ssoar_reference_gesis-ssoar-60934_study_program_90_search

Property	Value
?:linkContext	In the last decades, European societies have experienced a cultural shift toward an increase in egalitarian values that reflect changing attitudes toward gender roles. In this context, Italy, with its familistic and traditional culture regarding the societal role of women, represents a good case for investigating change in gender role attitudes through a path-dependency approach. In this paper, I aim to assess whether the postmaterialist shift is taking place in Italy and if variation in support for the role of females in the public sphere is part of a broader cultural change related to general postmaterialist values. Modernization processes varied across Italian regions and here I also investigate if different gender cultures developed accordingly. The analyses present a controversial situation: support for the role of females in the public sphere does not mirror change in variation in generic postmaterialist human values and, overall, the shift toward more egalitarian and postmaterialist values is not developing as one would expect. On the contrary, the results demonstrate a reinforcement of materialist/traditionalist values mainly due to the intracohort change. In the final section of the paper, I discuss these unexpected results. Gender role attitudes can be defined as the cognitive representation of what is believed appropriate for male and female roles in a specific cultural context, rooted in historic development and differs across societies . This representation can refer to the domestic sphere, expressing the preference for a certain family model, or the public sphere, showing more or less support for female or male roles in the fields of education, politics, or the labor market. Despite the public debate on this topic, there is a lack of knowledge regarding Italian attitudes on the role of women in the public sphere and how this changes over time. This paper aims to fill this gap by analyzing two specific aspects: support for the dual role of women (as workers and caregivers) and support for equality in the labor market. In this study, I adopt a societal-oriented approach to read the change in support for female roles in the public sphere as part of a broader cultural and social transformation, connected with modernization processes that lead to transformations in the social structure. As pointed out by several authors , modernization factors such as higher educational levels among the population, the rise of female labor market participation, and secularization are strictly connected with the shift toward more egalitarian gender values both at the individual and societal levels: people that directly experience higher female presence in the public space tend to assume more egalitarian values and, at the same time, a postmaterialist culture tends to morally legitimate and promote greater womens involvement in public life. This impacts on the individuals' socialization context and in their direct experience of gender relations, changing in this way the social representation of 'proper roles' for men and women . The cultural shift toward egalitarian/postmaterialist values can be explained through societal level perspectives as those proposed by and . Inglehart's theory of value change (1971) is based on the idea that when individual survival needs can be taken for granted, room for immaterial needs such as autonomy and selfrealization is then possible. In the case of gender relations, fulfilling materialist needs (with the necessity of a sexual division of work for the survival of the individual and family) opens space for postmaterialist needs including self-realization for all family members, pushing toward a more egalitarian perspective. Therefore, individuals define their priorities according to their immediate needs ('scarcity hypothesis'). This change is affected by the existential condition in that specific period. Nevertheless, this change is not immediate but gradual and reflects the conditions faced by people during their formative years. The change in values is then led by the cohort succession ('socialization hypothesis'). is extremely helpful to disentangle these two aspects. He addresses the relation between social change and two population processes: demographic metabolism and individual life cycles considered as aggregated. The first one, related to the 'cohort replacement', refers to the dynamics of fertility and mortality with respect to the possibility of each society to renew itself, generation by generation, where new individuals interact with the contemporary cultural heritage and enact it in a new way. The arrival of a new cohort makes social change possible. Italy is a country with an aging population and a weak cohort replacement: people older than 64 years old represent an increasing part of the population, while those below 14 years old are decreasing together with a fertility rate barely steady around 1.3 (see in the Appendix). However, each generational cohort lives its life cycle in a specific historical period, characterized by political, economic and social issues that affect the interpretation of life stages, producing different experiences regarding the transition to adult life, work and family, and public and private gender roles. The recent literature refers to this process as the 'period effect' or 'intracohort change' . Did the modernization processes affect the gender role patterns and the shift toward egalitarian/postmaterialist values in Italy? Postmaterialist theory underlines how an economic situation defines the context for scarcity or prosperity in living conditions, and how this impacts human values. But the economic context also has an important role in understanding patterns of gender relations established in a specific context and period. examines that prevailing gender role patterns are strictly connected to the dominant production model in a particular context. According to , three main patterns of gender relations can be found in Europe. The first one is based on the well-known male breadwinner/female homemaker model, a result of Western capitalism and modernization processes that led to the establishment of the urban bourgeoisie with clearly separated spheres of life for men (public) and women (domestic), based on an idea of complementarity. This model is typical in many European countries though it formed in different historical periods with the establishment of the bourgeois class. The second pattern, prevalent in Eastern countries, is based on the dual-worker model. In this case, egalitarian values developed mainly with regard to the public dimension of life but not in the private one, as explained by Voicu and Tufiş in the Romanian case (2012). Finally, the family-economy model refers to the family-economy model refers to the organization of work and familial relations in rural areas where the value given to work carried out by women and men is the same because both contribute to the survival of the family: the relationship between partners is then based on a mutual dependency. To explain this model, Pfau-Effinger (2004) refers to Finland. Italy presents a story of a fragmentary industrial and cultural path that reflects the development of gender role patterns. Since the Unification of Italy in 1861, policy-makers have addressed building a common Italian culture. Even today Italy's gender culture is typically oriented toward a traditional division of work between men and women based on a complementarity gender role model . The origin of this model is deeply rooted in the history of the country and institutionally promoted. For example, in the Italian fascist era female roles in the public sphere were highly discouraged. Even if there was some maternity protection introduced in this period, other sets of rules excluded or limited female presence in the labor force to jobs mainly related to caregiving. In the following decades, women's opportunities improved but still gender cultures are affected by the fact that all the economic, labor, and family policies have been generally implemented regardless of the previous local history that led to different modernization processes in Italian areas. Across Italy, despite agriculture being the main economic sector in the twenty-first century, three main economic models can be detected according to the political powers that according to the political powers that ruled the states and city-states that eventually merged into the modern nation of Italy in 1861. Since that time, the development of Italy knows three different rhythms, speeds, and peculiarities rooted in the different historical pathways that characterized the so-called Three Italies . In the North, thanks to the political strategies of the Austrian and Pedmont governments as well as the introduction of new technologies, agriculture became intensive and marketoriented. The Centre, ruled by the papal state, tended to underestimate the manufacturers from small inland villages and the potentiality of strategic harbors on the coasts and the society kept its sharecrop structure. The South, under the rule of the Kingdom of the Two Sicilies, remained anchored to the latifundium feudal regime. According to , because of capitalism's structuring of the economy together with urbanization, the North is the typical scenario for the establishment of the urban bourgeoisie with gender role patterns based on complementarity, because of the capitalist address given to the economy together with the urbanization. In the sharecrop system of the Centre, the model is close to that of family-economy even if patriarchal structures remain and women's lives are strictly bound to men. Nevertheless, in this type of society, the older female of the family could have a strong authority on the other women. It is the case of the 'massaia' in Tuscany or the 'arzdaura' in the region of Emilia Romagna , who managed all the female activities and domestic chores. At the same time she worked in the fields for the crop and grape harvests. The southern economic system typically employed seasonal workers, mainly employing only men. Women's activities remained centered in the domestic space . At the end of the twenty-first century, modernization processes led to the stabilization of the manufacturing sector, which gave job opportunities to women in the North and less in the Centre. In the South, the economy remained linked to the agriculture system and the gender role patterns did not change much. According to these different paths, the acceptability of certain male and female roles may therefore differ within Italian areas. In more recent times, the three different modernization paths are still visible. We could consider three modernization factors related to changing gender role attitudes in the frame of modernization processes: the rise of female employment, increase in the population's level of education, and secularization. Graphs in (a) and 1(b) show the labor market participation for men and women 24-54 years old by geographical area in the period 1993-2008 (data refer to the historical series available from the Italian National Institute of Statistics -ISTAT 2016). The 'three Italies' differ in the level of employment rates among the active population both in the male and female cases. But, while women who live in the North and Centre steadily increase their economic participation, those who live in the South show a more static trend. On the contrary, male economic participation in the South steadily drops since 2002. In the other areas, men who participate in the labor force slightly increases, even if there are some fluctuations. According to Inglehart's 'scarcity hypothesis', one could expect that men living in the South of Italy show a shift toward more traditionalist and materialist values, as a consequences of their reduced economic participation, and they may see the increase of working women as a growth of competition, as happened in the case of Germany after the reunification . (a) and 2(b) refers to the proportion of graduates among men (a) and women (b) aged 25 in the three areas. In all cases, we can see a similar dynamic with a growth in education levels until 2005. Then it dramatically drops. However, the density of those highly educated is much greater in the Centre, especially among women , than in the other areas. In particular, the South presents the lowest level of graduates both for men . The structural explanation argues that the shift toward more egalitarian positions is also due to the fact that women and men increasingly study alongside each other and in an increasing number of fields and levels. One could therefore expect that the increase in education levels among the Italian population reflects in more egalitarian and postmaterialist values. This change would be less relevant in the South and more evident in the Centre. Secularization is often a relevant modernization factor affecting gender role attitudes and basic human values . Processes related to urbanization and industrialization contributed to reducing the normative authority of religious institutions which have historically supported gender role patterns based on the complementary functions of reproduction (women) and production (men). As an indicator of the process of secularization, variation in the proportion of people over the age of 6 out of the entire population who never attend religious services was selected. (a) and 3(b) shows these trends respectively for males and females. The dynamics look similar for both: in the North and Centre the secularization process is not linear and presents recurring fluctuations. In the South, these fluctuations appear less evident and the amount of people who never attend religious services is strongly lower. In all areas, the secularization process involves men more than women, who, generally speaking, are still more religious. In the South, for example, males who never attend religious services in 1995 pass from 12% to 17.2% in 2008; women from 5.2% in 1995 to 9.4% in 2008. In the North, the proportion of non-attending males increases from 21.6% in 1995 to 27.1% in 2008 while for the females it rises from 13.5% to 18.8%. In the Centre, the percentage for men goes from 19.5% in 1995 to 28.1% in 2008, for women from 12% to 17.2%. However, these trends are not indicative of a dramatic change with regard to the secularization process; therefore, one could expect that this modernization factor would not particularly affect the change in gender role attitudes, especially in the South and for women. I assume that the change in a specific domain of human values, such as the support for female roles in the public sphere, is part of a broader cultural change, which is the shift toward postmaterialist values. So I suppose that: H1a: The variation in support for the role of females in the public sphere follows a dynamic similar to the variation in the postmaterialist position. H1b: The composition of the mechanisms of social change (intracohort change and cohort replacement) involved in the variation of the support for the female role in the public sphere is similar to those of the variation in the postmaterialist position. The economic situation of Italy, delineated by the macro indicators, does not refer to a period of scarcity between 1988 and 2008: I therefore expect that: H2: The results show a shift toward more egalitarian attitudes and postmaterialist values. Considering the demographic dynamics developed in Italy, with the slow cohort turnover and the empirical studies based on Ryder's perspective , I suppose that: H3: The intracohort change is the prevalent mechanism of change, while the cohort replacement, even if present, has a lower impact. Finally, according to the different historical pathways of economic development and gender role patterns of the 'Three Italies' as well the recent dynamics of the modernization factors shown, I expect that: H4a: Men in the South reinforce their materialist values and traditional gender role attitudes because of the drop in economic opportunities and the lower level of tertiary educated people. This is not the case for men living in the other areas. H4b: Women increase their egalitarian attitudes and generic postmaterialist values because of their increased economic participation and above all because of the growth of women with tertiary education among the population, in particular in the Centre. H4c: The secularization process is not affecting the variation in support for the female role in the public sphere, nor the shift toward postmaterialist values because its development is not dramatic, in particular for women and especially in the South of Italy. In order to test my hypothesis, I use pooled data collected in Italy in six observation points by different cross-national and repeated surveys: 1988 and 1994 from the International Social Survey Program , 2005 from the World Values Survey from the European Values Survey (EVS 2011). shows the details of the sample sizes. To assess if the change in gender role attitudes is part of the wider cultural change, I compare the trend and the decomposition of the variation in support for the role of females in the public sphere to the variation of the postmaterialist index , chosen as indicator of a general values system. I examine two aspects of the egalitarian support toward the female public role. The first is fully referred to as the public sphere and measures the support for equality in the labor market, tapped by the item 'When jobs are scarce, men have more right to a job than women' collected in 1990 (EVS), 1999 (EVS), 2005 (WVS) and 2008 (EVS). In all the surveys, the respondents can choose between three categories of response: 'agree', 'disagree', 'neither agree nor disagree'. The second dimension refers to the support for the dual role of women as worker and caregiver and it is covered by the item 'A job is alright but what most women really want is a home and children', available in 1988 (ISSP), 1994 (ISSP), 1999 (EVS), and 2008 (EVS). One could say that this dimension is between the public and the domestic domains if compared to the previous item. So together they allow for grasping different nuances in support for women in the public sphere. This item does not have the same response categories in all the surveys. In EVS, they are: 'agree strongly', 'agree'; 'disagree', 'disagree strongly'. ISSP has an additional middle category: 'not agree nor disagree'. For both dimensions, I considered only the disagreement categories, here indicating egalitarian support/postmaterialist values, recoded as dummies . Sample size by context and year. (disagree = 1; 0). In the case of support for the dual role, the two categories 'very disagree' and 'disagree' have been previously collapsed together. The 4-item Post-materialism index is available for 1990 (EVS), 1999 (EVS), 2005 (WVS), and 2008 (EVS). It is built on the basis of the national priorities assigned by the respondents and has three categories: 1 = materialist; 2 = postmaterialist, and 3 = mixed. This study considers only the postmaterialist position and transformed the index in a dummy variable (1 = postmaterialist, 0 = other). show the variation in support for equality in the labor market, for the dual role of women and in the postmaterialist position from 1988 (1990 when data are not earlier available) to 2008. On the one side, support for equality in the labor market increases over time for both men and women. This trend perfectly adheres to the postmaterialist theory. But, on the other side, variations in the postmaterialist position and in support for the dual role of women drops apparently until the end of the 1990s and starts to slightly increase in 2008. The graphs in the appendix ( ) provide visual information about these dynamics. This controversial figure tells us that in Italy the postmaterialist transition is not taking place in the way shaped by Inglehart's perspective. Even if the economic situation does not look to push toward a loss of prosperity, the variations show a reinforcement of material values and conservative gender role norms in the domestic sphere, while gender role attitudes concerning the public sphere appear more egalitarian. The decomposition of the components of social change will assess these patterns. I adopt decomposition modeling to decompose the intracohort change and the change due to processes related to the cohort replacement, comparing a generic postmaterialist position and a specific domain of human values, that is attitudes toward female roles in the public sphere. According to , this kind of analysis does not require panel data and repeated cross-section data are suitable, assuming that among the same birth cohort the probability of death is not related to variable of interest. Therefore, the use of pooled data is applicable. Firebaugh's linear decomposition model originally employs OLS regressions. Here, following the example of previous studies , the model is adapted for dichotomous variables. I perform the same modeling for the two aspects of support for women in the public sphere and for the postmaterialist position, separately for women and men and for geographic areas. The basic assumption of this technique is that the social change occurred between the final moment of observation and the initial one, due to the sum of the intracohort change, the cohort replacement and their joint effect (e). The modeling consists in sequential logistic regressions. In the first one, only the effects of period and cohort are tested (M0). In the next steps, control variables can be sequentially added. Here I controlled for modernization factors that are relevant to the change in gender role attitudes: the increase in educational level, labor market participation, and secularization . I firstly added the educational level (M1), and secondly the labor market participation (M2). Finally, I added religiosity as an additional control variable. Independent and controlling variables, collected in all the surveys, have been harmonized as follows: -Period variables refer to the survey year (ISSP88, EVS90, ISSP94, EVS99, WVS05, EVS08), the first conducted as reference. -Educational level presents three categories: low (below low secondary school), middle (secondary school) and high (tertiary and upper) levels. These levels are recoded in dummies and the lowest is the reference. -Employment status differs in: full-time worker, part-time worker (less than 30 hours/week); non-paid worker . Also in this case the variable is transformed into dummies and the not working category is adopted as reference. -To indicate religiosity, the variable regarding attendance at religious services has been chosen and firstly recoded in four categories: high attendance (at least once a week), moderate attendance (two or three times a month), rare attendance (one or two times a year), and never. Then these categories are recoded as dummies. People who never go to religious services are considered as reference in the analyses. The logistic regression equation used is: ln (  l −  ) =  0 +  1 survey year +  2 birth year In this case, the logistic regression esteems the probability  of assuming postmaterialism/egalitarian. The amount of change due to processes related to the intracohort mechanism is computed as a product of the survey year effect and the length of the period of observation; the amount of change due to the cohort replacement results from the product of the birth year effect and the difference between the average of birth year at the final point of observation and the initial one. In the procedure applied, these components have been computed in each step. The overall social change is defined by the difference between the logic at the final observation (2) and the initial (1): Results from the decomposition modeling confirm the ambivalent shift earlier detected by observing the variation in postmaterialist values and gender role attitudes. The logistic regression results are available on request. The mechanisms of change in support for equality in the labor market appear different between men and women (Table 5(a) and 5(b)). Testing for the period effect and the cohort effect (M0), for men living in the North and in the South of the country, the intracohort change prevails on the cohort replacement. On the contrary, for women and men living in Central Italy, the change registered is due only to the cohort replacement: the period effect in these cases is not significant. This means that here the increase of support for equality in the labor market is not related to the contextual circumstances but only because new cohorts, those more educated and actively participating more in the labor force, are slowly replacing the older. So, my third hypothesis is confirmed for men living only in the North and South. Controlling for education (M1) in all these groups, the size of the cohort replacement component decreases consistently. In the case of men living in the Centre and Northern women, economic participation (M2) seems to have a relevant impact, while for the women who live in the Centre and South this is marginal. This can be explained by the fact that in these areas the growth of female economic participation is very slow. When controlling for religiosity (M3), the situation slightly differs, this can be interpreted as a confirmation of the hypothesis H4c that assumes that this process is too slowly developed in Italy to affect this social change. For men living in the North and South , both the mechanisms contribute to social change. In this case, male support for equality in the labor market is affected by the contextual situation as well by the replacement of the old cohort by younger and more educated cohorts. In effect, only the educational level (M1) seems to make some differences: controlling for labor market participation (M2) and religiosity (M3), the composition of the change remains the same. The decomposition results of the variation in support for the dual role of women reveal that despite the dynamics shown in the variation's trends and ), the overall social change for men living in the Centre and South, and for women living in the Centre is not significant (Table 6(a) and 6(b)). In the other cases, results refer to a negative change, that is, the drop in egalitarian support earlier observed in this dimension. Testing the effect of survey year (period effect) and respondent's birth year (cohort effect), in M0 it is clear that for all the groups considered, this drop is due to the period effect that in the case of men is two times the cohort replacement component. Also for women, the intracohort change is 1.5 times higher than the cohort replacement. Controlling for education (M1), we can see that this is a potential driver of change, especially for women in the South, where the size of the cohort replacement component passes from 0.546 to 0.348. The increased economic participation seems to affect only women who live in the North of the country, while controlling for religiosity does not make relevant changes. These results address education as the main predictor of the shift toward more egalitarian values. Nevertheless, the replacement of the older, less educated cohort, is not enough to set off against something happening in the context of Italy in the observed period. The analysis of the decomposition of the variation in the postmaterialist position overlaps with the decomposition results of the support for the dual role of women. Here the overall social change that occurred between 1990 and 2008 is significant only for men living in the North and in the South (a)) and for women living in the South (b)). In all the cases, the period effect is largely prevailing, in particular for men in the North. Education is the main predictor of change (M1) for both men and women, while the change in economic participation (M2) and religiosity (M3) between generations seems to not be relevant. This paper analyzed the change in two dimensions of support for female roles in the public sphere in Italy, assuming that the change in this specific domain of values is part of a broader cultural change. To consider this general change, I referred to the postmaterialist shift proposed by . It briefly described the economic development of the country in consideration of three different Italian contexts, characterized by the gender role patterns developed according to the different historical pathways. The results refer to a controversial situation. First of all the assumption that changes in gender role attitudes are part of a wider cultural change is only partially confirmed. Despite that the support for the dual role of women as workers and caregivers presents trend and decomposition results similar to variation in the postmaterialist position, this is not the case for the support for equality in the labor market Moreover, only the latest dimension actually shows a growth in egalitarian support. This first result confirms that gender role attitudes are a complex concept and it is good practice to separately analyze the different dimensions that compose this concept, as already argued in previous studies . Probably this item, which is particularly related to economic issues and job opportunities, is more affected by the living condition that, as shown, in the period here considered do not refer to a period of scarcity and people tend to show more egalitarian values. Then, for men living in the South, who faced a drop in employment rate associated with the rise of female employment, the social change is smaller and this confirms my hypothesis 4a. But despite all the assumptions that drove my second hypothesis, results concerning variation in the postmaterialist position and support for the dual role of women address a worsening of egalitarian support and of postmaterialist values, both higher in the past. Differently from other controversial situations traced in the literature, where results addressed fluctuations , backlash or persistence , here data refer to a reinforcement of the materialist/traditionalist values. What can explain this scenario? According to the decomposition results, the causes of this reinforcement should be searched for in contextual conditions. In fact, as expected according to my third hypothesis, the intracohort change is the prevalent mechanism of change, except from women and men living in the Centre for which the increase in support for equality in the labor market is explained only by the arrival of new, higher educated cohorts. For the variation in support for the dual role of women and the postmaterialist position, the effect of the period is dominant and the effect of increased educational level and female participation are not enough to balance the negative effect of the period, neither for women living in the Centre where these modernization factors are quite relevant, as supposed in H4b. In all the cases, including the change in religiosity did not make much difference in the model probably because, as argued in the last hypothesis H4c, this process is quite slow in Italy and non-religious people continue to be a minority. In this study, the period effect has been delineated mainly considering structural and economic components that should define the context wherein people define their own priorities that are then reflected in attitudes and values. The Italian demographic dynamics may suggest slow value changes because of the poor cohort turnover, but this could be addressed by the persistence of materialist/traditionalist values or by a very slow postmaterialist/egalitarian shift, rather than a materialist reinforcement. In the observed period, the economic circumstances do not refer to a drop in economic prosperity and this should support the postmaterialist/egalitarian shift. Further studies should investigate additional elements that characterized the cultural context between 1988 and 2008, for example, considering the effect of the Christian-Catholic heritage that is still prevalent within Italian culture but endorsed differently by geographic area. At the national level, the Demochristian Party led the government until 1993, interrupted only for 4 years then followed by another conservative party, occasionally spaced out by one or two years of Centre-left wing governments. This could likely have had an impact on the establishment of many Italian institutions and policies that have an important role in transmitting gender relation models . In 2014, a new Centre-left wing government introduced several structural reforms. Future studies will be able to assess their impact on the modernization of the country and on the Italian gender culture. In addition, in the last decades, a gradual reform process of the welfare system is leading to a regional differentiation of welfare functions, with the result that the institutional support for dual earner families may differ by region, generally with a bit more generous provisions for child care services and work-life balance policies in the North and in the Centre. Will this reinforce the regional differences in gender relation patterns? Finally, according to further contextual elements could be searched for in 'popular culture', concerning the impact of mainstream gender roles spread by mass-media. Italian television commercials had great success between the 80s and 90s in advocating females as idealized perfect housewives or, as counterpart, objectified women and this still characterizes mainstream female models in advertisements as well as national and local TV entertainments ). According to , popular culture can be considered as a component of the period effect. Therefore, the reinforcement of traditional gender role attitudes could be the case of a joint effect of conservative policies and the 'popular culture' that, even if affecting the whole country, shows strong regional differences given by history. No potential conflict of interest was reported by the author. Dr Attitudes, beliefs, and childbearing The New Population Problem: Why Families in Developed Countries Are Shrinking and What It Means Support for traditional female roles across 32 countries: Female labour market participation, policy models and gender differences' Comparative Sociology Tre Italie: la problematica territoriale dello sviluppo italiano Scelte lavorative e familiari nella Valdichiana senese Gender attitudes and modernization processes International Journal of Public Opinion Research Change in gender-ideology, 1977-1996: The contributions of intracohort change and population turnover Journal of Marriage and Family The transformation of US gender role attitudes: Cohort replacement, social-structural change, and ideological learning Social Science Research The social and ideological bases of middle-class political realignment in the United States American Sociological Review Attitudes towards gender roles in cross-cultural surveys: Content validity and cross-cultural measurement invariance Social Indicators Research The end of the gender revolution? Gender role attitudes from American Journal of Sociology European values study longitudinal data file 1981-2008 (EVS 19812008)', GESIS Data Archive, Cologne. ZA4804 Data File Version EVS Methods for estimating cohort replacement effects Sociological Methodology Where does social change come from? Population Research and Policy Review Ofelia e Parsifal: modelli e differenze di genere nel mondo dei media L'arzdaura: donne e gestione familiare nella realta contadina The silent revolution in Europe: Intergenerational change in postindustrial societies American Political Science Review The Silent Revolution: Changing Values and Political Styles Among Western Publics The persistence of materialist and post-materialist value orientations: Comments on Van Deth's analysis European Journal of Political Research Modernization and Postmodernization: Cultural, Economic, and Political Change in 43 Societies Changing values among western publics from 1970 to West European Politics International Social Survey Programme: Family and Changing Gender Roles I -ISSP 1988. GESIS Data Archive, Cologne. ZA1700 Data File Version International Social Survey Programme: Family and Changing Gender Roles II -ISSP 1994. GESIS Data Archive, Cologne. ZA2620 Data File Version Serie Storiche Country differences in sex-role attitudes: Cultural and economic explanations The Cultural Diversity of European Unity: Findings, Explanations and Reflections from the European Values Study Trasformazioni del lavoro nella contemporaneità. Gli uomini nei lavori «non maschili Gender cultures and the gender arrangement -A theoretical framework for cross-national gender research', Innovation: The European Journal of Social Science Research Socio-historical paths of the male breadwinner model -An explanation of cross-national differences The British Journal of Sociology The cohort as a concept in the study of social change American Sociological Review Modernization and Variations in Emancipative Values in European Societies in 1995-2008: Test of Inglehart's Socialization Hypothesis SSRN Scholarly Paper ID 2479869 Essere Uomo, Essere Donna Beliefs about women's labour in the reunified Germany European Sociological Review Separate spheres or increasing equality? Changing gender beliefs in postwar Japan Journal of Marriage and Family Generational changes in gender-role attitudes: Britain in a cross-national perspective Sociology The role of family policy institutions in explaining gender-role attitudes: A comparative multilevel analysis of thirteen industrialized countries Journal of European Social Policy Ma le donne no Age and sex differences in gender role attitudes in Luxembourg between Work, Employment & Society Religion and gender across Europe Social Compass Value Contrasts and Consensus in Present-Day Europe : Painting Europe's Moral Landscapes, European Values Studies 15 Trends in gender beliefs in Romania Current Sociology The Sociology of Gender: An Introduction to Theory and Research Rich Democracies: Political Economy, Public Policy, and Performance World Values Survey Association (www.worldvaluessurvey.org) Aggregate File Producer: Asep/JDS Gender role attitudes in Italy: 1988-2008. A path-dependency story of traditionalism Vera Lomazzi vera.lomazzi@gesis.org GESIS -Leibniz Institute for Social Sciences Unter Sachsenhausen 6-8 50667 Cologne Germany GESIS -Leibniz Institute for Social Sciences Cologne Germany Gender role attitudes in Italy: 1988-2008. A path-dependency story of traditionalism 10.1080/14616696.2017.1318330 GROBID - A machine learning software for extracting information from scholarly documents Gender role attitudes modernization traditionalism intracohort change Italy path-dependency Considering gender role attitudes as part of a broader cultural change related to the modernization process, this study adopts a path-dependency approach to analyze the support for the role of women in the public sphere in Italy since 1988. Modernization processes varied across Italian regions and the paper explores how different gender patterns developed accordingly. Using pooled data from European Values Survey, World Values Survey, and International Social Survey Program, the author assesses if this specific change is part of the postmaterialist shift and investigates the mechanisms of change carrying out cohort decomposition methods. The results address a reinforcement of traditionalism mainly due to the period effect that shows regional differences given by history. Introduction In the last decades, European societies have experienced a cultural shift toward an increase in egalitarian values that reflect changing attitudes toward gender roles. In this context, Italy, with its familistic and traditional culture regarding the societal role of women, represents a good case for investigating change in gender role attitudes through a path-dependency approach. In this paper, I aim to assess whether the postmaterialist shift is taking place in Italy and if variation in support for the role of females in the public sphere is part of a broader cultural change related to general postmaterialist values. Modernization processes varied across Italian regions and here I also investigate if different gender cultures developed accordingly. The analyses present a controversial situation: support for the role of females in the public sphere does not mirror change in variation in generic postmaterialist human values and, overall, the shift toward more egalitarian and postmaterialist values is not developing as one would expect. On the contrary, the results demonstrate a reinforcement of materialist/traditionalist values mainly due to the intracohort change. In the final section of the paper, I discuss these unexpected results. A theoretical framework Change in support for the role of women in the public sphere Gender role attitudes can be defined as the cognitive representation of what is believed appropriate for male and female roles in a specific cultural context, rooted in historic development and differs across societies (Alwin 2005; Constantin and Voicu 2014; Schultz Lee et al. 2010) . This representation can refer to the domestic sphere, expressing the preference for a certain family model, or the public sphere, showing more or less support for female or male roles in the fields of education, politics, or the labor market. Despite the public debate on this topic, there is a lack of knowledge regarding Italian attitudes on the role of women in the public sphere and how this changes over time. This paper aims to fill this gap by analyzing two specific aspects: support for the dual role of women (as workers and caregivers) and support for equality in the labor market. In this study, I adopt a societal-oriented approach to read the change in support for female roles in the public sphere as part of a broader cultural and social transformation, connected with modernization processes that lead to transformations in the social structure. As pointed out by several authors (André et al. 2013; Bergh 2006; Inglehart 1977; Kalmijn 2003; Valentova 2013; Voicu and Constantin 2014; Wilensky 2002) , modernization factors such as higher educational levels among the population, the rise of female labor market participation, and secularization are strictly connected with the shift toward more egalitarian gender values both at the individual and societal levels: people that directly experience higher female presence in the public space tend to assume more egalitarian values and, at the same time, a postmaterialist culture tends to morally legitimate and promote greater womens involvement in public life. This impacts on the individuals' socialization context and in their direct experience of gender relations, changing in this way the social representation of 'proper roles' for men and women (Brooks and Bolzendhal 2004) . The cultural shift toward egalitarian/postmaterialist values can be explained through societal level perspectives as those proposed by Inglehart (1971) and Ryder (1965) . Inglehart's theory of value change (1971) is based on the idea that when individual survival needs can be taken for granted, room for immaterial needs such as autonomy and selfrealization is then possible. In the case of gender relations, fulfilling materialist needs (with the necessity of a sexual division of work for the survival of the individual and family) opens space for postmaterialist needs including self-realization for all family members, pushing toward a more egalitarian perspective. Therefore, individuals define their priorities according to their immediate needs ('scarcity hypothesis'). This change is affected by the existential condition in that specific period. Nevertheless, this change is not immediate but gradual and reflects the conditions faced by people during their formative years. The change in values is then led by the cohort succession ('socialization hypothesis'). Ryder's work (1965) is extremely helpful to disentangle these two aspects. He addresses the relation between social change and two population processes: demographic metabolism and individual life cycles considered as aggregated. The first one, related to the 'cohort replacement', refers to the dynamics of fertility and mortality with respect to the possibility of each society to renew itself, generation by generation, where new individuals interact with the contemporary cultural heritage and enact it in a new way. The arrival of a new cohort makes social change possible. Italy is a country with an aging population and a weak cohort replacement: people older than 64 years old represent an increasing part of the population, while those below 14 years old are decreasing together with a fertility rate barely steady around 1.3 (see Figure A1 in the Appendix). However, each generational cohort lives its life cycle in a specific historical period, characterized by political, economic and social issues that affect the interpretation of life stages, producing different experiences regarding the transition to adult life, work and family, and public and private gender roles. The recent literature refers to this process as the 'period effect' or 'intracohort change' (Brooks and Bolzendahl 2004; Firebaugh 1992; Schultz Lee et al. 2007; Voicu and Tufiş 2012) . Did the modernization processes affect the gender role patterns and the shift toward egalitarian/postmaterialist values in Italy? Economic development and gender role patterns in Italy Postmaterialist theory (Inglehart 1971 (Inglehart , 1977 (Inglehart , 1997 (Inglehart , 2008 underlines how an economic situation defines the context for scarcity or prosperity in living conditions, and how this impacts human values. But the economic context also has an important role in understanding patterns of gender relations established in a specific context and period. Pfau-Effinger (2004) examines that prevailing gender role patterns are strictly connected to the dominant production model in a particular context. According to Pfau-Effinger (2004) , three main patterns of gender relations can be found in Europe. The first one is based on the well-known male breadwinner/female homemaker model, a result of Western capitalism and modernization processes that led to the establishment of the urban bourgeoisie with clearly separated spheres of life for men (public) and women (domestic), based on an idea of complementarity. This model is typical in many European countries though it formed in different historical periods with the establishment of the bourgeois class. The second pattern, prevalent in Eastern countries, is based on the dual-worker model. In this case, egalitarian values developed mainly with regard to the public dimension of life but not in the private one, as explained by Voicu and Tufiş in the Romanian case (2012). Finally, the family-economy model refers to the family-economy model refers to the organization of work and familial relations in rural areas where the value given to work carried out by women and men is the same because both contribute to the survival of the family: the relationship between partners is then based on a mutual dependency. To explain this model, Pfau-Effinger (2004) refers to Finland. Italy presents a story of a fragmentary industrial and cultural path that reflects the development of gender role patterns. Since the Unification of Italy in 1861, policy-makers have addressed building a common Italian culture. Even today Italy's gender culture is typically oriented toward a traditional division of work between men and women based on a complementarity gender role model (Perra and Ruspini 2014) . The origin of this model is deeply rooted in the history of the country and institutionally promoted. For example, in the Italian fascist era female roles in the public sphere were highly discouraged. Even if there was some maternity protection introduced in this period, other sets of rules excluded or limited female presence in the labor force to jobs mainly related to caregiving. In the following decades, women's opportunities improved but still gender cultures are affected by the fact that all the economic, labor, and family policies have been generally implemented regardless of the previous local history that led to different modernization processes in Italian areas. Across Italy, despite agriculture being the main economic sector in the twenty-first century, three main economic models can be detected according to the political powers that according to the political powers that ruled the states and city-states that eventually merged into the modern nation of Italy in 1861. Since that time, the development of Italy knows three different rhythms, speeds, and peculiarities rooted in the different historical pathways that characterized the so-called Three Italies (Bagnasco 1977) . In the North, thanks to the political strategies of the Austrian and Pedmont governments as well as the introduction of new technologies, agriculture became intensive and marketoriented. The Centre, ruled by the papal state, tended to underestimate the manufacturers from small inland villages and the potentiality of strategic harbors on the coasts and the society kept its sharecrop structure. The South, under the rule of the Kingdom of the Two Sicilies, remained anchored to the latifundium feudal regime. According to Pfau-Effinger's view (2004) , because of capitalism's structuring of the economy together with urbanization, the North is the typical scenario for the establishment of the urban bourgeoisie with gender role patterns based on complementarity, because of the capitalist address given to the economy together with the urbanization. In the sharecrop system of the Centre, the model is close to that of family-economy even if patriarchal structures remain and women's lives are strictly bound to men. Nevertheless, in this type of society, the older female of the family could have a strong authority on the other women. It is the case of the 'massaia' in Tuscany (Banini 2013) or the 'arzdaura' in the region of Emilia Romagna (Guidicini and Alvisi 1994) , who managed all the female activities and domestic chores. At the same time she worked in the fields for the crop and grape harvests. The southern economic system typically employed seasonal workers, mainly employing only men. Women's activities remained centered in the domestic space (Scaraffia 1988) . At the end of the twenty-first century, modernization processes led to the stabilization of the manufacturing sector, which gave job opportunities to women in the North and less in the Centre. In the South, the economy remained linked to the agriculture system and the gender role patterns did not change much. According to these different paths, the acceptability of certain male and female roles may therefore differ within Italian areas. In more recent times, the three different modernization paths are still visible. We could consider three modernization factors related to changing gender role attitudes in the frame of modernization processes: the rise of female employment, increase in the population's level of education, and secularization. Graphs in Figure 1 (a) and 1(b) show the labor market participation for men and women 24-54 years old by geographical area in the period 1993-2008 (data refer to the historical series available from the Italian National Institute of Statistics -ISTAT 2016). The 'three Italies' differ in the level of employment rates among the active population both in the male and female cases. But, while women who live in the North and Centre steadily increase their economic participation, those who live in the South show a more static trend. On the contrary, male economic participation in the South steadily drops since 2002. In the other areas, men who participate in the labor force slightly increases, even if there are some fluctuations. According to Inglehart's 'scarcity hypothesis', one could expect that men living in the South of Italy show a shift toward more traditionalist and materialist values, as a consequences of their reduced economic participation, and they may see the increase of working women as a growth of competition, as happened in the case of Germany after the reunification (Schultz Lee et al. 2007) . Figure 2 (a) and 2(b) refers to the proportion of graduates among men (a) and women (b) aged 25 in the three areas. In all cases, we can see a similar dynamic with a growth in education levels until 2005. Then it dramatically drops. However, the density of those highly educated is much greater in the Centre, especially among women (15.5% in 1992, 64.5% in 2005, 57.8% in 2008 -men 14.5% in 1992, 46.6% in 2005, 48.6% in 2008) , than in the other areas. In particular, the South presents the lowest level of graduates both for men (7.2% in 1992, 23.21% in 2005, and 21.3% in 2008) and women (8.2% in 1992, 35.7% in 2005, and 34.1% in 2008) . The structural explanation (Bergh 2006) argues that the shift toward more egalitarian positions is also due to the fact that women and men increasingly study alongside each other and in an increasing number of fields and levels. One could therefore expect that the increase in education levels among the Italian population reflects in more egalitarian and postmaterialist values. This change would be less relevant in the South and more evident in the Centre. Secularization is often a relevant modernization factor affecting gender role attitudes and basic human values (Kalmijn 2003; Voicu 2009) . Processes related to urbanization and industrialization contributed to reducing the normative authority of religious institutions which have historically supported gender role patterns based on the complementary functions of reproduction (women) and production (men). As an indicator of the process of secularization, variation in the proportion of people over the age of 6 out of the entire population who never attend religious services was selected. Figure 3 (a) and 3(b) shows these trends respectively for males and females. The dynamics look similar for both: in the North and Centre the secularization process is not linear and presents recurring fluctuations. In the South, these fluctuations appear less evident and the amount of people who never attend religious services is strongly lower. In all areas, the secularization process involves men more than women, who, generally speaking, are still more religious. In the South, for example, males who never attend religious services in 1995 pass from 12% to 17.2% in 2008; women from 5.2% in 1995 to 9.4% in 2008. In the North, the proportion of non-attending males increases from 21.6% in 1995 to 27.1% in 2008 while for the females it rises from 13.5% to 18.8%. In the Centre, the percentage for men goes from 19.5% in 1995 to 28.1% in 2008, for women from 12% to 17.2%. However, these trends are not indicative of a dramatic change with regard to the secularization process; therefore, one could expect that this modernization factor would not particularly affect the change in gender role attitudes, especially in the South and for women. Hypotheses I assume that the change in a specific domain of human values, such as the support for female roles in the public sphere, is part of a broader cultural change, which is the shift toward postmaterialist values. So I suppose that: H1a: The variation in support for the role of females in the public sphere follows a dynamic similar to the variation in the postmaterialist position. H1b: The composition of the mechanisms of social change (intracohort change and cohort replacement) involved in the variation of the support for the female role in the public sphere is similar to those of the variation in the postmaterialist position. The economic situation of Italy, delineated by the macro indicators, does not refer to a period of scarcity between 1988 and 2008: I therefore expect that: H2: The results show a shift toward more egalitarian attitudes and postmaterialist values. Considering the demographic dynamics developed in Italy, with the slow cohort turnover and the empirical studies based on Ryder's perspective (Ryder 1965; Schultz Lee et al. 2007; Schultz Lee et al. 2010; Voicu and Tufiş 2012) , I suppose that: H3: The intracohort change is the prevalent mechanism of change, while the cohort replacement, even if present, has a lower impact. Finally, according to the different historical pathways of economic development and gender role patterns of the 'Three Italies' as well the recent dynamics of the modernization factors shown, I expect that: H4a: Men in the South reinforce their materialist values and traditional gender role attitudes because of the drop in economic opportunities and the lower level of tertiary educated people. This is not the case for men living in the other areas. H4b: Women increase their egalitarian attitudes and generic postmaterialist values because of their increased economic participation and above all because of the growth of women with tertiary education among the population, in particular in the Centre. H4c: The secularization process is not affecting the variation in support for the female role in the public sphere, nor the shift toward postmaterialist values because its development is not dramatic, in particular for women and especially in the South of Italy. Strategy of analysis: data and methods Data In order to test my hypothesis, I use pooled data collected in Italy in six observation points by different cross-national and repeated surveys: 1988 and 1994 from the International Social Survey Program (ISSP Research Group 1990 , 1994 , 2005 from the World Values Survey (WVS 2008 (WVS ), 1990 (WVS , 1999 (WVS , and 2008 from the European Values Survey (EVS 2011). Table 1 shows the details of the sample sizes. To assess if the change in gender role attitudes is part of the wider cultural change, I compare the trend and the decomposition of the variation in support for the role of females in the public sphere to the variation of the postmaterialist index (Inglehart 1971 (Inglehart , 1977 , chosen as indicator of a general values system. I examine two aspects of the egalitarian support toward the female public role. The first is fully referred to as the public sphere and measures the support for equality in the labor market, tapped by the item 'When jobs are scarce, men have more right to a job than women' collected in 1990 (EVS), 1999 (EVS), 2005 (WVS) and 2008 (EVS). In all the surveys, the respondents can choose between three categories of response: 'agree', 'disagree', 'neither agree nor disagree'. The second dimension refers to the support for the dual role of women as worker and caregiver and it is covered by the item 'A job is alright but what most women really want is a home and children', available in 1988 (ISSP), 1994 (ISSP), 1999 (EVS), and 2008 (EVS). One could say that this dimension is between the public and the domestic domains if compared to the previous item. So together they allow for grasping different nuances in support for women in the public sphere. This item does not have the same response categories in all the surveys. In EVS, they are: 'agree strongly', 'agree'; 'disagree', 'disagree strongly'. ISSP has an additional middle category: 'not agree nor disagree'. For both dimensions, I considered only the disagreement categories, here indicating egalitarian support/postmaterialist values, recoded as dummies Table 1 . Sample size by context and year. 1988 ISSP 1990 EVS 1994 ISSP 1999 EVS 2005 WVS 2008 (disagree = 1; 0). In the case of support for the dual role, the two categories 'very disagree' and 'disagree' have been previously collapsed together. The 4-item Post-materialism index is available for 1990 (EVS), 1999 (EVS), 2005 (WVS), and 2008 (EVS). It is built on the basis of the national priorities assigned by the respondents and has three categories: 1 = materialist; 2 = postmaterialist, and 3 = mixed. This study considers only the postmaterialist position and transformed the index in a dummy variable (1 = postmaterialist, 0 = other). Tables 2-4 show the variation in support for equality in the labor market, for the dual role of women and in the postmaterialist position from 1988 (1990 when data are not earlier available) to 2008. On the one side, support for equality in the labor market increases over time for both men and women. This trend perfectly adheres to the postmaterialist theory. But, on the other side, variations in the postmaterialist position and in support for the dual role of women drops apparently until the end of the 1990s and starts to slightly increase in 2008. The graphs in the appendix ( Figure A2 ) provide visual information about these dynamics. This controversial figure tells us that in Italy the postmaterialist transition is not taking place in the way shaped by Inglehart's perspective. Even if the economic situation does not look to push toward a loss of prosperity, the variations show a reinforcement of material values and conservative gender role norms in the domestic sphere, while gender role attitudes concerning the public sphere appear more egalitarian. The decomposition of the components of social change will assess these patterns. Methods I adopt decomposition modeling (Firebaugh 1992) to decompose the intracohort change and the change due to processes related to the cohort replacement, comparing a generic postmaterialist position and a specific domain of human values, that is attitudes toward female roles in the public sphere. According to Firebaugh (1992: 2) , this kind of analysis does not require panel data and repeated cross-section data are suitable, assuming that among the same birth cohort the probability of death is not related to variable of interest. Therefore, the use of pooled data is applicable. Firebaugh's linear decomposition model (1989, 1992) originally employs OLS regressions. Here, following the example of previous studies (Brooks and Manza 1997; Voicu and Tufiş 2012) , the model is adapted for dichotomous variables. I perform the same modeling for the two aspects of support for women in the public sphere and for the postmaterialist position, separately for women and men and for geographic areas. The basic assumption of this technique is that the social change occurred between the final moment of observation and the initial one, due to the sum of the intracohort change, the cohort replacement and their joint effect (e). The modeling consists in sequential logistic regressions. In the first one, only the effects of period and cohort are tested (M0). In the next steps, control variables can be sequentially added. Here I controlled for modernization factors that are relevant to the change in gender role attitudes: the increase in educational level, labor market participation, and secularization (Schultz Lee et al. 2007; Voicu and Tufiş 2012) . I firstly added the educational level (M1), and secondly the labor market participation (M2). Finally, I added religiosity as an additional control variable. Independent and controlling variables, collected in all the surveys, have been harmonized as follows: -Period variables refer to the survey year (ISSP88, EVS90, ISSP94, EVS99, WVS05, EVS08), the first conducted as reference. -Educational level presents three categories: low (below low secondary school), middle (secondary school) and high (tertiary and upper) levels. These levels are recoded in dummies and the lowest is the reference. -Employment status differs in: full-time worker, part-time worker (less than 30 hours/week); non-paid worker (students, housewives, retired, unemployed...) . Also in this case the variable is transformed into dummies and the not working category is adopted as reference. -To indicate religiosity, the variable regarding attendance at religious services has been chosen and firstly recoded in four categories: high attendance (at least once a week), moderate attendance (two or three times a month), rare attendance (one or two times a year), and never. Then these categories are recoded as dummies. People who never go to religious services are considered as reference in the analyses. The logistic regression equation used is: ln (  l −  ) =  0 +  1 survey year +  2 birth year In this case, the logistic regression esteems the probability  of assuming postmaterialism/egalitarian. The amount of change due to processes related to the intracohort mechanism is computed as a product of the survey year effect and the length of the period of observation; the amount of change due to the cohort replacement results from the product of the birth year effect and the difference between the average of birth year at the final point of observation and the initial one. In the procedure applied, these components have been computed in each step. The overall social change is defined by the difference between the logic at the final observation (2) and the initial (1): CS = [ln (  1 −  )] 2 − [ln (  1 −  )] 1 Results Results from the decomposition modeling confirm the ambivalent shift earlier detected by observing the variation in postmaterialist values and gender role attitudes. The logistic regression results are available on request. The mechanisms of change in support for equality in the labor market appear different between men and women (Table 5(a) and 5(b)). Testing for the period effect and the cohort effect (M0), for men living in the North and in the South of the country, the intracohort change prevails on the cohort replacement. On the contrary, for women and men living in Central Italy, the change registered is due only to the cohort replacement: the period effect in these cases is not significant. This means that here the increase of support for equality in the labor market is not related to the contextual circumstances but only because new cohorts, those more educated and actively participating more in the labor force, are slowly replacing the older. So, my third hypothesis is confirmed for men living only in the North and South. Controlling for education (M1) in all these groups, the size of the cohort replacement component decreases consistently. In the case of men living in the Centre and Northern women, economic participation (M2) seems to have a relevant impact, while for the women who live in the Centre and South this is marginal. This can be explained by the fact that in these areas the growth of female economic participation is very slow. When controlling for religiosity (M3), the situation slightly differs, this can be interpreted as a confirmation of the hypothesis H4c that assumes that this process is too slowly developed in Italy to affect this social change. For men living in the North and South (Table 5(a)) , both the mechanisms contribute to social change. In this case, male support for equality in the labor market is affected by the contextual situation as well by the replacement of the old cohort by younger and more educated cohorts. In effect, only the educational level (M1) seems to make some differences: controlling for labor market participation (M2) and religiosity (M3), the composition of the change remains the same. The decomposition results of the variation in support for the dual role of women reveal that despite the dynamics shown in the variation's trends (Table 3 and Figure A2 ), the overall social change for men living in the Centre and South, and for women living in the Centre is not significant (Table 6(a) and 6(b)). In the other cases, results refer to a negative change, that is, the drop in egalitarian support earlier observed in this dimension. Testing the effect of survey year (period effect) and respondent's birth year (cohort effect), in M0 it is clear that for all the groups considered, this drop is due to the period effect that in the case of men is two times the cohort replacement component. Also for women, the intracohort change is 1.5 times higher than the cohort replacement. Controlling for education (M1), we can see that this is a potential driver of change, especially for women in the South, where the size of the cohort replacement component passes from 0.546 to 0.348. The increased economic participation seems to affect only women who live in the North of the country, while controlling for religiosity does not make relevant changes. These results address education as the main predictor of the shift toward more egalitarian values. Nevertheless, the replacement of the older, less educated cohort, is not enough to set off against something happening in the context of Italy in the observed period. The analysis of the decomposition of the variation in the postmaterialist position overlaps with the decomposition results of the support for the dual role of women. Here the overall social change that occurred between 1990 and 2008 is significant only for men living in the North and in the South (Table 7 (a)) and for women living in the South (Table 7 (b)). In all the cases, the period effect is largely prevailing, in particular for men in the North. Education is the main predictor of change (M1) for both men and women, while the change in economic participation (M2) and religiosity (M3) between generations seems to not be relevant. Discussion This paper analyzed the change in two dimensions of support for female roles in the public sphere in Italy, assuming that the change in this specific domain of values is part of a broader cultural change. To consider this general change, I referred to the postmaterialist shift proposed by Inglehart (1971) . It briefly described the economic development of the country in consideration of three different Italian contexts, characterized by the gender role patterns developed according to the different historical pathways. The results refer to a controversial situation. First of all the assumption that changes in gender role attitudes are part of a wider cultural change is only partially confirmed. Despite that the support for the dual role of women as workers and caregivers presents trend and decomposition results similar to variation in the postmaterialist position, this is not the case for the support for equality in the labor market Moreover, only the latest dimension actually shows a growth in egalitarian support. This first result confirms that gender role attitudes are a complex concept and it is good practice to separately analyze the different dimensions that compose this concept, as already argued in previous studies (Voicu and Constantin 2014; Voicu and Tufiş 2012) . Probably this item, which is particularly related to economic issues and job opportunities, is more affected by the living condition that, as shown, in the period here considered do not refer to a period of scarcity and people tend to show more egalitarian values. Then, for men living in the South, who faced a drop in employment rate associated with the rise of female employment, the social change is smaller and this confirms my hypothesis 4a. But despite all the assumptions that drove my second hypothesis, results concerning variation in the postmaterialist position and support for the dual role of women address a worsening of egalitarian support and of postmaterialist values, both higher in the past. Differently from other controversial situations traced in the literature, where results addressed fluctuations (Brewster and Padavic 2000; Cotter et al. 2011; Scott et al. 1996) , backlash (Savelyev 2014; Schultz Lee et al. 2007) or persistence (Inglehart 1983) , here data refer to a reinforcement of the materialist/traditionalist values. What can explain this scenario? According to the decomposition results, the causes of this reinforcement should be searched for in contextual conditions. In fact, as expected according to my third hypothesis, the intracohort change is the prevalent mechanism of change, except from women and men living in the Centre for which the increase in support for equality in the labor market is explained only by the arrival of new, higher educated cohorts. For the variation in support for the dual role of women and the postmaterialist position, the effect of the period is dominant and the effect of increased educational level and female participation are not enough to balance the negative effect of the period, neither for women living in the Centre where these modernization factors are quite relevant, as supposed in H4b. In all the cases, including the change in religiosity did not make much difference in the model probably because, as argued in the last hypothesis H4c, this process is quite slow in Italy and non-religious people continue to be a minority. In this study, the period effect has been delineated mainly considering structural and economic components that should define the context wherein people define their own priorities that are then reflected in attitudes and values. The Italian demographic dynamics may suggest slow value changes because of the poor cohort turnover, but this could be addressed by the persistence of materialist/traditionalist values or by a very slow postmaterialist/egalitarian shift, rather than a materialist reinforcement. In the observed period, the economic circumstances do not refer to a drop in economic prosperity and this should support the postmaterialist/egalitarian shift. Further studies should investigate additional elements that characterized the cultural context between 1988 and 2008, for example, considering the effect of the Christian-Catholic heritage that is still prevalent within Italian culture but endorsed differently by geographic area. At the national level, the Demochristian Party led the government until 1993, interrupted only for 4 years (1983) (1984) (1985) (1986) (1987) then followed by another conservative party, occasionally spaced out by one or two years of Centre-left wing governments. This could likely have had an impact on the establishment of many Italian institutions and policies that have an important role in transmitting gender relation models (Pfau-Effinger 1998; Sjöberg 2004; Wharton 2005) . In 2014, a new Centre-left wing government introduced several structural reforms. Future studies will be able to assess their impact on the modernization of the country and on the Italian gender culture. In addition, in the last decades, a gradual reform process of the welfare system is leading to a regional differentiation of welfare functions, with the result that the institutional support for dual earner families may differ by region, generally with a bit more generous provisions for child care services and work-life balance policies in the North and in the Centre. Will this reinforce the regional differences in gender relation patterns? Finally, according to Cotter and colleagues' suggestions (2011) further contextual elements could be searched for in 'popular culture', concerning the impact of mainstream gender roles spread by mass-media. Italian television commercials had great success between the 80s and 90s in advocating females as idealized perfect housewives or, as counterpart, objectified women and this still characterizes mainstream female models in advertisements as well as national and local TV entertainments (Grossi and Ruspini 2007; Soffici 2010 ). According to Cotter and colleagues (2011) , popular culture can be considered as a component of the period effect. Therefore, the reinforcement of traditional gender role attitudes could be the case of a joint effect of conservative policies and the 'popular culture' that, even if affecting the whole country, shows strong regional differences given by history. Disclosure statement No potential conflict of interest was reported by the author. Notes on contributor Dr Figure 1 . 1 Labor market participation 1993-2008. (a) Men 24-54 years old by area. (b) Women 24-54 years old by area. Source: ISTAT -Italian National Institute of Statistics (2016). Figure 2 . 2 Proportion of (a) male graduates every 100 men aged 25 by geographical area and (b) female graduates every 100 women aged 25 by geographical area. Italy 1992-2008. Source: ISTAT -Italian National Institute of Statistics (2016). Figure 3 . 3 Proportion of (a) males and (b) females over 6 years of age who never attend religious services by geographical area out of 100 persons of the same subdivision. Italy 1995-2008. Source: ISTAT -Italian National Institute of Statistics (2016). Figure A2 . A2 Vera Lomazzi achieved her Ph.D. in Sociology and Social Research Methods at Catholic University of the Sacred Hearth, Milan (Italy) in 2015. As Quantitative Research Assistant she joined the ArabTransition project (FP7) at the University of Aberdeen (UK) where she still acts as Honorary Research Fellow. Since May 2016, she works at GESIS-Leibniz Institute for the Social Science in Cologne (DE) as PostDoc Researcher for the PROMISE project -PROMoting youth Involvement and Social Engagement: Opportunities and challenges for 'conflicted' young people across Europe (Horizon2020). Among her very recent publications: the commentary published for Journal of Global Faultlines and a peer reviewed monography (in Italian). Lomazzi, V. (2016) 'Middle East refugee response and gender equality challenges', Journal of Global Faultlines 3(1): 56-64. Lomazzi, V. (2016) Donne e sfera pubblica. I valori degli europei a confronto [Women and public sphere. Comparing Europeans' values], Milano: Franco Angeli. ORCID Vera Lomazzi http://orcid.org/0000-0003-2699-2768 Variations in support for equality in the labor market, support for the dual role of women and postmaterialist position 1988/1990-2008, by gender and area. Table 2 .Table 4 . 24 Support for equality in the labor market (%) by gender and context 1990-2008. Postmaterialist position (%) by gender and context 1990-2008. 1990 1999 2005 2008 Men North Italy 46.5 54.6 57.6 68.2 Centre Italy 41.7 59.5 52.2 62.8 South Italy 34.4 46.2 45.2 51.2 Women North Italy 58.4 61.5 65.2 73.8 Centre Italy 49.0 58.3 66.3 69.0 South Italy 46.7 55.3 58.7 61.2 Table 3. Support for the dual role of women as worker and caregiver (%) by gender and context 1988-2008. 1988 1990 1994 1999 2008 Men North Italy 42.8 29.9 33.9 28.9 29.8 Centre Italy 40.5 22.7 28.1 32.4 29.2 South Italy 31.1 26.7 32.4 22.1 16.1 Women North Italy 46.0 37.4 39.1 34.7 40.1 Centre Italy 37.9 28.2 41.2 30.6 38.6 South Italy 34.5 28.0 34.7 26.1 26.3 Table 5 . 5 Support for gender equality in the labor market 1990-2008: decomposition results for (a) men and (b) women. SC IC M0 CR IC M1 CR IC M2 CR IC M3 CR (a) North 0.906* 0.544 0.419* 0.595* 0.343* 0.563 0.372* 0.634* 0.358* Centre 0.857* 0.487 0.422* 0.478 0.336* 0.467 0.335* 0.472 0.372* South 0.693*** 0.409* 0.304* 0.542 0.210* 0.536 0.211* 0.531 0.229* (b) North 0.693* 0.294 0.450* 0.253 0.345* 0.295 0.279* 0.335 0.265* Centre 0.838* 0.299 0.587* 0.303 0.307 0.350 0.259 0.508 0.201 South 0.590* 0.236 0.466* 0.417* 0.264*** 0.422* 0.258*** 0.435* 0.254* Note: SC: social change; IC: intracohort change; CR: cohort replacement. p  .001. p  .01..01  p  .1. Table 6 . 6 Support for the dual role of women 1988-2008: decomposition results for (a) men and (b) women SC IC M0 CR IC M1 CR IC M2 CR IC M3 CR (a) North -0.568* -0.935* 0.470* -0.933* 0.381* -0.894* 0.357* -0.757* 0.313*** Centre -0.502 -0.480* 0.375*** -0.546* 0.318*** -0.501* 0.298* -0.312 0.299* South -0.855 -1.164* 0.364* -1.111* 0.244 -1.102* 0.237 -1.072*** 0.186* (b) North -0.243* -0.658* 0.380* -0.688* 0.252* -0.657* 0.203* -0.522 0.166** Centre 0.031 -0.539* 0.691*** -0.486* 0.566* -0.451 0.503* -0.145 0.441*** South -0.389* -0.728* 0.546* -0.673* 0.348* -0.691* 0.346* -0.564 0.327* Note: SC: social change; IC: intracohort change; CR: cohort replacement. *p  .001. p  .01. .01  p  .1. Table 7 . 7 Postmaterialist position 1990Postmaterialist position -2008: decomposition results for (a) men and (b) women SC IC M0 CR IC M1 CR IC M2 CR IC M3 CR (a) North -0.369 -0.704* 0.217* -0.677* 0.131* -0.632* 0.101 -0.592 0.099 Centre -0.183 -0.581* 0.189** -0.543 0.129* -0.543 0.124 -0.541 0.149* South -0.649* -0.805 0.275* -0.719 0.187* -0.718** 0.187* -0.716** 0.181* (b) North -0.169 -0.466* 0.361* -0.518 0.213 -0.515** 0.192* -0.495** 0.188* Centre -0.176 -0.811 0.450* -0.819 0.282 -0.801 0.266 -0.818 0.266 South -0.010* -0.384* 0.376*** -0.228* 0.224 -0.253 0.226 -0.251 0.239 Note: SC: social change; IC: intracohort change; CR: cohort replacement. p  .001. p  .01..01  p  .1. Appendix Figure A1 . Demographic trends in Italy 1988-2008. Source: The World Data Bank. Attitudes, beliefs, and childbearing D F Alwin The New Population Problem: Why Families in Developed Countries Are Shrinking and What It Means A. Booth and A. C. Crouter Mahwah, NJ Lawrence Erlbaum Associates Support for traditional female roles across 32 countries: Female labour market participation, policy models and gender differences' S André M Gesthuizen P Scheepers Comparative Sociology 12 4 Tre Italie: la problematica territoriale dello sviluppo italiano A Bagnasco Bologna: Il Mulino Scelte lavorative e familiari nella Valdichiana senese A Banini C&P Adver Effigi Arcidosso Grosseto Dopo la mezzadria Gender attitudes and modernization processes J Bergh International Journal of Public Opinion Research 19 1 Change in gender-ideology, 1977-1996: The contributions of intracohort change and population turnover K L Brewster I Padavic Journal of Marriage and Family 62 2 The transformation of US gender role attitudes: Cohort replacement, social-structural change, and ideological learning C Brooks C Bolzendahl Social Science Research 33 The social and ideological bases of middle-class political realignment in the United States C Brooks J Manza American Sociological Review 62 2 Attitudes towards gender roles in cross-cultural surveys: Content validity and cross-cultural measurement invariance A Constantin M Voicu Social Indicators Research 123 3 The end of the gender revolution? Gender role attitudes from D Cotter J M Hermsen R Vanneman American Journal of Sociology 117 1 European values study longitudinal data file 1981-2008 (EVS 19812008)', GESIS Data Archive, Cologne. ZA4804 Data File Version 10.4232/1.11005 EVS Methods for estimating cohort replacement effects G Firebaugh Sociological Methodology 19 Where does social change come from? G Firebaugh Population Research and Policy Review 11 1 Ofelia e Parsifal: modelli e differenze di genere nel mondo dei media G Grossi E Ruspini Cortina Libreria Milano L'arzdaura: donne e gestione familiare nella realta contadina P Guidicini C Alvisi FrancoAngeli Milano The silent revolution in Europe: Intergenerational change in postindustrial societies R Inglehart American Political Science Review 65 4 The Silent Revolution: Changing Values and Political Styles Among Western Publics R Inglehart Princeton University Press Princeton, NJ The persistence of materialist and post-materialist value orientations: Comments on Van Deth's analysis R Inglehart European Journal of Political Research 11 1 Modernization and Postmodernization: Cultural, Economic, and Political Change in 43 Societies R Inglehart Cambridge University Press 19 Cambridge Changing values among western publics from 1970 to R Inglehart West European Politics 31 1-2 International Social Survey Programme: Family and Changing Gender Roles I -ISSP 1988. GESIS Data Archive, Cologne. ZA1700 Data File Version Issp Research Group 10.4232/1.1700 International Social Survey Programme: Family and Changing Gender Roles II -ISSP 1994. GESIS Data Archive, Cologne. ZA2620 Data File Version Issp Research Group 10.4232/1.2620 Serie Storiche ISTAT -Italian National Institute of Statistics Country differences in sex-role attitudes: Cultural and economic explanations M Kalmijn The Cultural Diversity of European Unity: Findings, Explanations and Reflections from the European Values Study W. Arts and L. Halman Leiden Brill Trasformazioni del lavoro nella contemporaneità. Gli uomini nei lavori «non maschili M S Perra E Ruspini FrancoAngeli Milano Gender cultures and the gender arrangement -A theoretical framework for cross-national gender research', Innovation: The B Pfau-Effinger European Journal of Social Science Research 11 2 Socio-historical paths of the male breadwinner model -An explanation of cross-national differences B Pfau-Effinger The British Journal of Sociology 55 3 The cohort as a concept in the study of social change N B Ryder American Sociological Review 30 6 Modernization and Variations in Emancipative Values in European Societies in 1995-2008: Test of Inglehart's Socialization Hypothesis Y Savelyev SSRN Scholarly Paper ID 2479869 Rochester, NY Social Science Research Network Essere Uomo, Essere Donna L Scaraffia P. Melograni (ed) La Famiglia Italiana dall'Ottocento a Oggi Bari: Laterza Beliefs about women's labour in the reunified Germany K Schultz Lee D F Alwin P A Tufiş European Sociological Review 23 4 Separate spheres or increasing equality? Changing gender beliefs in postwar Japan K Schultz Lee P A Tufiş D F Alwin Journal of Marriage and Family 72 1 Generational changes in gender-role attitudes: Britain in a cross-national perspective J Scott D F Alwin M Braun Sociology 30 3 The role of family policy institutions in explaining gender-role attitudes: A comparative multilevel analysis of thirteen industrialized countries O Sjöberg Journal of European Social Policy 14 2 Ma le donne no C Soffici Feltrinelli Editore Milano Age and sex differences in gender role attitudes in Luxembourg between M Valentova Work, Employment & Society 27 4 Religion and gender across Europe M Voicu Social Compass 56 2 Value Contrasts and Consensus in Present-Day Europe : Painting Europe's Moral Landscapes, European Values Studies 15 M Voicu A Constantin W. Arts and L. Halman Brill Leiden Attitudes towards gender roles in Europe: Modernization and social institutions Trends in gender beliefs in Romania M Voicu P A Tufiş Current Sociology 60 1 The Sociology of Gender: An Introduction to Theory and Research A S Wharton Blackwell Oxford Rich Democracies: Political Economy, Public Policy, and Performance H L Wilensky University of California Press Berkeley World Values Survey Association (www.worldvaluessurvey.org) Aggregate File Producer: Asep/JDS Madrid, Spain 5 official aggregate v (xsd:string)
?:linkScore	0.0 (xsd:float)
?:linkSource	GEFUREX pipeline (xsd:string)
?:linkingMethod	search (xsd:string)
?:mainEntity	<https://data.gesis.org/gesiskg/resource/study_program_90>
?:name	WVS (xsd:string)
is ?:referenceMetadata of	<https://data.gesis.org/gesiskg/resource/ssoar_reference_gesis-ssoar-60934_study_program_90>
?:toSource	gesis (xsd:string)
rdf:type	<https://data.gesis.org/gesiskg/schema/LinkMetadata>