditionalism and Equal Labour. The model was run first on the 1999 data set and this analysis provided the coefficients (regression weights and covariation) presented in Figure 1. The model fits the data, as indicated by the Fit indexes presented below the path diagram. Austria, Ireland and Northern Ireland have been excluded because for them a different scale was used for
working mother with children
children suffer with working mother
women want Vy home and children
being a housewife
job independence women
importance of God
orthodoxy <-(e9)
confidence in church
household income contribution
FIGURE 1
Structural equation model: the effect of religiosity on gender values
Fit indexes: delta2 IFI = 0.945; CFI = 0.945; RMSEA = 0.05; P test for close fit = 0.484
measuring attitudes towards gender roles. Turkey has been excluded from the analysis as well.
The next step was to constrain all the parameters according to the results for the entire 1999 data set and to run the model for each of the countries included in the survey. The model fits the data for most European countries included in the 1999 survey. For a few countries, the model does not fit the data and these countries have been excluded from the cross-sectional comparisons. These countries are Portugal, Denmark, Iceland, Romania and Malta. The model's fit is presented in Table A1 in the Appendix.
The third step consisted in running the model on the 1990 data set, constraining the parameters to the values resulting from the 1999 analysis on the entire sample. Sweden, Estonia and Lithuania have been excluded from the analysis because the questions concerning religious issues were not asked. As indicated in Table A1, the model fits the data for most of the countries included in the sample, excepting Iceland, Latvia and Poland. Therefore for most countries the longitudinal and cross-sectional comparisons are valid.
In the next step, the effects of religiosity on Household Traditionalism and on Equal Labour were compared for different countries and for different moments in time (1990 and 1999). The model was run for each country included in the 1999 sample and, in the first stage, the values of coefficients indicating religiosity effects on gender values were compared for European countries. The next phase kept constrained the model parameters for each country on the level revealed by the analysis of 1999 data, and compared the effects of religiosity on gender values in 1990 and 1999, country by country. The indices for the model's fit in comparing each country in 1990 and 1999 are presented in the Table A2 in the Appendix.
The last step analyzed the impact of religiosity on the level of gender values, controlling for the socio-demographic variables of age, sex, education and country of residence. For each dependent variable (Household Traditionalism and Equal Labour) an Ordinary Last Square Regression Analysis was carried out for the entire sample for both the 1999 and the 1990 data. Additional OLS Regression Analyses were carried out for each country which provided comparable data for 1990 and 1999, and for which the model fit the data in both waves. The goal was to detect the effect of religious values on private and public gender values controlling for some individual characteristics, and to detect changes occurring over the decade.
3. Results
The path diagram shown in the Figure 1 indicates a strong positive impact of religiosity on traditionalist orientations towards private gender relationships, while the effect on gender values related to women's employment is weak and not significant. The data support hypothesis H2, which predicts a stronger effect of religiosity on private gender roles but a non-significant effect on public ones. Therefore, one can say that in Europe religiosity still plays an important role in determining gender relationships within families, but the values related to women's involvement in paid labour have been more strongly affected by secularization.
Figures 2 and Figure 3 present the variation of unstandarsized regression effects of religiosity on Household Traditionalism and on Equal Labour among European countries in 1999, while Figures 4 and 5 show the variation of these regression coefficients between 1990 and 1999 in various countries. Looking at the impact of religiosity on private and public gender values country by country, a similar pattern can be observed in many European countries. The data reported in Figure 2 indicate a strong impact of religiosity on attitudes related to private gender relationships in most European countries. In almost all European societies, these relationships are strongly influenced by Christian religiosity, which encourages a traditional pattern of relationships and role division within the household.
However, the impact differs from one country to another. The data indicate a higher secularization of the private sphere in some of the ex-communist countries. The impact is non-significant in Russia, Ukraine, Bulgaria and Latvia; but Belarus, Lithuania and Croatia have lower scores compared to other countries in Europe. Not all the post-communist countries are similar in this respect, Romania and Hungary registering higher effects of religiosity on gender values; but in the ex-Soviet bloc countries, the private sphere is more secularized compared with either Western, Central, or Eastern Europe.
The data presented in Figure 3 show that the public sphere is more secularized with respect to gender relationships than the private one. For many European countries, the impact of religious values on values related to women's participation in the labour market is weak and non-significant. Only in a few countries is religiosity important in determining attitudes to public gender relationships.
Iceland Greece Belgium Spain Romania Hungary Italy Malta Luxembourg Netherlands Slovenia Germany France Poland Sweden Slovakia Denmark United Kingdom Czech Republic Estonia Portugal i Croatia Belarus Lithuania Russia Ukraine Bulgaria I Finland Latvia
FIGURE 2
Unstandardized effects of religiosity on Household Traditionalism (1999)
Romania Belarus Russia Hungary Latvia Iceland Sweden Croatia Estonia Ukraine Slovenia Lithuania Luxembourg Czech Republic Bulgaria Slovakia Portugal Greece Poland Belgium Italy
United Kingdom Denmark France Netherlands Finland Malta Spain Germany
-0.5 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3
FIGURE 3
Unstandardized effects of religiosity on Equal Labour (1999)
Belgium Hungary Slovenia Slovakia Spain Latvia Poland France Bulgaria Finland Czech Republic Italy Denmark Netherlands United Kingdom Portugal
-0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3
FIGURE 4
Differences between unstandardized effects of religiosity on Household Traditionalism (1990, 1999)
Slovenia Bulgaria Hungary Slovakia Czech Republic Poland Latvia Finland Denmark Belgium Netherlands Portugal Spain Italy
United Kingdom France
■1990 D1999
FIGURE 5
Unstandardized effects of religiosity on Equal Labour—comparison
1990-1999
Thus, in Germany, Spain, Malta, Finland, The Netherlands, France, Belarus and Romania, religiosity plays an important role in influencing attitudes to women's involvement in the labour market (the effect of Religiosity on Equal Labour is more that +/-0.1). In the first six mentioned countries a higher level of religiosity discourages women's participation in paid work, while in Romania and Belarus it encourages it. Again, the countries from the ex-Soviet Union have higher levels of secularization of the labour market, but they are not unique in this respect.
Generally speaking, the data confirm the hypothesis of higher secularization of the values related to women's employment and of the significant impact of religion on private life. In almost all the countries, the effect of religion on private gender values is more important than on opinions about women's labour market involvement. In Ukraine, Russia, Bulgaria and Latvia, both dimensions are strongly affected by secularization, the impact of religiosity being weak.
Looking to the data from a longitudinal perspective, in many European countries one notices a decrease in the effect of religiosity on Household Traditionalism between 1990 and 1999. EVS data indicate a higher decline for The Netherlands, the United Kingdom and Portugal and an increase in case of Belgium (see Figure 4). For the other countries the variation was significantly less and for many of them the impact of religious values on private gender values also declined over the 10 years.
The variation of religiosity's impact on values related to women's employment does not follow a common pattern in Europe (see Figure 5). In the case of The Netherlands, Belgium, Finland, Italy, France and Poland there is a significant decrease in the effect of religion on values related to women's employment outside the home. For Spain the data show an increase in the effect of religiosity on Equal Labour, while for the other countries the variation is not significant. Therefore, no significant variation can be observed.
Figures 6 and 7 show the variation of Household Traditionalism and Equal Labour by religiosity among European Countries in 1990 and 1999. The data in suggest a transformation in the effect of religious values on gender values during this decade in most European countries. Figure 6 indicates a higher impact of religiosity on Household Traditionalism in 1990 than in 1999. The data in Figure 7 show a similar pattern. The data point to a decrease in religiosity and an increase in support for women's involvement into the paid labour market.
Is this change the effect of religiosity decreasing or do other factors play an important role here? To find out the effect of religiosity on the dependent variables while controlling for other significant factors, an OLS Regression Analysis was employed. The results of the analysis, included in Table 1, indicate that the impact of religiosity on gender values is still strong when controlling for age, education, gender and country. On both dimensions of gender values, the effect of religion is significant, and a decrease occurs only in the case of values related to women's participation in paid labour. In 1990, the impact was stronger than in 1999, while in case of private gender values a slight increase is discernible.
154 Social Compass 56(2) 5.50 —f
■■z 4.50-
I 4.00-
1.75 Religiosity
FIGURE 6
Household Traditionalism by religiosity (1990, 1999)
On the other hand, the OLS Regression Analysis indicates that other factors, such as gender, age and education, play an important role in determining public and private gender values in both data sets. Age is not significant for values related to labour market participation in both 1990 and 1999. In addition, country has an important effect on public and private gender values. The effect of religious denomination on gender values was tested too, but the impact was not significant and it did not add supplementary explanation for the model.
m 3.80-o
1.75 Religiosity
FIGURE 7 Labour Equality by religiosity (1990, 1999)
An OLS Regression Analysis was also carried out for each country in both years (the results are not included in the present paper, but the author can provide them upon request). The regression was run on 12 countries, with Household Traditionalism and Equal Labour as dependent variables. The goal was to analyze the variation of religiosity's effect on gender values, controlling for age, education and gender. The results sustain the findings reported.
TABLE 1
Results of Ordinary Last Square Regression Analysis. Dependent variables: Equal Labour and Household
Traditionalism
1999 1990
Equal Labour Household Traditionalism Equal Labour Household Traditionalism
B Beta B Beta B Beta B Beta
(Constant) 4.031 3.722 3.998 4.24
Religiosity -0.12 -0.138 0.396 0.362 -0.144 -0.172 0.387 0.342
Gender (female) 0.203 0.129 -0.236 -0.119 0.193 0.123 -0.223 -0.105
Age 0.001 0.026(ns) 0.007 0.127 0 0.006(ns) 0.011 0.173
Education 0.006 0.04 -0.022 -0.113 0.006 0.037 -0.03 -0.148
United Kingdom -0.523 -0.117 -0.402 -0.071 -0.411 -0.125 -0.691 -0.155
Germany -0.082 -0.026 -0.628 -0.156
West Germany -0.406 -0.145 -0.374 -0.099
East Germany -0.204 -0.059 -0.421 -0.09
Italy -0.193 -0.06 -0.154 -0.038 -0.174 -0.061 -0.355 -0.092
Spain -0.114 -0.028 -0.514 -0.1 -0.081 -0.032** -0.588 -0.172
Netherlands -0.736 -0.166 -0.577 -0.103 -0.838 -0.213 -0.46 -0.086
Belgium -0.06 -0.018** -0.215 -0.052 -0.32 -0.13 -0.214 -0.064
Finland -0.304 -0.07 -0.253 -0.046 -0.076 -0.015(ns) -0.479 -0.069
Slovakia 0.121 0.031 -0.137 -0.028 -0.275 -0.074 -0.111 -0.022**
Hungary 0.050 0.011(ns) 0.085 0.015** -0.329 -0.083 0.222 0.041
Bulgaria 0.139 0.031 -0.042 -0.007(ns)
Slovenia 0.053 0.012(ns) -0.182 -0.032
Poland -0.04 -0.009(ns) -0.067 -0.012(ns)
Sweden 0.157 0.036 -0.533 -0.095
Estonia -0.228 -0.051 0.111 0.02
Lithuania -0.169 -0.038 0.217 0.039
Croatia 0.008 0.002(ns) -0.331 -0.059
Greece 0.140 0.034 -0.082 -0.015*
Russia -0.203 -0.07 0.23 0.063
Luxembourg -0.112 -0.028 -0.09 -0.018*
Ukraine -0.075 -0.018** 0.211 0.041
Belarus -0.096 -0.022** 0.088 0.016*
Latvia -0.036 -0.008(ns) -0.092 -0.016**
Portugal Denmark Norway Czech Republic
R 0.294 0.509
Adjusted R2 0.085 0.258
* significant for p < 0.05, all other coefficient excepting (ns) are significant at p < 0.001. ** significant for p < 0.01.
-0.374 -0.096 0.652 0.124
-0.038 -0.010(ns) -0.001 0.000(ns)
0.456 0.125 -0.277 -0.056
-0.122 -0.031 -0.743 -0.14
-0.188 -0.052 -0.612 -0.126
-0.250 -0.089 0.324 0.086
0.337 0.562
0.113 0.315
158 Social Compass 56(2) Conclusions
This articles has investigated the impact of religion on gender values, using 1990 and 1999 EVS data. The approach was a comparative cross-sectional and longitudinal one, trying to find what impact traditional Christian religiosity has on gender values in different European countries, and to identify possible variation. The data support the hypothesis of a stronger effect of religion on private gender values in all European countries. Even controlling for the influence of other factors, the effect remains. The impact of religiosity on values related to women's labour market involvement is weaker and the data indicate a decrease in this effect over the last decade. There is no universal pattern, but the analysis shows a common trend in many countries. The available data indicate that, in Europe, while traditional Christian religion has a significant place in the private sphere, the labour market is free of "supernatural" forces.
This fact has a direct impact on women's status and on the negotiation of gender roles. The results indicate that traditional Christian religion does not represent a hindrance for women's involvement in the labour market in many European societies. On the other hand, in most European countries, the Christian religion still prescribes how domestic work is shared within the family and domestic roles assigned, encouraging the preservation of traditional family relationships.
The article assumes that secularization has reduced gender inequality, both within the family and in the labour market; the empirical data confirm this hypothesis. However, both phenomena are part of the modernization process and the decrease in traditional values may be the leading force behind them. Due to the lack a solid indicator for traditionalism in the available data, the current research could not investigate the complex relationships among religiosity, gender equality and traditionalism, but this topic can be addressed in the future.
The results of the present research are also limited to the effect of traditional Christian values on gender values in European countries. They cannot be extrapolated to countries outside Europe or to the effect of other religious traditions. This could be an interesting topic for further research. On the other hand, although the current analysis emphasizes the effect of Christian religiosity on value orientations related to women's position in the labour market, the public sphere cannot be reduced only to paid employment. Unfortunately, the lack of empirical data does not allow an extension of the analyses to other domains, such as women's political participation or involvement in civil society. This also will be the subject of further research.
ACKNOWLEDGEMENT
The research was supported by the Grant ID-56/2007 provided by the Romanian National University Research Council. The author addresses special acknowledgement to Loek Halman for his very helpful comments on the draft of this paper.
REFERENCES
Berger, P. (1969) The Sacred Canopy. New York: Anchor Books.
Dobbelaere, K. (1981) "Secularization: A Multi-dimensional Concept", Current Sociology 29(2): 1-213.
Dobbelaere, K. and Jogodzinski, W. (1995a) "Secularization and Church Religiosity", in J. van Deth and E. Scarbrough (eds) The Impact of Values, pp. 76-119. Oxford: Oxford University Press.
Dobbelaere, K. and Jogodzinski, W. (1995b) "Religious and Ethical Pluralism", in J. van Deth and E. Scarbrough (eds) The Impact of Values, pp. 218-49. Oxford: Oxford University Press.
Dogan, M. (1999) Sociologie politico. Bucharest: Alternative.
Ester, P., Halman, L. and de Moor, R. (1993) "Value Shift in Western Societies", in P. Ester et al. (eds) The Individualizing Society: Value Change in Europe and North America, pp. 1-20. Tilburg: Tilburg University Press.
Gay, D., Ellison, C. and Powers, D. (1996) "In Search of Denominational Sub-cultures: Religious Affiliation and 'Pro-family' issues revised", Review of Religious Research 38(1): 3-17.
Ghazel, R.J. (2003) "The Sources of Gender Role Attitudes among Christian and Muslim Arab-American Women", Sociology of Religion 64(2): 207-22.
Halman, L. and Petterson, T. (1996) "The Shifting Sources of Morality: From Religion to Post-materialism", in L. Halman and N. Nevitte (eds) Political Value Change in Western Democracies, pp. 261-84. Tilburg: Tilburg University Press.
Halman, L., Petterson, T. and Verweij, J. (1999) "The religious factor in contemporary society", International Journal of Comparative Sociology 40(1): 141-60.
Halman, L. and Draulans, V. (2004) "Religious beliefs and practices in contemporary Europe", in W. Arts and L. Halman (eds) European Values at the Turn of the Millennium, pp. 283-316. Leiden: Brill.
Halman, L. and Petterson, T. (2004) "Normative orientations towards the differentiation between religion and politics", in W. Arts and L. Halman (eds) European Values at the Turn of the Millennium, pp. 317-37. Leiden: Brill.
Hayes, B., McAllister, I. and Studlar, D. (2000) "Gender, Post-materialism and Feminism in Comparative Perspective", International Political Science Review 21(4): 425-39.
Hertel, B. and Hughes, M. (1987) "Religious Affiliation, Attendance and Support for 'Pro-family Issues' in the United States", Social Forces 65(3): 858-82.
Hervieu-Léger, D. (1998) "Transmission and Formation of Socio-religious Identities", International Sociology 13(2): 213-28.
Hofstede, G. (1980) Culture's Consequences: International Differences in Work Related Values. London: Sage.
Hofstede, G. (1991) Culture and Organization: Software of Mind. London: McGraw Hill.
Inglehart, R. (1990) Culture Shift in Advanced Industrial Society. Princeton, New Jersey: Princeton University Press.
Inglehart, R., Halman, L. and Welzel, C. (2004) "Introduction", in R. Inglehart et al. (eds) Human Beliefs and Values, pp. 1-20. Mexico City: Siglo XXI.
Inglehart, R. and Norris, P. (2003) Rising Tide: Gender Equality and Cultural Change around the World. Cambridge: Cambridge University Press.
Luckman, T. (1967) The Invisible Religion. London: Macmillan.
Lundmark, C. (1995) "Feminist Political Orientation", in J. van Deth and E. Scarbrough (eds) The Impact of Values, pp. 250-74. Oxford: Oxford University Press.
Peek, C., Lowe, G. and Williams, S. (1991) "Gender and God's Word: Another Look at the Religious Fundamentalism and Sexism", Social Forces 69(4): 1205-21.
Rémond, R. (2003) Religie §i societate în Europa. Ia§i: Polirom.
Scheepers, P., Te Grothenhuis, M. and van der Silk, F. (2002) "Education, Religiosity and Moral Attitudes: Explaining Cross-national Effect Differences", Sociology of Religion 63(2): 157-76.
Sherkat, D. (2000) "'That They Be Keepers of the Home': The Effect of Conservative Religion on Early and Late Transition into Housewifery", Review of Religious Research 41(3): 344-58.
Sherkat, D. and Ellison, C. (1999) "Recent Developments and Current Controversies in the Sociology of Religion", Annual Review of Sociology 25: 363-94.
Thompson, K. (1996) "Religion, Values and Ideologies", in S. Hall et al. (eds) Modernity: An Introduction to Modern Societies, pp. 321-66. Oxford: Blackwell.
Thornton, A., Alwin, D. and Comburn, D. (1983) "Causes and Consequences of Sex-role Attitudes and Attitude Change", American Sociological Review 48(2): 211-27.
Tschannen, O. (1991) "The Secularization Paradigm", Journal for the Scientific Study of Religion 30(4): 395-415.
Verweij, J., Easter, P. and Nauta, R. (1997) "Secularization as an Economic and Cultural Phenomenon: A Cross-cultural Analysis", Journal for the Scientific Study of Religion 36(2): 309-24.
Voicu, M. (2004) "Work and Family Life in Europe: Values Patterns and Policy-making", in L. Halman and W. Arts (eds) European Values at the Turn of the Millennium, pp. 231-54. Leiden: Brill.
Voicu, M. and Voicu, B. (2002) "Gender values dynamics: Towards a common European pattern", Romanian Journal of Sociology 1-2: 40-63.
Wilcox, C. and Jelen, T. (1991) "The Effects of Employment and Religion on Women's Feminist Attitudes", International Journal for the Psychology of Religion 1(3): 161-71.
Willaime, J.-P. (2001) Sociologia religiilor. Ia§i: Institutul European.
Wilson, B. (2000) Religia din perspectivä sociologicä. Bucharest: Trei.
Mälina VOICU obtained her PhD in 2004 at the University of Bucharest. She is now Principal Research Fellow in the Research Institute for Quality of Life in Bucharest. She is a member of the European Values Study Executive Committee and Program Director of the European Values Study. Her recent publications deal with the comparative study of values, particularly focusing on religion and gender. Her publications include Religious Romania: On the European Wave or behind it? (2007, European Institute) and Romanian Values: 1993-2006 (2007, European Institute, co-edited with Bogdan Voicu). ADDRESS: Research Institute for Quality of Life, Romanian Academy of Science, Calea 13 Septembrie, 13, Sector 5, Bucharest, Romania. [email: malina@iccv.ro]
Appendix
TABLE A1
Model fit for 1990 and 1999 constraining all the parameters according to the results for the entire sample in 1999
Country 1999 1990
CFI RMSEA CFI RMSEA
France 0.944 0.049 0.937 0.056
United Kingdom 0.903 0.066 0.926 0.058
Germany 0.888 0.091
Italy 0.889 0.065 0.925 0.062
Spain 0.967 0.040 0.950 0.051
Portugal 0.697 0.091 0.832 0.082
Netherlands 0.929 0.610 0.917 0.073
Belgium 0.923 0.058 0.961 0.042
Denmark 0.786 0.092 0.801 0.084
Sweden 0.876 0.067
Finland 0.872 0.066 0.871 0.082
Iceland 0.793 0.075 0.771 0.082
Estonia 0.945 0.046
Latvia 0.892 0.057 0.676 0.094
Lithuania 0.878 0.065
Poland 0.821 0.078 0.785 0.086
Czech Republic 0.929 0.056 0.938 0.061
Slovakia 0.964 0.049 0.950 0.053
Hungary 0.938 0.056 0.912 0.069
Romania 0.412 0.109
Bulgaria 0.949 0.046 0.901 0.068
Croatia 0.894 0.063
Greece 0.874 0.071
Russia 0.897 0.064
Malta 0.012 0.156
Luxembourg 0.911 0.058
Slovenia 0.957 0.048 0.920 0.061
Ukraine 0.813 0.084
Belarus 0.924 0.050
West Germany 0.925 0.066
East Germany 0.946 0.063
Entire sample 0.971 0.037 0.955 0.048
TABLE A2
Indexes of fit for the constrained model by country
Model fit 1990 Model fit 1999
CFI RMSEA CFI RMSEA
France 0.926 0.062 0.979 0.039
United Kingdom 0.937 0.055 0.973 0.044
Germany 0.981 0.049
West Germany 0.994 0.020
East Germany 0.983 0.038
Italy 0.934 0.060 0.947 0.057
Spain 0.972 0.039 0.995 0.020
Portugal 0.908 0.063 0.909 0.070
Netherlands 0.973 0.043 0.977 0.044
Belgium 0.946 0.051 0.958 0.054
Denmark 0.948 0.044 0.960 0.051
Sweden 0.985 0.033
Finland 0.825 0.099 0.956 0.049
Iceland 0.916 0.061
Estonia 0.991 0.024
Latvia 0.832 0.070 0.995 0.016
Lithuania 0.981 0.033
Poland 0.914 0.044 0.981 0.032
Czech Republic 0.969 0.045 0.990 0.024
Slovakia 0.961 0.048 0.987 0.037
Hungary 0.952 0.053 0.979 0.041
Romania 0.934 0.046
Bulgaria 0.888 0.074 0.984 0.033
Croatia 0.980 0.035
Greece 0.972 0.043
Russia 0.983 0.034
Malta 0.945 0.047
Luxembourg 0.960 0.050
Slovenia 0.974 0.036 0.993 0.025
Ukraine 0.991 0.01
Belarus 0.982 0.031
