A Quarter Century of Change in Family and Gender-Role Attitudes in Hungary A Quarter Century of Change in Family and Gender-Role Attitudes in Hungary* Zsolt Spéder Abstract: Our study examines how attitudes towards family and gender roles have changed since the ultimate collapse of communism in Hungary. With respect to evaluating the effects of the regime change, it is important to note that Hungary is unique in having pre-transition measures on attitudes from the International Social Survey Program. In analyzing the nature of value shifts, an arithmetic method that decomposes the changes into population turnover and individual (period) components is used. According to the results, period effects  uctuated over the quarter of century, while the population turnover effects point continuously and clearly towards liberalization of family and gender-role attitudes. Since the period effects were usually stronger, they shaped the  uctuating nature of overall change. Namely, there is a clear trend towards re-traditionalization immediately following the regime change and liberalization thereafter, although there are also signs of continued support for traditional values. The series of repeated modules of the ISSP allowed us to examine a key premise of the Second Demographic Transition (SDT) theory in the case of Hungary. We concluded that the detected direction of the attitude change does not support the examined premise of the SDT. Keywords: Gender role · Attitude change · Regime change · Second Demographic Transition theory 1 Introduction It is a commonplace that in the time from 1989/1990 onwards, profound changes altered all aspects of life in the societies of post-communist countries. In a nutshell, a democratic system based on free elections and a market economy has replaced Comparative Population Studies Vol. 48 (2023): 741-770 (Date of release: 20.12.2023) Federal Institute for Population Research 2023 URL: www.comparativepopulationstudies.de DOI: https://doi.org/10.12765/CPoS-2023-29 URN: urn:nbn:de:bib-cpos-2023-29en0 * This article belongs to a special issue on “Demographic Developments in Eastern and Western Europe Before and After the Transformation of Socialist Countries”. • Zsolt Spéder742 the command economies and one-party political systems (Kornai 1992; Offe 1994). Are we witnessing equally profound changes in family life – more speci cally in values related to gender roles, and attitudes towards the family? Institutional and structural changes have probably not left beliefs and ideas untouched, but there is also good reason to believe that people’s mindsets change little or only slowly in the short run (Dahrendorf 1990). The aim of our analysis is to describe the changes in attitudes towards gender roles, family life, partnership forms and childbearing over a quarter of a century. Furthermore, using a decomposition procedure we aim to identify the extent to which the changes can be linked to the political-economic regime change or to historical events (periods), and whether and to what extent population turnover played a role in the changes. Since the 1970s, attitudes towards gender roles and family life in the USA and other Western countries have been characterized by the powerful spread of the idea of gender equality, more liberal expectations regarding family life, and increasing tolerance towards atypical family forms (Thornton/Young-DeMarco 2001; Scott 2006; Brooks/Bolzendahl 2004). However, after the turn of the millennium the tendency peaked (Thornton/Young-DeMarco 2001), and in fact, a reversal took place in the USA and some Western European countries. A US study even proposes the “end of the gender revolution” (Cotter et al. 2011), while a Western European comparative analysis reveals a “conservative turn” in the 1990s (Scott 2006). Findings regarding post-communist countries are non-conclusive (Chromkova Manea/Rabusic 2020). While several studies indicate a shift towards liberalization (Rabusic 2001; Voicu/Tu s 2012; Lee et al. 2007), there are others that reveal stable attitudes (Schwartz et al. 2000), or a shift towards traditional gender-role expectations (Spéder/Kapitány 2014). Research on attitudinal change has a tradition in its own right, but it is of particular relevance for demographic analysis. In his comprehensive review, van de Kaa discussed cultural/ideational approaches as one of seven research traditions in the understanding of fertility development (van de Kaa 1996). According to the most popular current theory, the Second Demographic Transition (SDT) theory, value change is the driver of demographic behavior. Values play a key role in the “developmental idealism” approach (Thornton 2005), and they are also present in the “anomie” approach employed for understanding changes in the post-communist countries (Philipov et al. 2006). Nineteen country studies, published in the early 2000s as a Special Collection of the journal Demographic Research, edited by Tomas Frejka and Jan Hoem (Frejka 2008; Hoem 2008), discussed long-term trends in fertility using a similar structure and standard indicators. All but three of these studies, to a greater or lesser extent, have addressed the role of value change. Of the 19 studies, 12 explicitly touch on SDT, and for 10 studies SDT theory was considered a relevant approach for understanding changes. Two country studies (France, Ukraine) conclude that SDT is not a suitable explanatory framework. This clearly underlines the fact that research on attitudinal and value change should be given more emphasis in population research than is currently the case (cf. Liefbroer/Billari 2010). A Quarter Century of Change in Family and Gender-Role Attitudes in Hungary • 743 In this paper we will give a comprehensive and detailed answer to the question of how attitudes towards family and gender roles have changed in Hungary during the quarter of a century since the end of the communist system and following the change of regime. This is made possible by the four available “Family and changing gender roles” modules of the International Social Survey Program (ISSP), which will be analysed using a method that is capable of differentiating between cohort and individual (period) effects (Firebaugh 1997). We focus on Hungary because data that allows a consistent comparison of the times before and after the regime change exists only in the case of this country. However, there are signs that our results can be generalized to other post-communist countries. First, a number of analyses have concluded that 40 years of political control and command economy have left their mark on the world of values (Inglehart/Baker 2000). Furthermore, studies on attitudes towards gender roles in the family have concluded that attitudes in the former socialist countries are similar, well separated and still distinct from the general attitudes in Western European countries (Lück/ Hofäcker 2003; Harrison/Fitzgerald 2010; Chromkova Manea/Rabusic 2020). We do not deny, however, that there are differences within the former socialist countries (Lück/Hofäcker 2003) and that international comparisons of attitudes should be treated with great caution (van Vlimmeren et al. 2017). Still, we assume that the attitudinal footprints of the common structures, and the similar natures of structural and institutional changes, justify considering that the Hungarian results may be relevant for, or generalizable to, some of the post-communist countries. We begin our analysis by giving an overview of theories about attitude change and highlighting the role of values and attitudes in the SDT theory and in its application for post-communist fertility transition. We also devote attention to some speci c contextual factors around the regime change that may have in uenced the directions of change in attitudes towards gender roles and family life. Then follows a short description of the data and methods. The statistical analysis comprises two steps: descriptive statistics and the decomposition of the change into population turnover and individual effects. The summary section will discuss the results from the perspective of general attitudinal change as well as from the aspect of the relevance of demographic approaches. 2 Approaches to understanding attitudinal change and the relevance of values for demographic analysis 2.1 General theories about attitude changes According to the relevant literature, three different avenues of thinking can be distinguished when considering change in attitudes towards gender roles and family arrangements (Brooks/Bolzendahl 2004; Ross/Sacker 2010). Structural approaches claim that the attitudes of individuals are strongly connected to the social structure; to institutional arrangements. More speci cally, gender-role and family expectations are linked to employment structures and labor market relations • Zsolt Spéder744 as well to the existing family systems. A certain social status can only manifest itself if individuals can adapt to it; different social statuses strengthen certain attitudes while simultaneously weakening others. This approach is based on macro- and micro-level research  ndings. Macro-level analyses link gender-role attitude changes to the growing female participation in the labor market, educational expansion and the increasing proportion of women in higher education (Cotter et al. 2011). Micro-level analyses reveal that individuals with higher quali cations, women and especially employed women and people living in atypical partnership and family forms have more permissive attitudes and expectations towards gender roles and partnerships than the average. Furthermore, interest-based approaches state that attitudes change because it is in the interest of the individual, whereas exposure-based explanations assume that attitudes change as a consequence of personal experiences and exposure to new circumstances (cf. Voicu/Tu s 2012). The second approach views attitudes towards gender roles and partnerships as an inherent element of a general value change, the mechanisms of which, following Brooks and Bolzendahl, can be described as ideological learning. This approach is supported by research  ndings that establish a close relationship between the liberalization of gender roles and the idea of equality and liberty (Thornton/Young- DeMarco 2001), individualization and civic values (Brooks/Bolzendahl 2004: 112), or secularization (Scott 2006). There are three aspects worth emphasizing. First, attitudes are interrelated and only change in conjunction with each other, but not separately. Second, new attitudes will only spread if they are considered “more favorable” than the related “old” attitudes. Finally, learning (the reproduction of attitudes) includes not only cognitive but also emotional elements, which helps people accept new attitudes. Speci cally, gender-role equity is a result but also an element of the increasingly supported, commonly shared human rights values. That is, belief in human rights helped spread the idea of equal gender roles (Brooks/ Bolzendahl 2004). Finally, according to approaches following the theory of cohort replacement delineated by Inglehart in The Silent Revolution (1997), values change with the appearance of younger generations and the disappearance of older generations. This concept is based on two pillars. On the one hand, individuals establish their fundamental beliefs and value system by socialization, especially during what are called the formative years. During that period, adolescents are extremely sensitive and open to what is new; their basic values and attitudes are formed and consolidated through constantly re ecting on their social environment. On the other hand, there is no fundamental change in general values, beliefs and attitudes; that is, there is neither remarkable aging nor period effects in the later life course. (According to Inglehart, the generation born and raised during the period of prosperity following the Second World War, that never experienced any  nancial deprivation, will embrace and later advocate post-material values.) The theory suggests that attitudes towards gender roles and partnership forms shift with the emergence of open-minded adolescents and the disappearance of older cohorts with a traditional mind-set. However, the “scarcity hypothesis“ is also present from the outset in this approach (Inglehart 1977: 22); namely that the actual circumstances, in particular A Quarter Century of Change in Family and Gender-Role Attitudes in Hungary • 745 scarcity, are re ected in preferences, which allows to interpret period effects within the framework (Inglehart 1985: 494 ff.). 2.2 Value shift as a core element of Second Demographic Transition (SDT) theory The SDT theory was developed to describe and understand the new forms of relationships (cohabitation, singleness) and changing childbearing practices (out- of-wedlock births, low fertility) that increasingly diffused in the last third of the 20th century in Western Europe (van de Kaa 1987, 2002; Lesthaeghe 1995). The key element of the theory is that the driving force behind these changes is the shifts of values. Speci cally, behavioral change is a consequence of the change in values that diffused and became actor-motivating in Western Europe beginning in the mid- 1960s. Lesthaeghe stresses that in the SDT, “the major role is given to the ideational factors, and the dynamics of cultural shift” (Lesthaeghe 2014: 1). The speci cities of value change were “borrowed” and are founded on established concepts in social science. First, according to Aries, motivations for having children changed; “individualistic” motivations replaced classical “altruistic” childbearing motivations in the 1960s; young people have children if it  ts in with their life goals (Aries 1980; van de Kaa 2001). At the same time, the quality of the relationship replaced the “centrality of kids” in family life. Secondly, in accordance with the works of Edward Shorter, new and free sexual behavior was spreading. Lesthaeghe 1995 generalized the weakening of norms, and stresses that the essence of the change in values is the liberation from all forms of authority (family, kinship, church, norms) and the creation of individual autonomy. Thirdly, drawing on Maslow’s hierarchy of needs, the spread of higher order needs such as self-actualization is highlighted (Lesthaeghe 1995). The spread of post-material values conceptualized and detailed by Inglehart (1977) is integrated into the SDT. Finally, although less strongly emphasized, the spread of the idea of gender equality is also noted, especially in later works (Lesthaeghe 2010). In the latest “concise” summary, Lesthaeghe sums up: “These new needs are centered on a triad: self-actualization in formulating goals, individual autonomy in choosing means, and a claim of recognition for their realization.” (Lesthaeghe 2014: 3) The SDT has been somewhat modi ed and re ned over the past decades. Structural conditions, the welfare state, and historical traditions have been “incorporated” in the explanation of new demographic behavior, although their role is often to explain the variation of behavior, but the theoretical role of value change has remained, namely “...cultural change ...as a necessary additional force with its own exogenous effects on demographic outcomes” (Lesthaeghe 2010: 242). The results are not clear-cut when it comes to specifying the exact nature of value change assumed by the theory. The proponents of the SDT do not make explicit whether they link value shift to period or cohort change. According to our understanding, value change is seen as general, pervasive and linked to a speci c • Zsolt Spéder746 historical period, starting in the mid-1960s.1 The fact that in their empirical analyses the proponents of the SDT mostly interpret the change in attitudes of the whole population (van de Kaa 1987: 5; 2001; Lesthaeghe 1995: 28-29, 34 ff.) and that they rarely examine attitudes by birth cohort (e.g. van de Kaa 2001) highlights their inclination to the period-related nature of value shifts. However, it is undeniable that in some places we  nd formulations referring to epochal, cohort and recursive (selection and adaptation) related value changes (Lesthaeghe 2010: 242). In interpreting the fertility decline in formerly communist countries, Lesthaeghe stressed the relevance of SDTs as opposed to structural explanations, highlighting that cultural shifts also played a role in the transformation of demographic behavior (Lesthaeghe 2010: 243). He assumes a historical effect by linking freedom of choice to political regime change (“Similarly, the rise of individual autonomy and freedom of choice has legitimized the adoption of nontraditional living arrangements in a very short time” (Lesthaeghe 2010: 243). Although Leshaeghe and Surkyn’s research focused on the relationship between value orientations and partnership forms, a chapter of their paper reported on attitude change over time, using the European Value Survey data from 1990 and 1999 (Leshaeghe/Surkyn 2002). It reports that perceptions of the family have changed in the direction of tolerance, while there is little change in many other areas, such as trust in institutions, socialization values and identi cation orientations. There is no explicit formulation in the analysis of whether the value changes may be cohort- or period-related. All in all, we conclude that the SDT interpretation of family formation and fertility changes in post-communist countries basically assumes a general period shift in values. There should be a separate paper to delineate and discuss criteria and aspects by which a theory is judged to be relevant, adequate or useful (cf. Lesthaeghe 1997; Graham 2021). That is beyond the scope of this paper. It is also out of scope to make a general evaluation of the SDT concept, as done by Zaidi and Morgan 2017. However, the present study is suitable to examine whether the value shifts assumed by the SDT theory could be the driver of fertility pattern change in Hungary, speci cally in the Hungarian post-communist fertility transition. According to our understanding, the relevance of the SDT theoretical model, also in the case of post-communist countries, can be corroborated if the following macro- and micro-level criteria are valid simultaneously: (1a) there is a substantial macro-level shift in the value patterns; (1b) the direction of the shifts is towards non-conformity, tolerance, post- material values; (2) in general, but especially in the case of the formerly communist countries, the dominance of period effect as opposed to the population turnover effect is expected; 1 Value change is described as an “epochal shift”, a shift from material to post-material values due to unprecedented economic growth and according to the Maslowian hierarchy of needs (van de Kaa 2001, 2002). A Quarter Century of Change in Family and Gender-Role Attitudes in Hungary • 747 (3a) value orientations and attitudes play an independent (net) role in the choice of relationship form and in fertility behavior (micro-level criteria); (3b) the direction of attitudinal effects must  t in with, and not contradict, the direction of macro-level (2b) changes. With our analysis, which aims at exploring and analyzing the trends and changing nature of attitudes towards gender and family roles over a quarter of century, we can make a substantive contribution only to the  rst three (1a, 1b and 2) criteria. Testing the last two criteria (3a and 3b) is not possible with the data used in the present analysis. Such an analysis would require a panel survey with a wide range of questions to measure attitudes (cf. Lesthaeghe/Moors 2002). 3 The post-communist context: research results, processes, assumptions 3.1 Review of relevant literature: ideational change in post-communist countries Findings on attitudinal and value changes regarding post-communist countries are non-conclusive. While several studies indicate a shift towards liberalization, there are others that reveal stable attitudes or a shift towards traditional gender-role expectations. Rabusic (2001) found that in the Czech Republic premarital sexual relationships and non-traditional partnership forms became more widely accepted after the change of regime. In Romania the acceptance of a modern, egalitarian family model and female employment increased; however, general ideas (ideologies) about gender roles remained unchanged between 1993 and 2004 (Voicu/Tu s 2012). A shift towards supporting atypical partnership forms is also found in a Hungarian qualitative empirical study, which focuses on the life goals of young generations before and after the political regime change (H. Sass 2003). Liberalization and increased tolerance were also apparent in Eastern Germany (Lee et al. 2007). A comparative study of the Baltic states and Western and Eastern Germany revealed an obvious liberalization in the case of Latvia (Motiejünaité/Höhne 2004). Finally, using two waves of the European Value Survey, Thornton and Philipov (2009) showed a liberal shift in some family related attitudes in some countries between 1990 and 1999. Shalom Schwartz in the mid 1990s posed the question as to whether profound political regime change may in uence basic value orientations (Schwartz et al. 2000). Using his well known value-index, they compared value orientation of enrolled college students between 1989 and 1996 in three post-communist countries (Poland, Hungary, Slovakia). The research concluded, somewhat unexpectedly, that the value system was almost “unaffected” by the change of regime (Schwartz et al. 2000). Using the two waves of the ISSP data collection in 1988 and 1994, Tóth hypothesized that perhaps a re-familization was underway (Tóth 1995). In an earlier analysis we stated that the assumption of overall liberalization does not apply to Hungary (Spéder/Kapitány 2014); however we were cautious, since from three • Zsolt Spéder748 waves of data collection it was too early to draw any conclusions about tendencies. A comparative study on the Baltic states found that family attitudes in Estonia and Lithuania shifted towards traditional family roles following the change of regime, and then reversed towards liberalization (Motiejünaité/Höhne 2004). All in all, research results about the tendencies of attitudes towards gender roles are mixed, and the share of cohort turnover and intra-individual (period-related) components in overall shifts are hardly considered. (Only Voicu and Tu s (2012) separated the changes into population turnover and period effects, and stress the dominant role of the period effect towards tolerance after 1993.) 3.2 Assumptions (hypotheses) based on key socio-political and structural changes Keeping in mind the outlined theoretical approaches on attitude change, we highlight some political and structural changes in post-communist countries that may have brought shifts in gender and family role expectations. Our aim is also to make assumptions on whether a shift in orientations may be a period effect, in uencing people from all walks of life, or may be operating by population turnover, or both mechanisms apply. From repression to freedom Both the transition from single-party totalitarianism to parliamentary democracy and the change from a redistributing economy to a market economy promised the arrival and an ever-increasing degree of freedom that would replace repression. Supposedly, freedom is not to be limited to politics or the economy, but would also affect other aspects of life, including gender relations and partnership forms. Consequently, and in accordance with the ideological learning approach, norms and attitudes about unconventional family relationships become more and more tolerated. This may in uence the general population; i.e., all kinds of people, which corresponds to Lesthaghe’s idea on the possible effects of regime change. However, a speci c realm of freedom, namely religious freedom, may not have encouraged further liberalization of family and gender-role expectations. It is widely known that the totalitarian regime supressed the civil sphere, which had dire consequences for religious institutions. Under communism, both secularization and “forced secularization” were apparent. The change of regime brought freedom for the churches and, despite numerous obstacles, the potential for a religious revival (Tomka/Zulehner 2000). Given the strong correlation between religion and beliefs about family (Scott 2006), an increased appreciation of traditional family values was expected as a result of a religious revival. This counter move, the acknowledgement of the conventional family, may impact a speci c part of the population: the people associated with the church. Although in different directions, both assumptions should have a period effect. The  rst would suggest a shift in attitudes towards liberalization after the regime change; the second, towards traditional beliefs (cf. Table 1). A Quarter Century of Change in Family and Gender-Role Attitudes in Hungary • 749 Employment relations: legacies and structural changes Gender relations in the labor market are critical for understanding gender roles, and are key aspects for structural approaches (c.f Brooks/Bolzendahl 2004). Even an overview of the legacies of rapid expansion, and the continued high level of female employment under communism2 and its relationship to gender-role expectations would, require a separate analysis (cf. Panayotova/Bray eld 1997). The consequences regarding attitudes towards female employment (and gender roles), very simplistically, appear to be ambivalent. On the one hand, since female participation was “enforced” by extensive industrialization, it was full-time and poorly remunerated (Berend 1996: 79 ff.). At the same time homemaking remained the responsibility of women, and attitudes toward employment and gender roles remained traditional (Panayotova/Bray eld 1997). On the other hand, it is undeniable that female employment provided a kind of autonomy and  nancial independence for women (Pongrácz/Molnár 1976). Thus a signi cant, albeit minority group of women in employment considered female employment as important and supported it. The regime change brought abrupt changes: a sharp contraction of the labor market within a very short time (the number of people employed decreased by about 20 percent between 1991 and 1992) led to intensi ed competition in that market. A gendered adjustment would have been favored by males since it would have helped them to maintain their main earner position, but also by a section of women who were exhausted from the “double shifts”. Consequently, according to the interest-based approach, increasing support for a move from the “dual earner- female carer” to the “male breadwinner” family model, with a shift towards more traditional gender-role attitudes, was not unexpected. However, the sectoral feature of the labor market contraction may have motivated stability or counter moves in gender-role expectations. Work opportunities shrank mainly in the primary and secondary sectors, but not in the service sector. Since the service sector had been dominated by female employment, participation in the labor market did not show a selection effect against women as a whole. The attitudinal consequences of this profound transformation of employment relations, if they are detectable, should be re ected in the period effects (cf. Table 1). Educational expansion Structural approaches emphasize the increasing level of education of the population as being a driver of attitudinal change. The expected relation is clear: the more highly educated, the greater the shift of attitudes towards gender equity and towards the tolerance of atypical living arrangements. Since educational expansion started in the late 1990s and gained momentum around the millennium: the participation in 2 For example, in 1970, when female employment in Western Europe was 46.4 percent, in Hungary and Poland it was 79.5 percent. At that time, the rate in Northern Europe (53.8 percent) was also far below that in the communist countries (Kornai 1992: 207). • Zsolt Spéder750 Ta b . 1 : S el ec te d p ro ce ss es a nd m ec h an is m s an d t he ir li ke ly e ff ec ts o n ch an gi ng a tt itu d es to w ar d s ge n d er a n d fa m ily ro le s Le ge nd : “+ ” de no te s an e xp ec te d e ffe ct o f t he m ec ha ni sm o n va lu e ch an ge . S ou rc e: a ut h or ’s o w n el ab or at io n M ec h an is m s in d iv id u al ( p er io d ) ch an g e p o pu la ti o n t u rn o ve r ch an g e su p p o rt in g s h if t to w ar d s m o d er n iz at io n / tr ad it io na l v al ue s m o d er n iz at io n / tr ad it io na l v al ue s lib er al iz at io n lib er al iz at io n V al u e ch an ge s, id eo lo g ic al le ar n in g In cr ea se d f re ed om a s a co n se q ue n ce o f t h e + p o lit ic al r eg im e ch an g e P o ss ib ili ty o f r el ig io us r ev iv al S tr uc tu ra l m e ch an is m s In cr ea se d c o m p e tit io n o f g en d e rs in t he + la b o r m ar ke t S ta b ili ty a n d e xt en si o n o ft h e (f em al e do m in at ed ) + se rv ic e se ct o r E d u ca tio n al e xp an si o n + C oh o rt r ep la ce m en t G en er al te n d en ci es + S o ci al iz at io n d u ri ng e co n o m ic r e ce ss io n + A Quarter Century of Change in Family and Gender-Role Attitudes in Hungary • 751 tertiary education more than tripled. The liberalizing effect of this change is expected to appear later and embedded in a cohort replacement mechanism. Cohort replacement A feature of social change often emphasized by demographers is that the dominant driver of change is the entry of new cohorts and the exit and death of older cohorts (Ryder 1964). It is generally assumed that the entering cohorts are more open and tolerant in their attitudes than older cohorts leaving society. Accordingly, it is assumed that cohort replacement brings with it continuous liberalization of attitudes. To summarize, the change of the political, institutional and economic system was such a fundamental historic transformation that it was highly justi ed to expect a general period shift in values and attitudes as well as a parallel shift in views about gender roles and family. However, as we highlighted, the assumed nature (direction, depth, inclusiveness) of the shift was not straightforward. (See the assumed effects of the different mechanisms in Table 1). The approach used in the next chapter makes it possible to study the nature of shifts in attitudes. It enables us to disentangle whether cohort replacement or the period effect played a more signi cant role in overall change. We will also be able to identify the direction of change, which will help to distinguish between more likely and less likely mechanisms shaping attitudes. However, this will not tell us whether and what kind of structural or cultural changes caused the shifts. 4 Data and methods The International Social Survey Program (ISSP) has included the “Family and changing gender roles” module in its program since 1988. The module has been repeated three times since then (in 1994, 2002 and 2013), with the intention of surveying changes in attitudes towards gender roles and family relations.3 The fact that Hungary is the only post-communist country where the module has been conducted during the communist time, clearly before the collapse of the communist system, and has been repeated three times thereafter, lends special signi cance to the analysis of the Hungarian data. Accordingly, using the ISSP module, it is only possible for Hungary to measure family-related attitudes before the ultimate collapse of the communist system and to investigate the potential effect of the political and economic regime change on changes in family related attitudes. Of course, several studies have analyzed certain waves of the Hungarian data;4 however, to the best of 3 The last (4th) “Family and changing gender role” module was due in 2012, but the Hungarian  eldwork was done in the  rst half of 2013; therefore we indicated the year differently. The sample sizes (N=) of the Hungarian surveys were 1737, 1494, 1023 and 1012 respectively. For more information on ISSP, see https://issp.org/. 4 For example, Tóth 1995, Dupcsik/Tóth 2008, Spéder 2003, Blaskó 2006, Rohr/Spéder 2014. • Zsolt Spéder752 our knowledge, we are the  rst to analyze and interpret changing ideas about family and gender roles within a twenty- ve-year timeframe, and who aim to understand the effects of regime change using decomposition. The “Family and changing gender roles” module of the ISSP covers a wide range of family and gender-role–related attitudes. Altogether the 11 items below were included in all four waves of the survey (Table 2). As a  rst step, general trends are presented using arithmetic means of the above scales, and then changes in arithmetic means will be decomposed. Firebaugh developed an arithmetical method for analyzing repeated cross-sectional surveys, which makes it possible to determine whether changes over time depend on population replacement or on changes in individual attitudes (Firebraugh 1992, 1997). More precisely, average change in opinions is a result of cohort replacement, individual, and joint effects; however, the joint effect can be distributed between population replacement and individual components. This approach enables us to calculate the extent to which the change in averages is the consequence of cohort replacement – that is, the fact that emerging younger cohorts have different attitudes than the disappearing older cohorts – or whether Tab. 2: Statements (items) measuring attitudes towards gender roles and partnership forms that were included in all four waves (1988, 1998, 2002, 2013) of the ISSP “Family and changing gender roles” module 1 A working mother can establish just as warm and secure a relationship with her children as a mother who does not work. 2 A pre-school child is likely to suffer if his or her mother works. (x) 3 All in all, family life suffers when the mother has a full-time job (x) 4 A job is all right, but what women really want is a home and children. (x) 5 Being a housewife is just as ful lling as working for pay. (x) 6 Both the man and the woman should contribute to the household income. 7 A man’s job is to earn money; a woman’s job is to look after the home and the family. (x) 8 Married people are generally happier than unmarried people. (x) 9 People who want children ought to get married. (x) 10 Watching children grow up is life’s greatest joy. (x) 11 People who have never had children lead empty lives. (x) Notes: The degree of agreement was measured on a  ve-point scale that ranged from strongly agree (5) through both agree and disagree (3) to strongly disagree (1). When interpreting responses we took lower values as re ecting support for traditional family and partnership forms, while a higher value stood for a modern attitude, supporting gender equality and the plurality of partnership forms. (Responses were recoded in the case of items marked with an x). Source: ISSP 1988, 1994, 2002, 2013 A Quarter Century of Change in Family and Gender-Role Attitudes in Hungary • 753 the opinions of individuals change over time.5 Firebaugh refers to the latter as “individual change” because it measures to what extent given individuals, as representing individuals of a society, change their opinion. Since attitude changes occur over a given period, interpretations consider changes as closely relating to the social changes during the period; hence they can be considered period effects. Notwithstanding, an ageing or life-cycle effect may not be fully eliminated from individual cases, but according to Firebaugh (Firebaugh 1992: 13) they are not reinforcing the individual and turnover effects.6 This technique has already been used successfully to assess changes in gender-role attitudes (Firebaugh 1992; Brewster/Padavic 2000; Voicu/Tu s 2012; Kraaykamp 2012). 5 Results 5.1 General tendencies The three graphs below (Fig. 1, 2 and 3) give an overview of twenty- ve-year-long trends by presenting the averages for the responses to each statement. (Averages can be found in Table A1 in the appendices.) As mentioned above, the acceptance or rejection of certain family roles are not presented based on responses to the original items, but they have been recoded. When recoding responses we took lower values as representing approval of the traditional family model; that is, a set of opinions that re ect support for the traditional division of gender roles: the preference for women in family roles, marriage as the desired partnership form and having children as an important life goal. High values indicate support for equal gender-role expectations or liberal family model, which includes the dual-earner partnership set-up, a favorable relation to female employment, and acceptance of cohabitation.7 Looking at changes in average values over time, it is apparent, especially when comparing two consecutive survey rounds (years), that most of the changes went in the same direction (c.f. Fig. 1-3). To make it easier to track changes, we have also created an indicator in the tradition of Inglehart in which we ignore those who could not choose between agreement and disagreement. In Table A2, we subtracted the 5 The basic premises of the method are that representative samples represent a given society well and that a repeated survey accurately reproduces changes in the characteristics and attitudes of a given society. 6 Firebaugh (1992) also mentions that the study of changes in gender-role attitudes using panel data gave the same results. We did some robustness checks using linear regression models of attitude change on pooled data, and there were no ageing effects, but period effects remained. 7 In certain cases it is not obvious whether a response to a given statement indicates traditional or modern family ideals. We took having children as an important life goal as belonging to the traditional family model; however, this could undeniably be part of a modern family ideal as well. Therefore, we must note that tendencies in changes – whether family-related attitudes have become more liberal – are more important in the present analysis. In this respect items can be categorized fairly well. • Zsolt Spéder754 proportion of those who hold conventional attitudes from the proportion who hold liberal attitudes, thus obtaining an indicator that is positive and increasing if liberal attitudes are in the majority or increasing, and negative and decreasing if support for traditional attitudes is in the majority or increasing. Caution should be exercised in interpreting the level of this indicator, as the meaning of the answers may depend on the positive or negative formulation of a given item. In any case, this indicator gives a more clear insight on the direction of the change within a given attitude item. General shifts can be summarized as follows: 1) Family attitudes for the period between 1988 and 1994, during the high time of societal transition and regime change, shifted towards increased support for the traditional family ideal. For 7 items of the 11, we can see an obvious decrease in average values, in two cases there is an increase, while there is no sizable change in the remaining two cases. Fig. 1: Changes in attitudes towards gender roles between 1988 and 2013 in Hungary Note: Responses to certain items have been recoded so that lower values represent approval of the traditional family model and higher values indicate support for a modern or liberal family model. Source: own calculation, ISSP 1988, 1994, 2002, 2013 A Quarter Century of Change in Family and Gender-Role Attitudes in Hungary • 755 2) During the period from 1994 to 2002 we observe a change in the opposite direction compared to the previous period; a certain reversal and rearrangement of attitudes. Average values increased in all cases; that is, opinions re ected a shift towards a more modern family ideal. 3) During the decade between 2002 and 2013 the support for a more modern family life and equity in gender relations continued to increase, although three attitudes/items showed a decrease (e.g. “A working mother can establish just as warm and secure a relationship with her children as a mother who does not work.”). In some cases stagnation seems more likely; however, in the majority of cases (7 items), values indicate a shift towards modernization. There are two items where shifts differ from the others. First, the proportion of those who disagree with the statement “Being a housewife is just as ful lling as Fig. 2: Changes in attitudes towards female employment between 1988 and 2013 in Hungary Note: Responses to certain items have been recoded so that lower values represent approval of the traditional family model and higher values indicate support for a modern or liberal family model. Source: own calculation, ISSP 1988, 1994, 2002, 2013 • Zsolt Spéder756 working for pay” shows an increase at each interval; that is, the rejection of women being restricted to the role of a housewife has increased continuously. Secondly, there is general and lasting support for the necessity of the “double earner” family; more precisely, there is increasing support for the idea that “Both the man and the woman should contribute to the household income”. The general acceptance of the last statement may re ect the constraints of those living in needy material circumstances as much as the gender equality expectation regarding female employment. Based on the descriptive statistics we may conclude that although there was a shift towards more traditional family ideals after the change of regime, support for gender equity and modern family relations increased later, and perhaps on the whole. However, a more nuanced conclusion can be drawn if we disentangle Fig. 3: Changes in attitudes towards family and marriage between 1988 and 2013 in Hungary Note: Responses to certain items have been recoded so that lower values represent approval of the traditional family model and higher values indicate support for a modern or liberal family model. Source: own calculation, ISSP 1988, 1994, 2002, 2013 A Quarter Century of Change in Family and Gender-Role Attitudes in Hungary • 757 population turnover and individual change using Glenn Firebaugh’s arithmetic decomposition method. 5.2 Decomposing change into population turnover and individual (period-related) effects Firebaugh’s approach, as mentioned earlier, reveals the extent to which the change in averages is the consequence of cohort replacement – that is, the fact that appearing younger cohorts have different attitudes to the disappearing older ones – or whether it results from the changing opinion of individuals (Firebaugh 1992, 1997). Table 3 shows, for all the 11 attitude items and for all the three periods how far changes in average values are the result of population turnover and to what extent they can be attributed to individual change across periods. While the  rst three columns show the direction and the amount of change as a consequence of cohort replacement in average values, the next three columns show the direction and degree of changes in individual opinion in the given periods. A minus sign (-) signi es a shift towards traditional ideas ascribed to the given effect; an unmarked value, a shift toward liberal attitudes. The values of the two effects are comparable: if the absolute value ascribed to the population turnover is lower than that of individual change, then it had a weaker effect on the average change. For example: The average for the responses to the statement “A pre-school child is likely to suffer if his or her mother works” decreased by -0.28 during the period between 1988 and 1994 (see Table A1, second row). As for the directions in Table 3, a negative intra- individual effect shows a shift towards traditional attitudes, whereas population replacement shifted towards liberal attitudes. As far as the size of the effects is concerned, values in Table 3 show that the individual (period) effect was  ve times larger (-0.35) than the population turnover effect (0.07). Together, the individual effect masked the population turnover effect. During the next period (1994-2002) the average increased by 0.29, of which 0.10 is related to cohort replacement and 0.19 to the individual/period component. Both components shifted the average towards a more liberal direction, and once more, if at a lower size, individual effect was stronger. Finally, in the last period, cohort replacement and period effects contributed to the overall 0.38 increase by 0.10 and 0.28, respectively. In Table 3 we have highlighted all the numbers of the two components that contributed statistically signi cantly to the change in arithmetic means. Positive values in italics and bold indicate a shift towards liberalization due to the given component, while negative values in bold indicate a shift towards supporting traditional family ideals. Lastly, not highlighted values indicate that the effect was statistically not signi cant. It is encouraging that the majority of cells are highlighted and there is a clear pattern. Changes in attitudes over time or just after the regime change are central in our investigation. Taking into account the decomposed effects, we  nd that the individual change in most attitudes in the 1988-1994 period (Table 3, column 4), was larger than the difference in arithmetic means between 1988 and 1994 (c.f. Table • Zsolt Spéder758 A1). This is due to the constant positive population turnover effect in this period (Table 3, column 1). It means that the population turnover effect weakened the overall conservative turn associated with the period effects. Speci cally, for seven items we see a very strong, statistically signi cant conservative turn regarding individual effects (Table 3, column 4). For two attitudes, the period effect points in the direction of liberalization. However, the population turnover effects in each item point towards liberalization, but their magnitude is far below the value of the Tab. 3: The size and direction of population replacement and individual (period) effects in changes in family attitude averages, Hungary (time periods 1988-94, 1994-2112, 2002-12) Population replacement Individual/period effect effect 1988- 1994- 2002- 1988- 1994- 2002- 1994 2002 2013 1994 2002 2013 A working mother can establish just as warm and secure a relationship with her children as a mother who does not work. 0.06 0.05 -0.03 0.15 0.32 -0.13 A pre-school child is likely to suffer if his or her mother works. 0.07 0.10 0.10 -0.35 0.19 0.28 All in all, family life suffers when the mother has a full-time job. 0.08 0.10 0.12 -0.29 0.22 0.06 A job is all right, but what women really want is a home and children. 0.02 0.08 0.10 -0.19 0.23 0.06 Being a housewife is just as ful lling as working for pay. 0.06 0.06 0.11 0.55 0.07 0.13 Both the man and the woman should contribute to the household income. -0.01 0.01 0.01 0.06 0.07 -0.19 A man’s job is to earn money; a woman’s job is to look after the home and the family. 0.03 0.14 0.06 -0.59 0.35 -0.23 Married people are generally happier than unmarried people. 0.07 0.09 0.19 -0.30 0.08 0.22 People who want children ought to get married. 0.12 0.21 0.22 -0.02 0.21 -0.25 Watching children grow up is life’s greatest joy. 0.01 0.04 0.08 -0.38 0.04 0.10 People who have never had children lead empty lives. 0.02 0.08 0.08 -0.23 0.16 0.12 Notes: italics and bold: signi cant change towards liberalization; bold: signi cant change towards traditional/conventional attitudes; not marked: change not signi cant statistically (p>0.05). Source: own calculations, ISSP 1988, 1994, 2002, 2013 A Quarter Century of Change in Family and Gender-Role Attitudes in Hungary • 759 individual effects. On the whole, a clear “conservative turn” took place between 1988 and 1994, which was partly watered down by cohort replacement. The following period, from 1994 to 2002, can be basically considered a time of “reversal”. The pendulum swings back; for 9 items, there is a clear period effect towards liberalization. However, the reversal is not complete, since when only the period effects are compared, in 6 cases the values for the period 1994-2002 are smaller (Table 3, column 5) than the values for the previous period (Table 3, column 4). At the same time, for 4 items the degree of liberalization is larger than the rate of change in the previous period. On the whole, however, we can speak of a regression of the pendulum. When the effect of population turnover is also taken into account, the attitudes in 2002 were very close to those in 1988: the average value for 6 items was higher than in 1988, for 2 items the same and for 3 items lower than in 1988 (cf. Table A1). During the decade from 2002 to 2013 the period effect is mixed: although liberalization seems to be dominant, in the case of two items there is a conservative shift, while in three cases there is no signi cant change. At the same time, the two items where we see a renewed “rebound” seem to be of crucial importance. There was a clear increase (0.23) in support for “A man’s job is to earn money; a woman’s job is to look after the home and the family”, which could not be compensated by the 0.06 for population turnover. On the other hand, the support for the two-earner family model decreased: period effect -0.19, turnover effect 0.01. It is worth re ecting on three items in particular. There is no “conservative turn” during 1988 and 1994 or even later in relation to the statement “Being a housewife is just as ful lling as working for pay.” Rejection of the “exclusive housewife role” is not only increasingly high among new generations, but the opinion of older individuals also shows an increasing degree of disagreement. Finally, the change in responses regarding two items related to family models, to the “double earner” and “male breadwinner” gender roles, seems to be contradictory. On the one hand there is high support, and relatively also the highest support, for the statement “Both the man and the woman should contribute to the household income.” On the other hand, responses to the item “A man’s job is to earn money; a woman’s job is to look after the home and the family” are volatile over time. (Note that there is no  uctuation in population turnover effect: younger cohorts show consistently decreasing support for the “male breadwinner” model.) Considering the tendencies of population turnover and individual effect, various tendencies could be identi ed. As far as the individual (period) effects, the majority of effects are signi cant (27 of the 33 cells are highlighted), and changes in the trends are both easily visible and vast (there are dominant shifts in each column). It can be concluded that after the change of regime a clear conservative shift took place in attitudes towards gender roles, which was followed by a slow but dominant reversal that lasted until the  nal period investigated. These developments indicate an increasing level of support towards liberal gender roles and alternative family • Zsolt Spéder760 forms following 1994; however we cannot exclude presence of shifts towards traditional attitudes in the time after the turn of the millennium.8 On the other hand, shifts attributable to cohort replacement are uniform and broad. All 33 cells of the  rst three columns are highlighted, and in 31 cases we found that cohort replacement led to an increase in averages: that is, the support for non-conventional family ideals and for more equity in gender roles. This coincides with theories that consider generational changes as having a signi cant impact on social changes (Ryder 1964). This means that new generations growing up acquire and internalize more liberal ideas during their “formative years”, which functions as the “driving force” behind social changes (Inglehart 1977). 6 Discussion Our analysis was intended to answer two questions. First we examined whether the profound political, institutional and structural changes since 1989/90 have been accompanied by changes in attitudes, speci cally in the area of gender-role expectations and family-related beliefs, and if so, whether historical events or cohort replacement – the latter always present in population development – had a decisive role in determining the nature of these changes. Secondly, our analysis enabled us to examine a key element of SDT in Hungary; namely, to what extent attitudinal shifts are consistent with the assumptions of SDT theory. We concluded that gender-role expectations were changing continuously over the quarter of a century examined, but the direction of change, contrary to expectations, has not been unidirectional. The changes most closely resemble a pendulum movement. During the period of regime change, a conservative turn could be clearly discerned. However, this turn has proved to be temporary, insofar as at the turn of the millennium (2002), almost a decade and a half after 1988, the general characteristics (mean) of attitudes were very similar to the characteristics of the time before the regime change (1988). Thereafter, in the  rst decade of the new millennium, the modernization shift, the liberalization of attitudes, continued, but it was not comprehensive, and in some cases there was a re-return to traditional ideas. Our results regarding the conservative turn due to the regime change stand in contrast with the results of several other analyses (Voicu/Tu s 2012; Thornton/ Philipov 2009). These analyses identi ed liberalization of gender-role attitudes and growing tolerance towards non-conformist partnership forms in the last decade of the last century in the post-communist countries. However, the  rst measurement 8 Undoubtably, attitude change may be in uenced by other factors as well, which may be included in the pure cohort and period effects. Of course educational expansion strengthens the uncontrolled cohort effect, yet we have to note that the effects of birth cohort remain signi cant when we analyse changes in attitudes using linear regression on the pooled data of consecutive survey waves. In that exercise, we included gender, educational attainment, family status and economic activity in the models. A Quarter Century of Change in Family and Gender-Role Attitudes in Hungary • 761 date of these analyses was invariably after the regime change. To the best of our knowledge, the ISSP is the only data collection where measurements of gender- role attitudes in a post-communist country (Hungary) are available for both pre- and post-regime-change dates. In light of this, we consider our  ndings valid. Our results regarding changes measured in the 1994-2002 period are otherwise in line with the above results. In order to study the nature of the change, the trends of attitudinal averages were decomposed into two components: intra-individual (period) and population turnover (cohort) components, based on Firebaugh’s method. Comparing the magnitude of the two components for each attitudinal item in the respective periods, it became clear that the individual changes were in most cases clearly larger than the population turnover values, i.e. changes related to the historical events dominated the overall evolution of attitudes of the whole population. The change in individuals’ views, what we see as essentially period effect, has, however, clearly  uctuated over the twenty- ve years. There were both signi cant increases and decreases in the support for traditional gender roles and family model in 9 out of 11 items across the entire period.9 Population turnover, the entry and exit of new and old cohorts respectively, brought about shifts towards liberalization, increasing support for the equality of gender roles, and increasing tolerance towards different partnership forms. Due to the strength of the period effect, it was able to bounce along the averages over time. The population turnover effect was smaller but unidirectional, amplifying the period effects in some periods and dampening them in others. In order to examine the empirical evidence for SDT theory, we identi ed 5 criteria that we consider suf cient and at the same time necessary to see SDT theory as relevant for understanding the fertility transition in post-communist countries (cf. section 1.2). Our analysis enables re ection regarding three criteria. First, it has been shown that there were substantial changes in gender-role expectations as a result of, or in parallel with, historical events. Secondly, the direction of change in the  rst period of regime change is contrary to what the SDT theory assumes. Although the SDT assumes growing support for gender equity, attitudinal shifts between 1988 and 1994 clearly pointed towards more traditional role expectations. Thirdly, the period effects are larger, but at the same time, close to the regime change, they are opposite in direction to what was expected according to the SDT. It is a well-known fact, discussed in detail in the literature, that after regime change Hungary and other post-communist countries were characterized by a rapid transformation of family formation, a rapid spread of cohabitation and a decline in fertility (Frejka 2008; Sobotka 2011; Sobotka/Toulemon 2008). (For an illustration, the relevant Hungarian data can be found in Table A3.) Whether and how far the above behavioral and attitudinal changes at the macro level are related cannot be 9 This result is in accordance with our assumption that individual changes in opinion are mainly the result of period effects. It is inconceivable that the effect of ageing varies from period to period. • Zsolt Spéder762 analyzed using the ISSP data collection. But it may be possible to assess whether the measured attitudinal and behavioral changes point in the same direction. In our view, the spread of singleness, cohabitation and out-of-wedlock births is dif cult to reconcile with a macro-level conservative turn. On this basis, it is hard to imagine that changes in values motivated the behavioral changes in the period immediately after the regime change. Therefore, according to the results of the analysis of gender and family role expectation trends, we cannot corroborate that value changes initiated or enabled by the regime change triggered the behavioral change. We have to concur with those skeptical views about the relevance of the SDT in the case of all post-communist countries (cf. Rychterikova 2000; Rotariu 2006; Perelli-Harris 2008; Perelli-Harris/Gerber 2011). Two objections may challenge our assessment of the relevance of SDT in countries, including Hungary, where attitudes do not seem to be generally liberalized. First, it cannot be ruled out that liberal value orientations are playing an increasingly in uential role in behavioral shifts (in the rise of singleness, cohabitation, out-of- wedlock births) at the micro level, while society is undergoing a conservative shift at the macro level. Yet the opposite assumption seems more plausible to us: We cannot attribute a decisive role to the attitudinal change (conservative turn) following the regime change in the spread of new demographic behavior. This is not to say that we neglect the role of values in family and fertility decision making, nor their in uence on the long-term development of fertility and partnership relations. It is not impossible that the structural conditions of the communist system (high female employment, intense internal migration, political atomization of society etc.; cf. Spéder/Kamarás 2008) heavily contributed to the erosion of certain norms. In this case, however, attitudinal and behavioral changes are perhaps more a consequence of structural change. Secondly, it may be argued that we have misinterpreted SDT because it sees generational replacement as the vehicle for value change. Our results are consistent with this, as the effects of generational turnover is present in all periods and clearly points towards a liberalization of attitudes. We do not deny this role, but the higher pace with which behavior spreads suggests other mechanisms (also) at work. Furthermore, in this scenario it would be necessary to identify the speci c generations (cohorts) that triggered the changes. There are several limitations to our analysis. The ISSP data do not allow us to answer the question of the reasons for the conservative turn, nor for the reversal of opinion. We can perhaps be assured that the assumptions about liberalization (section 2.2), i.e. the liberalizing role of the generalization of freedom and its spill- over into gender-role expectations, must be rejected. Freedom and democracy have not “brought with them” further rapid liberalization of attitudes and ideas about family relations. We consider it likely that the pre-transition mode of gender division of labor (“double earner, female carer”) and the confrontation with the transformation of the labor market (contraction and competition) may have been decisive in the conservative turn. Full-time housework alongside full employment, the “double burden”, may have made more traditional role expectations attractive to many women. For men, a conservative turn may have been a response to the A Quarter Century of Change in Family and Gender-Role Attitudes in Hungary • 763 challenge of unprecedented intense competition in the labor market. However, these are all speculations; the ISSP data cannot be used to detect the reasons for the shifts, so we cannot test our hypotheses in a conventional manner. It is legitimate to ask to what extent the Hungarian results are generelisable to the former socialist countries. The wider relevance is supported by the studies that drew attention to the processes of familization following the regime change (Lobodzinska 1996; Motiejünaité/Höhne 2012). However, there are signi cant differences in the prevalence of post-materialist values among post-communist countries, although mainly in recent times (Chromkova Manea/Rabusic 2020). We should therefore be cautious about straightforward generalization. The question is also often raised as to how well an item set on gender-role expectations developed in the late 1980s can reliably measure role expectations a quarter of a century later, or whether a given item has the same meaning in all countries and cultures (van Vlimmeren et al. 2017). The standard answer to this question is that to measure change or invariance, it is essential to use the same measurement instrument. In the introduction, we raised the point that in analyzing fertility trends, we often refer to value change, to individualization, whether in scienti c studies, academic analyses or public life. All this requires a more systematic tracing of the nature of and changes in people’s attitudes, values and norms (cf. Liefbroer/Billari 2010). With our study we hope to have made a modest contribution to this task. References Alwin, Duane F.; McCamon, Ryan J. 2003: Generations, Cohorts, and Social Change. In: Mortimer, Jeylan T.; Shanahan, Michael J. (Eds.): Handbook of the Life Course. Boston: Springer: 23-50. https://doi.org/10.1007/978-0-306-48247-2_2 Alwin, Duane F.; Braun, Michael; Scott, Jacqueline 1992: The Separarion of Work and the Family: Attitudes Towards Women’s Labour-Force Participation in Germany, Great Britain, and the United States. In: European Sociological Review 8,1: 13-37. Aries, Philippe 1980: Two Successive Motivations for the Declining Birth Rate in the West. In: Population and Development Review 6,4: 645-650. https://doi.org/10.2307/1972930 Berend, Ivan T. 1996: Central and Eastern Europe 1944-1993. Detour from the Periphery to the Periphery. Cambridge Univ. Press. Blaskó, Zsuzsa 2006: Nők és fér ak – keresőmunka, házimunka. In: Kutatási jelentések 82. Braun, Michael; Scott, Jacqueline 2009: Changing public views of gender roles in seven nations, 1988-2002. In: Haller, Max; Jowell, Roger; Smith, Tom W. (Eds.): The International Social Survey Programme, 1984-2009. Charting the globe. Routledge: London: 358-377. Brewster, Karin L.; Padavic, Irene 2000: Change in Gender-Ideology, 1977-1996: The Contributions of Intracohort Change and Population Turnover. In: Journal of Marriage and Family 62,2: 477-487. https://doi.org/10.1111/j.1741-3737.2000.00477.x Brooks, Clem; Bolzendahl, Catherine 2004: The transformation of US gender role attitudes: cohort replacement, sociol-structural change, and ideological learning. In: Social Science Research 33,1: 106-133. https://doi.org/10.1016/S0049-089X(03)00041-3 • Zsolt Spéder764 Chromková Manea, Beatrice; Rabušic, Ladislav 2020: Value Modernisation in Central and Eastern Europeean Countries: How Does Inglehart’s Theory Work? In: Czceh Sociological Review 56,6: 699-740. https://doi.org/10.13060/csr.2020.033 Cotter, David; Hermsen, Joan M.; Vanneman, Reeve 2011: The End of the Gender Revolution? Gender Role Attitudes from 1977 to 2008. In: Amarican Journal of Sociology 117,1: 259-289. https://doi.org/10.1086/658853 Dahrendorf, Ralf 1990: Re ections on the Revolution in Europe. London: Chatto and Windus. Dupcsik, Csaba; Tóth, Olga 2008: Feminizmus helyett familizmus. In: Demográ a 51,4: 307-328. Firebaugh, Glenn 1992: Where does social change come from? In: Population Research and Policy Review 11,1: 1-20. https://doi.org/10.1007/BF00136392 Firebaugh, Glenn 1997: Analyzing repeated surveys. Sage University Paper: Quantitative Applications in the Social Science. Frejka, Tomas 2008: Determinants of family formation and childbearing during the societal transition in Central and Eastern Europe. In: Demographic Research 19,7: 139- 170. https://doi.org/10.4054/DemRes.2008.19.7 Graham, Elspeth 2021: Theory and explanation in demography: The case of low fertility in Europe. In: Population Studies 75,1: 133-155. https://doi.org/10.1080/00324728.2021.1971742 H. Sass, Judit 2003: Egy napom tíz év múlva. In: Educatio 14,2: 365-383. Harrison, Eric; Fitzgerald, Rory 2010: A chorus of disapproval? European attitudes to non-traditional family patterns. In: Park, Alison et al. (Eds.): British Social Attitudes. The 26th Report. Sage: 135-158. Hoem, Jan M. 2008: Preface: Childbearing Trends and Policies in Europe. In: Demographic Research 19,1: 1-4. https://doi.org/10.4054/DemRes.2008.19.1 Inglehart, Ronald 1977: The Silent Revolution. Changing Values and Political Styles among Western Publics. Princeton University Press Inglehart, Ronald 1985: New perspectives on Value change: Response to Lafferty and Knudsen, Savage and Böltken and Jagodzinski. In: Comparative Political Studies 17,4: 485-532. https://doi.org/10.1177/0010414085017004004 Inglehart, Ronald; Baker, Wayne E. 2000: Modernization, cultural change, and the persistence of traditional values. In: American Sociological Review 65,1: 19-51. https://doi.org/10.1177/000312240006500103 Kornai, János 1992: The Socialist System: The Political Economy of Communism. Princeton: Princeton University Press. Kraaykamp, Gerbert 2012: Employment status and family role attitudes: A trend analysis for the Netherlands. In: International Sociology 27,3: 308-329. https://doi.org/10.1177/0268580911423046 Lee, Kristen S.; Alwin, Duane F.; Tu ș, Paula A. 2007: Beliefs about Women’s Labour in the Reuni ed Germany, 1991-2004. In: European Sociological Review 23,4: 487-503. https://doi.org/10.1093/esr/jcm015 Lesthaeghe, Ron 1995: The Second Demographic Transition in Western Countries: An Interpretation. In: Oppenheim Masson, Karen; Jensen, An-Magritt (Eds.): Gender and Family Change in Industrialized Countries. Oxford University Press: 17-62. A Quarter Century of Change in Family and Gender-Role Attitudes in Hungary • 765 Lesthaeghe, Ron 1997: Imre Lakatos’ views on theory development: applications to the  eld of fertility theories. Interuniversity papers in demography. IPD-Working Papers 1997-1. Vrie Universiteit Brussel. Lesthaeghe, Ron 2010: The unfolding story of the second demographic transition. In: Population and Development Review 36,2: 211-251. https://doi.org/10.1111/j.1728-4457.2010.00328.x Lesthaeghe, Ron 2014: The second demographic transition: A concise overview of its development. In: PNAS 111,51: 18112-18115. https://doi.org/10.1073/pnas.1420441111 Lesthaeghe, Ron; Surkyn, Johan 2002: New Forms of Household Formation in Central and Eastern Europe: Are they related to newly emerging Value Orientations? In: UNECE (Ed.): Economic survey of Europe. Geneva: UNECE: 197-216. Lesthaeghe, Ron; Moors, Guy B. D. 2002: Life course transitions and value orientations: Selection and adaptation. In: Lesthaeghe, Ron (Ed.): Meaning and choice – Value orientations and life course decisions. Brussel: CBGS: 1-44. Liefbroer, Aart C.; Billari, Francesco C. 2010: Bringing norms back in: A theoretical and empirical discussion of their importance for understanding demographic behaviour. In: Population, Space and Place 16,4: 287-305. https://doi.org/10.1002/psp.552 Lobodzinska, Barbara 1996: Women’s Employment or Return to “Family Values” in Central-Eastern Europe. In: Journal of Comparative Family Studies 27,3: 519-544. https://doi.org/10.3138/jcfs.27.3.519 Lück, Detlev; Hofäcker, Dirk 2003: Rejection and Acceptance of the Male Breadwinner Model: Which Preferences Do Women Have under Which Circumstances? GLOBALIFE Working Paper 60. Bamberg: Universität Bamberg. Motiejūnaitė, Akvilė; Höhne, Anke 2012: Gender-role attitudes and female employment in East/West Germany and the Baltic countries. In: Motiejūnaitė, Akvilė (Ed.): Female employment, gender roles, and attitudes. The Baltic countries in a broader context. Stockholm University: 57-80. Offe, Claus 1994: Der Tunnel am Ende des Lichtes. Erkundungen der politischen Transformation im Neuen Osten. Campus Verlag. Panayotova, Evelina; Bray eld, April 1997: National Context and Gender Ideology Attitudes toward Woman’s Employment in Hungary and the United States. In: Gender and Society 11,5: 627-655. Perelli-Harris, Brienna 2008: Ukraine: On the border between old and new uncertain times, and below replacement. In: Demographic Research 19,29: 1145-1178. https://doi.org/10.4054/DemRes.2008.19.29 Perelli-Harris, Brienna; Gerber, Theodore P. 2011: Nonmarital Childbearing in Russia: Second Demographic Transition or Pattern of Disadvantage? In: Demography 48,1: 317-342. https://doi.org/10.1007/s13524-010-0001-4 Philipov, Dimiter 2003: Fertility in times of discontinuous societal change. In: Kotowska, Irena; Jóźwiak, Janina (Eds.): Population of Central and Eastern Europe. Challenges and Opportunities. Warsaw: Statistical Publishing Establishment: 665-689. Philipov, Dimiter; Spéder, Zsolt; Billari, Francesco C. 2006: Soon, later, or ever? The impact of anomie and social capital on fertility intentions in Bulgaria (2002) and Hungary (2001). In: Population Studies 60,3: 289-308. https://doi.org/10.1080/00324720600896080 Pongrácz, Tiborné; S. Molnár, Edit 1976: Népesedési kérdésekkel kapcsolatos közvéleménykutatás. KSH Népességtudományi Kutató Intézetének és az MTA Demográ ai Bizottságának közleményei 43. • Zsolt Spéder766 Rabušic, Ladislav 2001: Value Change and Demographic Behaviour in the Czech Republic. In: Czech Sociological Review 9,1: 99-122. Rohr, Adél; Spéder, Zsolt 2014: Stalling modernization of family values: a quarter century of change and stability. Poster presented at the European Population Conference. Budapets, 25-28 June. Ross, Andy; Sacker, Amanda 2010: Understanding the dynamics of attitude change. In: Park, Alison et al. (Eds.): British Social Attitudes. The 26th Report. Sage: 115-133. Rotariu, Traian 2006: Romania and the Second Demographic Transition. The Traditional Value System and Low Fertilty Rates. In: International Journal of Sociology 36,1: 10- 27. https://doi.org/10.2753/IJS0020-7659360102 Rychtarikova, Jitka 2000: Demographic transition or demographic shock in recent population development in the Czech Republic? In: Acta Universitas Caroliane Geographica 1: 89-102. Ryder, Norman B. 1964: The Process of Demographic Translation. In: Demography 1,1: 74-82. https://doi.org/10.1007/BF03208446. Schwartz, Shalom H.; Bardi, Anat; Bianchi, Gabriel 2000: Value Adaptation to the Imposition and Collapse of Communist Regimes in East-Central Europe. In: Duckitt, John; Renshon, Stanley A. (Eds.): Political Psychology: Cultural and Cross Cultural Foundations. London: Palgrave Macmillan: 217-237. https://doi.org/10.1057/9780230598744_13 Scott, Jacqueline 2006: Family and Gender Roles: How Attitudes are Changing. In: ARXIUS 15: 143-154. Sobotka, Tomás 2011: Fertility in Central and Eastern Europe after 1989: Collapse and Gradual Recovery. In: Historical Social Research 36,2: 246-296. https://doi.org/10.12759/hsr.36.2011.2.246-296 Sobotka, Tomás; Toulemon, Laurent 2008: Changing family and partnership behaviour: Common trends and persistent diversity across Europe. In: Demographic Research 19,6: 85-138. https://doi.org/10.4054/DemRes.2008.19.6 Spéder, Zsolt 2003: Fertility behaviour in a period of economic pressures and growing opportunities – Hungary, the 1990s. In: Kotowska, Irena E.; Józwiak, Janina (Eds.): Population of Central and Eastern Europe. Challenges and Opportunities. Warsaw: Statistical Publishing Establishment: 457-484. Spéder, Zsolt; Kamarás, Ferenc 2008: Hungary: Secular fertility decline with distinct period  uctuations. In: Demographic Research 19,18: 599-664. https://doi.org/10.4054/DemRes.2008.19.18 Spéder, Zsolt; Kapitány, Balázs 2014: Failure to realize fertility intentions: A key aspect of the post-communist fertility transition. In: Population Research and Policy Review 33,3: 393-418. https://doi.org/10.1007/s11113-013-9313-6 Thornton, Arland 2005: Reading History Sideways. The Fallacy and Enduring Impact of the Developmental Paradigm on Family Life. Population and Development Series. Chicago: University of Chicago Press. Thornton, Arland; Young-DeMarco, Linda 2001: Four Decades of Trends in Attitudes. Toward Family Issues in the United States: The 1960s through the 1990s. In: Journal of Marriage and Family 63,4: 1009-1037. https://doi.org/10.1111/j.1741-3737.2001.01009.x Thornton, Arland; Philipov, Dimiter 2009: Sweeping changes in marriage, cohabitation and childbearing in Central and Eastern Europe: New insights from the developmental idealism framework. In: European Journal of Population 25,2: 123-156. https://doi.org/10.1007/s10680-009-9181-2 A Quarter Century of Change in Family and Gender-Role Attitudes in Hungary • 767 Tomka, Miklós; Zulehner, Paul Michael 2000: Religion im gesellschaftlichen Kontext Ost(Mittel)Europas. Schwabenverlag: Ost ldern. Tóth, Olga 1995: Attitűdváltozások a női munkavállalás megítélésében. In: Szociológiai Szemle 1: 71-86. Van De Kaa, Dirk 1987: Europe’s Second Demographic Transition. In: Population Bulletin 42,1: 1-57. Van De Kaa, Dirk 1996: Anchored Narratives: The Story and Findings of Half a Century of Research into the Determinants of Fertility. In: Population Studies 50,3: 389-432. https://doi.org/10.1080/0032472031000149546 Van De Kaa, Dirk 2001: Postmodern Fertility Preferences: from Changing Value Orientation to to New Behaviour. In: Population and Development Review 27: 290-331. Van De Kaa, Dirk 2002: The Idea of a Second Demographic Transition in Industrialized Countries. Paper presented at the Sixth Welfare Policy Seminar of the National Institute of Population and Social Security. Tokyo, Japan, 29 January 2002 [https://wp.aleteia. org/wp-content/uploads/sites/2/2013/08/kaa.pdf, 07.12.2023]. Van Vlimmeren, Eva; Moors, Guy B. D.; Gelissen, John P. T. M. 2017: Clusters of cultures: diversity in meaning of family value and gender role items across Europe. In: Quality and Quantity 51: 2737-2760. https://doi.org/10.1007/s11135-016-0422-2 Voicu, Malina; Tu is, Paula Andreea 2012: Trends in gender beliefs in Romania: 1993- 2008. In: Current Sociology 60,1: 61-80. https://doi.org/10.1177/0011392111426648 Zaidi, Batool; Morgan, S. Philip 2017: The Second Demographic Transition: a Review and Appraisal. In: Annual Review of Sociology 43: 473-492. https://doi.org/10.1146/annurev-soc-060116-053442 Date of submission: 30.08.2022 Date of acceptance: 03.11.2023 Zsolt Spéder (). Hungarian Demographic Research Institute. Budapest, Hungary. University of Pécs, Department of Sociology. Pécs, Hungary. E-mail: speder@demogra a.hu URL: http://www.demogra a.hu/en/staff-speder-zsolt • Zsolt Spéder768 Appendix Tab. A1: Average values of responses to family and gender-role attitudes Notes: The degree of agreement was measured on a  ve-point scale that ranged from strongly agree (5) through agree (4) neither agree nor disagree (3) and disagree (2) to strongly disagree (1). When interpreting responses we took higher values for equal gender relations, supporting female employment and non-conventional partnership forms, while a lower value stood for traditional gender roles, a male breadwinner family, and a conventional family (For the above reason values of response items were reversed which is marked by x). Source: ISSP 1988, 1994, 2002, 2013 1988 1994 2002 2013 Consequences of female employment A working mother can establish just as warm and secure a relationship with her children as a mother who does not work. 3.27 3.48 3.84 3.67 A pre-school child is likely to suffer if his or her mother works. (x) 2.21 1.93 2.22 2.59 All in all, family life suffers when the mother has a full-time job. (x) 2.42 2.21 2.55 2.70 Gender role, employment A job is all right, but what women really want is a home and children. (x) 2.11 1.93 2.25 2.39 Being a housewife is just as ful lling as working for pay. (x) 1.75 2.36 2.49 2.73 Both the man and the woman should contribute to the household income. 4.03 4.08 4.18 3.98 A man’s job is to earn money; a woman’s job is to look after the home and the family. (x) 2.88 2.32 2.81 2.64 Attitude towards partnership forms and children Married people are generally happier than unmarried people. (x) 2.53 2.30 2.47 2.88 People who want children ought to get married. (x) 2.38 2.49 2.92 2.88 Watching children grow up is life’s greatest joy. (x) 1.60 1.23 1.32 1.50 People who have never had children lead empty lives. (x) 1.89 1.68 1.93 2.13 A Quarter Century of Change in Family and Gender-Role Attitudes in Hungary • 769 Tab. A2: Percentage by which people supporting liberal gender roles outnumber people supporting conventional gender roles Notes: The degree of agreement was measured on a  ve-point scale that ranged from strongly agree (5) through agree (4) and disagree (2) to strongly disagree (1). When interpreting responses we took lower values as re ecting support for traditional family and partnership forms, while a higher value stood for a modern attitude, supporting gender equality and the plurality of partnership forms. (Responses were recoded in case of items noted by x). The numbers show the exceeding prevalence of liberal attitudes related to conventional attitudes, and the responses „neither agree nor disagree”(3) have been excluded. Source: ISSP 1988, 1994, 2002, 2013 1988 1994 2002 2013 A working mother can establish just as warm and secure a relationship with her children as a mother who does not work. 15.2 25.2 50.0 43.8 A pre-school child is likely to suffer if his or her mother works. (x) -52.7 -61.6 -47.4 -25.9 All in all, family life suffers when the mother has a full-time job. (x) -39.2 -46.5 -28.0 -20.0 A job is all right, but what women really want is a home and children. (x) -64.8 -64.8 -49.5 -42.7 Being a housewife is just as ful lling as working for pay. (x) -86.4 -39.7 -33.7 17.7 Both the man and the woman should contribute to the household income. 77.2 63.2 67.9 66.5 A man’s job is to earn money; a woman’s job is to look after the home and the family. (x) -4.0 -37.9 -10.2 -23.8 Married people are generally happier than unmarried people. (x) -31.4 -39.4 -35.9 -10.0 People who want children ought to get married. (x) -32.6 -26.1 -3.0 -6.8 Watching children grow up is life’s greatest joy. (x) -94.9 -96.7 -94.9 -91.7 People who have never had children lead empty lives. (x) -75.0 -73.1 -60.9 -55.7 • Zsolt Spéder770 Tab. A3: Trends of selected fertility and partnership indicators, 1988-2013, Hungary Source: + Vital statistics, HCSO; ++ Census, HCSO; +++ “Turning points of the life course”, estimations, HDRI Year Fertility indicators+ Partnership status at age 24-29++ Ratio of TFR Ratio of non- MAFB Single, Non-marital Married cohabitation marital births living alone cohabitation as  rst partnership+++ 1988 1.79 11.9 23.10 1989 1.78 12.4 23.07 38.0 1990 1.84 13.1 22.99 18.93 4.08 76.99 1991 1.85 14.1 23.02 1992 1.76 15.6 23.08 1993 1.68 17.6 23.11 45.5 1994 1.64 19.4 23.22 1995 1.57 20.6 23.43 1996 1.45 22.6 23.70 1997 1.37 25.0 23.93 1998 1.33 26.6 24.29 65.5 1999 1.29 28.0 24.69 2000 1.33 29.0 25.02 2001 1.31 30.3 25.33 32.99 14.62 52.39 2002 1.31 31.3 25.70 2003 1.28 32.3 26.10 81.7 2004 1.28 34.0 26.54 2005 1.32 34.9 26.96 2006 1.35 35.6 27.29 2007 1.32 37.5 27.55 2008 1.35 39.5 27.70 90.4 2009 1.33 40.8 27.92 2010 1.26 40.8 28.23 2011 1.24 42.3 28.34 48.06 25.27 26.67 2012 1.34 44.5 28.29 2013 1.34 46.2 28.23 Published by Federal Institute for Population Research (BiB) 65180 Wiesbaden / Germany Managing Publisher Dr. Nikola Sander 2023 Editor Prof. Frans Willekens Managing Editor Dr. Katrin Schiefer Editorial Assistant Beatriz Feiler-Fuchs Wiebke Hamann Layout Beatriz Feiler-Fuchs E-mail: cpos@bib.bund.de Scientifi c Advisory Board Kieron Barclay (Stockholm) Karsten Hank (Cologne) Ridhi Kashyap (Oxford) Natalie Nitsche (Rostock) Alyson van Raalte (Rostock) Pia S. Schober (Tübingen) Rainer Wehrhahn (Kiel) Comparative Population Studies www.comparativepopulationstudies.de ISSN: 1869-8980 (Print) – 1869-8999 (Internet) Board of Reviewers Bruno Arpino (Barcelona) Laura Bernardi (Lausanne) Gabriele Doblhammer (Rostock) Anette Eva Fasang (Berlin) Michael Feldhaus (Oldenburg) Alexia Fürnkranz-Prskawetz (Vienna) Birgit Glorius (Chemnitz) Fanny Janssen (Groningen) Frank Kalter (Mannheim) Stefanie Kley (Hamburg) Bernhard Köppen (Koblenz) Anne-Kristin Kuhnt (Rostock) Hill Kulu (St Andrews) Nadja Milewski (Wiesbaden) Roland Rau (Rostock) Thorsten Schneider (Leipzig) Tomas Sobotka (Vienna) Jeroen J. A. Spijker (Barcelona) Heike Trappe (Rostock) Helga de Valk (The Hague) Sergi Vidal (Barcelona) Michael Wagner (Cologne)