Gender Norms under Socialism and Capitalism: A Historical Examination of Attitudes towards Maternal Employment in the German Democratic Republic and the Federal Republic of Germany Gender Norms under Socialism and Capitalism: A Historical Examination of Attitudes towards Maternal Employment in the German Democratic Republic and the Federal Republic of Germany* Leonie Kleinschrot Abstract: Research on the Federal Republic of Germany (FRG) in the 1980s shows a high level of congruence between conservative social policy deterring mothers from employment and traditional societal gender norms. In contrast, little is known about whether people in the German Democratic Republic (GDR) agreed with the socialist idea of continuous full-time maternal employment. Based on unexploited GDR data from 1984 and a description of contemporary social policy, this study examines attitudes towards maternal employment, whether they were related to individual preferences for work or children, and their congruence with the socialist policy. The same questions are examined for the FRG using data from 1982. Results for the GDR indicate that one third of respondents rejected the socialist idea of maternal full-time employment, with individual work preferences being decisive for respondents’ assessments. In the FRG, there was a high degree of agreement with the gender norm of maternal non-employment, with this being dependent on individual preferences for children. These fi ndings complement post-reunifi cation evidence on East-West-differences in gender norms and provide insights into attitudes under Eastern European state socialism. Keywords: German Democratic Republic · Socialism · Gender Norms · Maternal Employment Comparative Population Studies Vol. 48 (2023): 151-182 (Date of release: 24.04.2023) Federal Institute for Population Research 2023 URL: www.comparativepopulationstudies.de DOI: https://doi.org/10.12765/CPoS-2023-07 URN: urn:nbn:de:bib-cpos-2023-07en3 * This article belongs to a special issue on “Demographic Developments in Eastern and Western Europe Before and After the Transformation of Socialist Countries”. • Leonie Kleinschrot152 1 Introduction The employment participation and reproductive behaviour of women in the German Democratic Republic (GDR) and the former Federal Republic of Germany (FRG) differed substantially. Women in the socialist GDR had children at an earlier age and were more frequently and extensively employed (Goldstein/Kreyenfeld 2011). Women in the capitalist FRG, on the other hand, focused more on housework and childcare. These behavioural differences are often attributed to the political frameworks (Pfau-Effi nger/Smidt 2011). While the GDR – which constantly sought to simultaneously compete with and demarcate itself from the West (Trappe 1996) – pushed for the full integration of women into the labour force and provided for institutional childcare from the fi rst months of life, social policy in the FRG strengthened the male-breadwinner model and the maternal provision of childcare, especially for young children (Adler et al. 2016). This article adds to the knowledge about the stark differences regarding maternal labour force behaviour and respective policy in the two former states by investigating individual attitudes towards maternal employment. Adopting a historical sociological perspective enables the prevailing gender norms to be examined in the former FRG, and the extent to which they corresponded to the policy framework. West German survey data starting in the early 1980s show that social policy and cultural norms concerning maternal employment were largely consistent. For example, shortly before German reunifi cation in 1990, almost three- quarters of respondents in the FRG agreed with the statement that pre-school aged children are bound to suffer if their mothers go out to work (Berth 2019). For the GDR, a similar investigation of gender norms has not yet been possible because the surveys developed in the West were only conducted in East Germany after reunifi cation. This study addresses this research gap, as it aims to assess the congruence of gender norms and social policy in the GDR by analysing attitudes towards maternal employment and by describing the GDR’s socialist women’s policy. In detail, the study descriptively examines the data and applies logistic regressions to fi nd out which attitudes existed towards the employment of mothers with young children in GDR society, what role individual preferences for work or children played here, and to what extent these attitudes corresponded to the socialist idea of continuous full-time maternal employment. Unexploited attitudinal data1 from the time of the GDR are used for these analyses. These data are from a survey (“Women’s Study”) conducted by the Central Institute for Youth Research (“Zentralinstitut für Jugendforschung”) in 1984. They offer a hitherto unexploited opportunity to analyse the attitudes of cohorts who were socialised in the GDR, were of reproductive age, and lived in the GDR at the time of the survey. However, the methodological 1 So far, the data have only been analysed descriptively in the context of project reports by the ZIJ at the time of the GDR. An overview of the study reports that are accessible can be found on the GESIS homepage: https://search.gesis.org/research_data/ZA6138?doi=10.4232/1.6138. Gender Norms under Socialism and Capitalism • 153 limitations of data collected by a politically controlled research institute in the GDR need to be considered. In addition, the results of an equivalent analysis addressing the same questions but using data from the FRG in the early 1980s, the German General Social Survey (GGSS) from 1982,2 are also presented. It should be noted that these two datasets do not allow a one-to-one comparison of the two former states. This study not only enhances German attitudinal research by including the period before reunifi cation, but it is also of interest to international discussions. First, it contributes to the comparatively scarce quantitative research on gender norms under socialist regimes in Eastern Europe (focusing on Hungary, e.g. Panayotova/ Brayfi eld 1997; Braun/Scott 2009), and lays some groundwork for interpreting cultural developments after 1989. Second, the survey data from the GDR allow for a period-specifi c examination of a socialist and a capitalist European state in the early 1980s. In addition, the article takes a step towards accessing the wealth of survey data collected by the Central Institute for Youth Research in the GDR during its almost 25 years of existence (Friedrich 1991). In the next section, an outline of the theoretical background of the study is given, followed by a description of the social policy frameworks in the GDR and FRG at the time of the surveys. In section 4, the current state of research, as well as the hypotheses for the analyses are presented. In section 5, the methodological approach is described. Finally, the results are reported and discussed. 2 Theoretical background In the literature, gender norms are understood as consensually defi ned and socially known beliefs and expectations concerning the behaviour of women and men within and outside the family (Duncan/Edwards 1997; Pfau-Effi nger/Smidt 2011; Grunow/ Veltkamp 2016), e.g. beliefs about the extent to which men should participate in childcare or whether women should actively engage in politics. This study will be looking at one of the aspects of gender norms, namely beliefs concerning maternal employment. Various mechanisms for the potential congruence of gender norms and social policy are theoretically discussed and, feminist literature in particular emphasises the complex interplay and continual mutual impacts of the two (Pfau-Effi nger 2005; Kremer 2007). Social policy refl ects and institutionalises the gender norms of a society, for instance, as demonstrated by the lack of institutional childcare for under 3-year-olds in the FRG in the 1980s. Since social policy instruments, according to this understanding, include consensual solutions to the motherhood-paid work arrangement, they function as an “institutional point of reference” (Grunow/ Veltkamp 2016: 12-13) for the appropriate employment behaviour of mothers (Pfau- Effi nger 2005; Kremer 2007), and serve “as normative anchors” (Gangl/Ziefl e 2015: 2 Since the relevant question was not asked in the GGSS 1984, the 1982 wave is used. • Leonie Kleinschrot154 519) in forming individual preferences and attitudes towards work and children (Sjöberg 2004). In this sense, social policy is embedded in and legitimised by the normative societal context, meaning that it may experience support or disapproval, for example in elections (Raven et al. 2011). Consequently, social policy is also assumed to be adapted in response to changing societal norms (Svallfors 2010). At the same time, specifi c social policy instruments, such as the extensive supply of institutional day-care in the GDR, determine parents’ scope of action (Sjöberg 2004). As a result of individuals’ experiences in role behaviour within this policy framework, they adapt their preferences accordingly. Following exposure theory, greater exposure to a work-related setting, e.g. by being employed, will tend to increase mothers’ preferences for work (Gangl/Ziefl e 2015). In addition to individual experiences, “social multiplier effects” (Gangl/Ziefl e 2015: 519) can arise, as observing relevant others’ role behaviour, e.g. other mothers working, leads an individual to internalise this behaviour and shape its preferences accordingly. Indeed, the GDR’s totalitarian regime even sought to actively harness the role of policy as a determinant of experiences and as a normative anchor. East German state socialism permeated social life and actively intervened in it. Freedom of the press, of speech and of assembly, but also free elections and culture were subject to the hegemony of the state party (Weber 2011). At the same time, there were areas, in which citizens could lead their “private” lives more or less autonomously, as long as they did not challenge the state’s monopoly on power (Wolle 1998; Lindenberger 2002; Bauerkämper 2005). Therefore, the question arises of how successful the GDR regime was in its endeavour to normatively shape society in terms of gender roles. This study tries to answer this broad question by looking at one part of it – i.e. attitudes towards maternal employment – in more detail. How experiences with gender roles can become crucial for individual attitudes and preferences is described by Bielby and Bielby (1989). They argue that actively fulfi lling family or work roles defi ned by gender norms leads to identifi cation with these roles, and to commitment to the life domains of family or work. The strength of commitment depends on the extent to which the gender role behaviour is meaningful and identity-forming for the individual. Thus, commitments bind individuals to the gender norms of the society they live in (Bielby 1992). If mothers are employed, identities in both domains emerge. And in a society with traditional gender norms, a confl ict arises, which forces mothers to trade off one domain against the other because they are expected to focus on care work. Working fathers, on the other hand, do not experience an identity confl ict because their work identity coincides with the normative expectation of men as the family breadwinners. In a way, Hakim (2002) argues the other way round, since she states that individual preferences for work or children are the main predictors of parents’ work and care behaviour. Differentiating between women according to their preferences, leads to three ideal typical categories in any society (Hakim 2002). First, adaptive women prefer to combine employment and children without giving absolute priority to either of the two. They therefore react strongly in their preferences to social and labour market policy. Second, there are work-centred women, for whom employment or an equivalent activity in the public sphere (e.g. sport, politics) is most important to Gender Norms under Socialism and Capitalism • 155 their lives, and thus they arrange their family life around their work and are more responsive to opportunities than to policies. Third, home-centred women prioritise children and the household and prefer not to work outside the home. This group is strongly responsive to social, but not to labour market policies. According to Hakim (2002), men differ signifi cantly from women in their preferences, as they are very homogeneously work-centred – something that is also shaped by societal norms. Even though an empirical disentanglement of the discussed mechanisms is beyond the scope of this paper and not possible with the data used here, the theoretical background on the complex interplay is important for understanding why not only gender role behaviour and social policy might be congruent, but also social policy and attitudes or preferences towards work and children. Before the hypotheses for the analysis are derived in section 4, the social policy framework of both former states in the early 1980s will be described. 3 Social policy in the FRG and GDR This section presents both states’ social policies focusing on maternal employment at the time of data collection (“GGSS” in 1982 and “Women’s Study” in 1984). The FRG in 19823 In the FRG, the policy framework of the conservative-corporatist welfare state not only supported the male-breadwinner model, but also conformed to the gender norm according to which women should prioritise childcare over employment, and children especially at a very young age need their mothers and the care only she can provide (Schütze 1986; Oechsle 1998; McDonald 2000; Pfau-Effi nger 2004). For many years, the domestic responsibilities of married women were even prescribed by the marriage and family law in the FRG. Only in 1977 was this law reformed, removing the wife’s duty to maintain the household, and the husband’s de jure fi nal decision on his wife’s employment. While many kindergarten places for 3-6 year-olds were created before 1982, there were hardly any day-care places for under 3-year-olds or school-aged children. The tax and social security system favoured the male-breadwinner model. Paid maternity leave for employed mothers consisted of just 14 weeks of full wage compensation, which could be followed by another four months in which 750 Deutsche Mark per month were paid, and child (sickness) benefi ts were available (Helwig 1993; Rosenfeld et al. 2004; Leitner et al. 2008).4 Women’s employment rates (51 percent in 1982) were considerably lower than those of men (83 percent) (Statistisches Bundesamt 1984: 96), and almost one third of employed women did not work in a full-time job of 40 hours per week, mainly for 3 The description for the FRG is briefer than for the GDR, but see Gerlach (2008) for a detailed description of individual family policy measures, and Ostner (1993) and Kolinsky (1989) for comprehensive overviews of social policy in the FRG in the 1980s. 4 The paid parental leave (“Erziehungsurlaub”) for all parents was only introduced in 1986. • Leonie Kleinschrot156 family reasons. Mostly, part-time jobs involved less than 21 hours of work per week (Statistisches Bundesamt 1984: 97; Rosenfeld et al. 2004). The male-breadwinner model, in which men were mainly responsible for paid work and women for care work, was the most frequent family model in the FRG (Rosenfeld et al. 2004). Despite this, in the FRG there were working mothers, mostly in part-time and often due to fi nancial reasons, but also because they did not want to leave the workplace (Becker- Schmidt 2010). In 1982, 36 percent of mothers whose youngest child was under six years of age were employed (Statistisches Bundesamt 1984: 103). Nevertheless, maternal employment remained the exception if there were children under the age of three, not least because of the lack of institutional childcare for this age group. In 1982 only 26 245 places were available for under 3-year-olds, which amounted to a childcare rate of 2 percent (Bundesminister für Jugend Familie, Frauen und Gesundheit 1986: 36). The GDR in 1984 The GDR state party regime followed “the Marxist conviction that women’s emancipation requires their formal participation in the labor force” (Adler/Brayfi eld 1997: 248), so the socialist gender norm was that of a mother who worked full- time throughout all phases of the family and who put her children (of any age) into institutional day-care (Gysi/Meyer 1993). Theoretically, generous policies were put in place to enable formal gender equality, but in real life women came to be treated as “reproducers and producers” while men were seen only as “producers” (Trappe 1996: 358). In fact, women remained primarily responsible for care work (Rosenfeld et al. 2004; Banaszak 2006; Lee et al. 2007). Although, since employment was regarded as crucial for emancipation, the regime did promote women’s employment for reasons of gender equality, economic motives were also at play, as women constituted an important resource for the socialist planned economy (Trappe 1996). Consequently, the GDR had one of the highest female employment rates5 in the world (Lee et al. 2007), standing at 79 percent in 1984 (Staatliche Zentralverwaltung für Statistik 1986), and at the same time, the total fertility rate of 1.7 was slightly higher than that of the FRG (1.5 in 1982) (Winkler 1990: 25; GENESIS online 2021). Although women were almost as likely as men to be employed in the GDR, they more often held positions of lower responsibility, often below their qualifi cations, and were paid less than men6 (Rosenfeld et al. 2004; Banaszak 2006; Lee et al. 2007). 5 The employment rates in the GDR and FRG cannot be compared directly, as the defi nition of employed persons differed (in the GDR mothers on “baby year”, apprentices and employees of certain sectors, e.g. state security, were excluded), as did data collection (the GDR had full data collection via employers, whereas the FRG statistics are based on surveys, partial and random sample collection and on secondary statistical collections). For details see Fritz (2000). There is no information on maternal employment in the published statistical yearbooks of the GDR. However, this should not differ greatly from the female employment rate, since childlessness was very low (Gysi/Meyer 1993; Goldstein/Kreyenfeld 2011). 6 Studies showed that the gender pay gap was quite similar in both states (Rosenfeld et al. 2004). For the time of reunifi cation, Szydlik (1994) calculated a gender pay gap of 18 percent in the FRG and 15 percent in the GDR. Gender Norms under Socialism and Capitalism • 157 Most employed women worked full-time, although part-time employment (which mostly meant working between 25 and 35 hours per week) was not entirely unusual (Winkler 1990). The 27 percent of women not working the regular 43.75 hours per week (Winkler 1990: 107) were mostly over 35, i.e. not mothers with young children, and worked less for health reasons, or because they were in their pre-retirement phase (Winkler 1990; Rosenfeld et al. 2004). One of the factors contributing to the high female employment rate was the high availability of day-care places, also for children under three years of age. In 1984, 330 812 young children attended day-care, which resulted in an offi cial childcare rate of 69 percent7 (Staatliche Zentralverwaltung für Statistik 1986). Further factors might have been that work avoidance was prosecuted by law and could even be punished by imprisonment, being employed was seen as a moral obligation in East German socialism, many social entitlements were connected to full-time employment, and often one wage alone was not suffi cient to support the family (Trappe et al. 2015). In addition, women worked for self-affi rmation and fi nancial independence, in order to be part of social networks (the nationally-owned enterprises even offered leisure events, holiday trips etc.), and because work was considered a core value in life (Gysi/Meyer 1993; Lindenberger 2002; Leitner et al. 2008). As birth rates declined from the mid-1960s onwards, a wide range of policy measures were introduced to better support women in reconciling family and paid work (Nickel 1998; Leitner et al. 2008; Trappe et al. 2015). Public childcare was further expanded (children could enter from the age of fi ve months); mothers with two children were allowed to reduce their weekly working hours to 40 without a wage reduction; maternity leave with full wage compensation covered 26 weeks and could be followed by a well-compensated (50-90 percent) paid “baby year” (before 1986 only mothers with a second child were eligible); a birth bonus of 1 000 Mark was paid; extra holidays and child (sickness) benefi ts were given to mothers; and mothers were entitled to one paid day off work per month for housework (Winkler 1990; Helwig 1993; Schmidt 1996; Nickel 1998). Since gender equality (that was predominantly focused on employment) was “issued” by the male-dominated political leadership, in contrast to equality evolving from a feminist movement, Nickel (1998: 26) speaks of “patriarchal equality” in the GDR. 4 Literature review and hypotheses The numerous social policy instruments in the GDR and state socialist propaganda about the importance of employment – which had already been in place for 35 years by the time of the survey and thus had had a great impact on the respondents’ 7 The childcare rate was the ratio of children in day-care to those who were eligible for institutional care. Thus, those children whose mothers were eligible for paid leave, e.g. during the “baby year”, were not taken into account. • Leonie Kleinschrot158 socialisation – can be expected to have functioned as a “normative anchor” (Kremer 2007; Gangl/Ziefl e 2015; Grunow/Veltkamp 2016), and therefore to have set the norm of combining motherhood and full-time employment. The exposure to employment and institutional childcare might have also strengthened maternal preferences for full-time employment, which, since almost all mothers in the GDR were employed, could have been reinforced by a “social multiplier effect” (Gangl/ Ziefl e 2015). Several empirical studies – e.g. using the available GGSS waves from 1991 until now – show that over this time period, people in East Germany were far less negative in their opinion about maternal employment than those in West Germany (Banaszak 2006; Lee et al. 2007; Bauernschuster/Rainer 2012; Ebner et al. 2020). In 2016, nearly 30 years after reunifi cation, East German women born between 1950 and the mid-1970s – who were primarily socialised in the GDR – are those who most often agree with the statement that a full-time working mother can usually have just as meaningful a relationship to their toddler as a mother who does not work (around 90 percent agreement), compared to the East cohorts born after 1975 (79 percent agreement), or West German cohorts born before 1950 (69 percent agreement), between 1950 and 1975 (63 percent), and after 1975 (58 percent) (Barth et al. 2020). Following the theoretical arguments and post-reunifi cation fi ndings, it is expected that the vast majority of people in the GDR supported the socialist idea of full-time maternal employment (hypothesis 1). Referring to the same theoretical mechanisms (social policy as a normative anchor and social multiplier effects), the FRG’s conservative-corporatist social policy, including the lack of institutional childcare and its tax and social security system that favoured the male-breadwinner model, could have supported the norm of maternal childcare. Furthermore, the role experience of being a housewife, as well as the observation of other non-working mothers in the FRG might have led individuals to adapt their attitudes accordingly. Thus, it is expected that in the early 1980s people in the FRG were more negative in their opinion about maternal employment than people in the GDR (hypothesis 2). Since living up to gender norms in the GDR meant full-time employment for mothers, following identity formation theory (Bielby/Bielby 1989) one could expect that women in the GDR were strongly committed to both work and children. And in fact, studies show that shortly after reunifi cation and until recently, people in East Germany had high preferences for children and work, and did not prioritise one over the other (Braun et al. 1994; Adler/Brayfi eld 1997; Arránz Becker et al. 2010; Beblo/Görges 2018; Görges 2021). This is particularly true for the oldest cohorts, who had experienced the social policy of the GDR regime (Beblo/Görges 2018; Görges 2021). Thus, it is expected, that people in the GDR with high preferences for work, had a higher likelihood of agreeing with the socialist idea of full-time maternal employment compared to those with low preferences for work (hypothesis 3a). With regard to preferences for children, it can be stated that female employment in the GDR was not affected by motherhood because of the extensive availability of institutional childcare and other reconciliation measures. Therefore, no association between individual preferences for children and the likelihood of agreeing with full- time maternal employment is expected (hypothesis 4a). Due to the equally high Gender Norms under Socialism and Capitalism • 159 participation rates of men and women in the labour market, which was probably also accompanied by a high commitment to employment among women, it is expected that the association between the preferences for work and the likelihood of agreeing with mothers’ employment did not differ between women and men in the GDR (hypothesis 5a). In the FRG, due to the high prevalence of traditional gender norms (Lee et al. 2007) and the corresponding social policy framework supporting the male-breadwinner model, one could expect an identity confl ict for mothers who planned to work (Bielby/Bielby 1989), which would compel them to trade off children for work. Thus, it is hypothesised that either people who had low preferences for work, or those who had high preferences for children, were more likely to agree that mothers with young children should not work (hypothesis 3b and 4b). However, a gender difference in the impact of preferences for children or work on attitudes towards maternal employment could be expected in the FRG. Empirically, it has been shown that women in the FRG rate work as less important, and having children as more important to their lives than men (Beblo/Görges 2018; Görges 2021). Following the theoretical arguments on identity formation, men in the FRG particularly emphasised their identity as the family breadwinner, and according to Hakim (2002) could be categorised as “work-centred”. Thus, regardless of whether they had high preferences for children or for work, they are expected to be more likely to agree that their female partners should not be employed, but should care for the children. In contrast, “work-centred” women would be expected to arrange their lives around employment (Hakim 2002), and to be less likely to agree that small children suffer if their mothers work. However, if they were “home-centred” (Hakim 2002) and had developed a strong family identity, it could be expected that they were more likely to agree to maternal non-employment to avoid a confl ict of identities. It is thus hypothesised that the impact of preferences for children or work on attitudes towards maternal employment differed by gender in the FRG (hypothesis 5b). 5 Method 5.1 Sample and measures: GDR In the GDR, social science research was strictly politically controlled. Representative studies were rare because an elaborate approval procedure at the highest political level was required, which interfered with research topics and survey instruments. The empirical studies were therefore mostly regionally confi ned so that only the subordinate political authorities had to be involved (Brislinger et al. 1998). The data collection then took place independently, and predominantly in the form of paper-pencil surveys conducted anonymously in group settings (Brislinger et al. 1998; Griese 1991). Even though respondents might not have had full confi dence in the promised anonymity, the surveys were also seen as an opportunity to • Leonie Kleinschrot160 express opinions without fear of consequences8 (see also Griese 1991). Besides, the questions of the 1984 “Women’s Study” were rather innocuous compared to those of other surveys, which asked, for example, about the personal evaluation of the socialist regime. Even though the results of studies had to be kept secret and evaluated in line with political convictions, empirical research in the GDR followed western scientifi c standards (Bertram 1997; Brislinger et al. 1998; Griese 1991). One of the two datasets used in this paper, the “Women’s Study” 1984, was conducted by the GDR’s Central Institute for Youth Research (ZIJ) (Zentralinstitut für Jugendforschung, Leipzig 1984).9 The ZIJ existed from 1966 to 1989 in Leipzig and undertook – mostly quantitative – social science research on the attitudes and living conditions of youths and young people (for example regarding work, family, education, sexuality or leisure) in the GDR. In its later phase, it additionally included a small working group on women’s studies (Friedrich 1991). The ZIJ “Women’s Study 1984” asked women and men about, among other things, their life planning, general goals in life, leisure activities, preferred ways of living, political opinions and attitudes, as well as behaviour relating to gender equality. The respondents answered the questionnaires partly at their workplace and partly at home. The sample came from the former districts of Erfurt and Karl- Marx-Stadt10 where respondents were approached via their workplace. The data are not representative of the population of the GDR as a whole, but consist of an initial sample of 1623 respondents who were between 18 and 37 years old. The following analyses only include respondents who gave information on their gender (N=112 missing values) and the dependent variable (N=192 missing). Due to some further missing values, the fi nal sample consists of 1300 respondents, of whom 76 percent are female and 24 percent are male (see Table 1). The overrepresentation of women needs to be taken into account when interpreting the results. The dependent variable for the GDR regression analyses is whether the respondent agreed with the socialist norm of full-time maternal employment. This is captured by respondents’ answers to the following question: “Is it desirable to limit your own/a mother’s employment activity if children under three are part of the family?” The three answer options were “by all means”, “under certain circumstances only”, “not at all”. To construct the binary coded dependent variable, the last two answers are combined and coded as “1”. Since mothers with at least two young children could reduce their working time to 40 hours per week on full pay, the answer option “under certain circumstances only” is also in line with the socialist idea of full-time maternal employment. The answer “by all means” indicates disagreement with full-time maternal employment with young children, as it prioritises childcare over employment. It is coded as “0”. 8 This assessment originates from Dr Uta Schlegel, with whom we spoke in preparation for this paper on 17.12.2020. She was a former researcher and department head at the ZIJ. 9 After the fall of the Berlin Wall, the majority of social science data collected in the GDR were checked, the content and methodology were specifi ed and were then archived for secondary analyses (Brislinger et al. 1998). 10 Karl-Marx-Stadt is now called Chemnitz. Gender Norms under Socialism and Capitalism • 161 The two central explanatory variables refl ect respondents’ preferences for work or children, measured by the question: “Indicate how important the life goals or areas of life mentioned are for you”. Respondents rated the items from 1 = totally unimportant to 4 = very important. For the analyses, the variables were dichotomised by grouping the fi rst three response categories as not very important as an indication of low preference, and the last category as very important (high preference), since the vast majority of responses for these items fell into category four. Tab. 1: Descriptive statistics GDR 1984 FRG 1982 Mean / % SD Mean / % SD Agreement with the Norm concerning Maternal Employment: No 36.5 17.8 Yes 63.5 82.2 Preferences for Work: Low 15.2 10.1 High 84.8 89.9 Preferences for Children: Low 14.3 11.4 High 85.7 88.6 Family Status: Single 13.0 41.7 LAT/Married, living apart 15.5 2.1 Cohabiting 13.6 2.8 Married 57.9 53.4 Employment Status: Not employed n.a. 37.2 Part-time 8.3 11.1 Full-time 91.7 51.7 Education: Low 9.9 10.3 Middle 76.9 62.2 High 13.2 27.5 Age 24.3 3.5 27.8 5.5 Gender: Female 76.2 53.6 Male 23.9 46.4 N 1300 1009 Source: Women’s Study 1984, GGSS 1982; own calculations • Leonie Kleinschrot162 The logistic regression models control for respondents’ gender, age, education, employment status, and family status, as several studies have shown that these factors are signifi cant predictors of attitudes towards maternal employment (Alwin/ Scott 1996; Panayotova/Brayfi eld 1997; Bolzendahl/Myers 2004; Brooks/Bolzendahl 2004). The problem of the limited cross-national comparability of educational qualifi cations has already been pointed out (e.g. Hadjar/Berger 2010; Schneider 2016). Here, the recommendation made by Hadjar and Berger (2010) of classifying education into three categories (low, middle, high) is followed (see Appendix Table 1). Employment status was categorised by respondents as part-time or full-time, but their weekly working hours were not recorded. Because respondents were recruited via their employers, the category “not employed” did not exist in the GDR data, which has to be kept in mind when interpreting the results. Nevertheless, this does not substantially limit the data, since in the GDR, unemployment in the surveyed age group was almost non-existent. Family status is coded into single, living-apart- together (LAT), cohabiting and married. 5.2 Sample and measures: FRG For the FRG analyses, the dataset of the German General Social Survey cumulation 1980-1986 for the year 1982 is used (GESIS – Leibniz-Institut für Sozialwissenschaften 2002). The survey was conducted in West Germany (including West Berlin), and only persons with German citizenship were interviewed. The population from which the sample was drawn, consisted of all persons eligible to vote and living in private households. In the early GGSS waves (1982-1992, 1998), the representative sample was generated through ADM sample design with the selection stages 1) electoral districts, 2) households, 3) persons in households. Since with this procedure, a respondent’s chance of selection depends on the size of the household, a weight should be applied for analyses at the individual level. In research practice, this requirement is usually not met because (un-)weighted results do not differ or do so only slightly (GESIS – Leibniz-Institut für Sozialwissenschaften 2021). This is also the case for this paper: the use of weighted data changes the results only marginally, thus the results presented are unweighted. In the 1982 GGSS wave, 2991 persons initially took part. To ensure better comparability with the GDR sample, the FRG sample is also adjusted to the age range of 18-37. The fi nal analytic sample for this paper comprises N=1009, of whom 54 percent are female and 46 percent are male respondents (see Table 1). The dependent variable for the FRG regression analyses is whether respondents agreed with the norm of maternal non-employment. It is operationalised by respondents’ agreement with the statement “A small child is bound to suffer if his or her mother goes out to work”11 on a four-point scale from “completely agree” to “completely disagree”. For comparability with the GDR analyses, the dependent 11 The English translations were taken from Terwey/Baumann (2013). Gender Norms under Socialism and Capitalism • 163 variable is also dichotomised. The two affi rmative response options are taken together and coded as “1”, which signifi es support for the idea that mothers with small children should not be employed. The two negative answer options are coded as “0”, meaning a person did not agree with the then prevailing norm concerning maternal employment. The two central explanatory variables are also the preferences for work and children. Respondents answered the question, “The cards here list various spheres of life. We would like to know how important each of these spheres of life is for you”, on a seven-point scale with only the two endpoints headed “(1) unimportant” and “(7) very important”. For the analyses, as it was not made explicit what the middle category actually meant and “important” was given one additional rating (“very”) which “unimportant” did not receive, the variables are dichotomised by grouping 1 to 3 as unimportant, meaning low preferences and 4 to 7 as important, meaning high preferences. The same control variables as in the GDR analyses are included. For the employment status respondents’ self-categorisation12 as “not employed” (which includes unemployed, housewife, doing compulsory military/alternative community service, retired, at school/university/in training, not working for other reasons), “part-time” and “full-time” is used. Family status is coded as “single”, “married but living separately” ,13 “cohabiting” and “married”. 5.3 Analytical strategy Binary logistic regression models are applied to agreement with the prevailing norm concerning maternal employment considering the preferences for work and children after controlling for gender, age, education, employment and family status. The models are estimated separately for each dataset, i.e. for each state. To be able to compare coeffi cients of the explanatory variables, average marginal effects (AME) are predicted, as AME are hardly affected by uncorrelated unobserved heterogeneity (Mood 2010; Wooldridge 2010). The question of whether there are differences between women and men is investigated by including interaction terms. For sensitivity analyses in both country models, the two main explanatory variables are also tested in separate models (see Appendix Table A2 and A3). However, the results do not differ from those reported below. To further test the robustness of the results, additional controls were included for the size of the community respondents lived in and the number of children they had, as individuals might differ in attitudes due to differing social environments concerning maternal employment rates (urban vs. rural), or role exposure experiences (parenthood vs. childlessness) (Bolzendahl/Myers 2004). However, including them in the FRG analysis neither changes signifi cance levels and coeffi cient sizes, nor improves model fi t. Including them in the GDR models results in better model fi ts and partly 12 Weekly working hours were not surveyed in GGSS 1982. 13 This category can represent LAT relationships, but also separated couples that are still married. • Leonie Kleinschrot164 reveals signifi cant associations with the dependent variable. Nevertheless, it was decided not to include community size and the number of children in the GDR regression analysis as the sample size would be dramatically reduced from 1300 observations to 786 or 871. 6 Results 6.1 Descriptive results on agreement with the prevailing norm concerning maternal employment In the GDR, 9 percent of the respondents thought that mothers with children under the age of three should not limit their employment, 55 percent thought they should only limit it under certain circumstances. Thus, 64 percent agreed with the socialist idea of full-time working mothers (see Table 1). This also means that more than one third (37 percent) of the respondents thought that a mother should by all means limit her employment activity if young children were part of the family (see Table 1). When looking at the attitudes of different subgroups, it becomes clear that they were quite homogenous (see Fig. 1). However, respondents in part-time and full- time employment differed signifi cantly (p=0.000) in their attitudes towards full-time maternal employment, with 44 percent of the former and 65 percent of the latter agreeing with it. The small number of part-time employees has to be considered here. Female and male respondents in the GDR neither differed signifi cantly in their attitudes towards maternal full-time employment (62 percent of women agreed and 67 percent of men), nor in their preferences for work or children. The majority of both female and male respondents had high preferences for work (84 percent of women and 87 percent of men), as well as for children (86 percent of women and 84 percent of men). In the FRG, women and men differed signifi cantly in their preferences for work (p=0.003) or children (p=0.004). 87 percent of women and 93 percent of men rated work as important, and 91 percent of women rated having children as important compared to 85 percent of men. But men and women did not differ in their assessment of maternal employment. In the FRG, 82 percent of the respondents completely agreed/tended to agree with negative consequences for young children if their mother went out to work, and only 18 percent tended to disagree/completely disagreed. Thus, the vast majority endorsed the corresponding norm of maternal non-employment (see Table 1). Figure 2 shows that in the FRG, the proportion of non-employed respondents who thought young children suffer because of their mother’s employment was signifi cantly greater (87 percent) than that of the part- time employed (75 percent, p=0.01) and that of the full-time employed (80 percent, p=0.04). Summarising the descriptive fi ndings, it appears that around two thirds of respondents in the GDR agreed with the socialist norm of full-time maternal employment if there are children under the age of three in the family. Yet, the fact that more than one third disagreed with this norm indicates greater heterogeneity in Gender Norms under Socialism and Capitalism • 165 attitudes than expected according to hypothesis 1. The second central descriptive fi nding is consistent with hypothesis 2: compared to respondents in the GDR, those in the FRG more often evaluated maternal employment negatively and assumed that young children would suffer if their mothers were employed. 6.2 GDR: Results on the relation between preferences for work or children and attitudes towards maternal employment The multivariate analyses examined whether agreeing with the maternal employment norm in the two states depended on the individual preferences for work or children (Hakim 2002). Due to the different data sets and measurements, the regression results of the two states cannot be compared directly. The infl uence of preferences for work or children can only be considered within the countries. Logistic regression Fig. 1: Responses to the question: “Is it desirable to limit your own/a mother’s employment activity if children under three are part of the family?”, in percent, GDR 1984 Sex Cohort 0 20 40 60 80 100 in percent male female not at all under certain circumstances only by all means 0 20 40 60 80 100 in percent 1947-50 1951-55 1956-60 1961-66 Family status Education 0 20 40 60 80 100 in percent single LAT cohabitation marriage 0 20 40 60 80 100 in percent low middle high Employment status 0 20 40 60 80 100 in percent part-time full-time Source: Women’s Study 1984, own calculations • Leonie Kleinschrot166 models are estimated for each state and the results are given as AME. In model 1, only the preferences for work and children are included, while model 2 additionally includes all control variables. To identify potential differences between women and men, interaction terms are also included. Table 2 shows the results for the GDR. On average across all respondents, those who had high preferences for work had a higher probability (11 percentage points) of agreeing with the norm of full-time maternal employment with young children than those who rated work as not very important.14 This fi nding remained robust Fig. 2: Responses to the question: “A small child is bound to suffer if his or her mother goes out to work”, in percent, FRG 1982 Sex Cohort 0 20 40 60 80 100 in percent male female Completely agree Tend to agree Tend to disagree Completely disagree 0 20 40 60 80 100 in percent 1945-1949 1950-1954 1955-1959 1960-1964 Family status Education 0 20 40 60 80 100 in percent single married, living apart cohabitation marriage 0 20 40 60 80 100 in percent low middle high Employment status 0 20 40 60 80 100 in percent not emloyed part-time full-time Source: GGSS 1982, own calculations 14 As mentioned above, the sample did not include people who were not employed. However, unemployment was – due to the reasons given previously – almost non-existent in the GDR in the age group considered, and the two variables “employment status” and “importance of career and work” were uncorrelated. Gender Norms under Socialism and Capitalism • 167 after including all control variables and is in line with hypothesis 3a. While the preferences for work were signifi cantly and substantially associated with agreeing with the norm of full-time maternal employment, preferences for children showed a non-signifi cant coeffi cient, as expected in hypothesis 4a. The interaction of the preferences for work and respondents’ gender, which was included in model 2, was signifi cant and tests confi rmed that the effect of the preferences for work among women differed signifi cantly from that of men. This Tab. 2: Agreement with the norm of full-time maternal employment, GDR 1984, average marginal effects Model (1) (2) Preferences for Work (ref.: low) high 0.119** 0.110** (0.038) (0.038) Preferences for Children (ref.: low) high 0.054 0.067 (0.039) (0.040) Sex (ref.: male) female -0.030 (0.032) Age 0.001 (0.004) Family Status (ref.: single) married -0.023 (0.043) LAT -0.039 (0.050) cohabiting 0.006 (0.051) Education (ref.: low) middle -0.004 (0.045) high 0.011 (0.055) Employment Status (ref.: part-time) full-time 0.200*** (0.052) Observations 1300 1300 Pseudo R2MF 0.007 0.019 Note: Standard errors in parentheses. * p < 0.05, ** p < 0.01, *** p < 0.001 Source: Women’s Study 1984, own calculations • Leonie Kleinschrot168 is in contrast to hypothesis 5a, which expected no difference. As can be seen in Appendix Figure A1, a signifi cant association between preferences for work and attitudes towards maternal employment was only found among women. For the GDR, it can be summarised that female respondents’ preferences for work (but not for children) were associated with their agreement with the socialist idea of full-time maternal employment with children under the age of three. In other words, women who placed high importance on being employed did not even see children as a reason for limiting their employment activity. 6.3 FRG: Results on the relation between preferences for work or children and attitudes towards maternal employment Analyses of the FRG data show that respondents’ preferences for work were not signifi cantly related to their attitudes towards maternal employment with young children, contradicting hypothesis 3b suggesting an association (see Table 3). Yet, the preferences for children were signifi cantly related to attitudes towards maternal employment, based on a comparatively powerful coeffi cient. On average across all respondents, those who had high preferences for children, had a higher probability (12 percentage points) of agreeing with the statement that young children suffer if their mothers work than those who had low preferences. This is in line with hypothesis 4b. To test whether the effect of the preferences for children differed signifi cantly between women and men, an interaction term with the respondents’ gender was integrated into the model. However, the interaction term was not signifi cant, and neither the models with and without the interaction term, nor the effect in the group of women and men differed signifi cantly (see Appendix Figure A2). Thus, there was no signifi cant difference in the effect of the preferences for children between women and men, which is inconsistent with hypothesis 5b assuming gender differences. For the FRG, it can be stated that respondents who had high preferences for children were more likely to agree with the norm that mothers of young children should not work. At the same time, the preferences for work were not relevant for their attitudes towards maternal employment. 7 Discussion and conclusion This study fi rstly investigated the congruence of social policy and gender norms in the German Democratic Republic (GDR) by analysing the attitudes of 18-37-year- old respondents towards maternal employment, and describing socialist women’s policy. Using previously unanalysed survey data, the “Women’s Study” from 1984, this study examined how GDR respondents viewed the employment of mothers with young children, whether this was related to individual preferences for work or children, and to what extent these attitudes were consistent with the socialist idea of continuous full-time maternal employment. Secondly, a similar analysis was carried out with same-age respondents from the Federal Republic of Germany (FRG), using Gender Norms under Socialism and Capitalism • 169 the German General Social Survey from 1982. The results were presented side by side, and are now interpreted separately because it is not possible to directly compare the two former states on the basis of the datasets used. The study found that more than one third (37 percent) of the respondents in the GDR rejected the socialist idea of mothers with children under the age of Tab. 3: Agreement with the norm of maternal non-employment, FRG 1982, average marginal effects Model (1) (2) Preferences for Work (ref.: low) high -0.043 -0.036 (0.036) (0.038) Preferences for Children (ref.: low) high 0.114* 0.124* (0.045) (0.049) Sex (ref.: male) female -0.042 (0.027) Age 0.002 (0.003) Family Status (ref.: single) married -0.006 (0.030) LAT 0.048 (0.073) cohabiting -0.039 (0.079) Education (ref.: low) middle 0.037 (0.044) high 0.005 (0.049) Employment Status (ref.: part-time) full-time 0.046 (0.049) not employed 0.128** (0.047) Observations 1009 1009 Pseudo R2MF 0.008 0.025 Standard errors in parentheses. * p < 0.05, ** p < 0.01, *** p < 0.001 Source: GGSS 1982, own calculations • Leonie Kleinschrot170 three working full-time. This proportion is surprising, since the GDR had already existed for 35 years at the time of the survey, and cohorts who grew up under East German state socialism were interviewed. Furthermore, theories about social policy being a normative anchor, about the consequences of exposure to work-related environments, as well as arguments on social multiplier effects arising from high maternal labour force participation (Pfau-Effi nger 2005; Kremer 2007; Gangl/Ziefl e 2015; Grunow/Veltkamp 2016), would have suggested a higher agreement with maternal full-time employment in the GDR. However, exposure to the double burden of paid and unpaid work that many mothers experienced, which was associated with time pressures and mental strain, may have promoted preferences for maternal non-employment (Adler/Brayfi eld 1997; Nickel 1998; Becker-Schmidt 2010; Adler et al. 2016). Similarly, experience with day-care centres for under 3-year-olds ran by the socialist state might have reinforced maternal non-employment preferences, and with it the desire to raise children in the family according to the family’s own values rather than to socialist ideas. Logistic regressions revealed that women in the GDR who had high preferences for work were substantially more likely to agree with full-time maternal employment with children under three years of age than those who rated work as not very important. This association could not be shown for men, which suggests that fulfi lling the socialist gender norm of being employed led to a high commitment to employment among women and led work to become identity- forming (Bielby/Bielby 1989). This is in line with what other researchers (Gysi/ Meyer 1993; Leitner et al. 2008) have already pointed out: employment was highly valued in the GDR, especially for emancipation. Thus, women not only worked for fi nancial, but also for personal reasons like self-affi rmation. This and the second fi nding from the study, that individual preferences for children were not associated with attitudes towards full-time maternal employment, indicate a lack of confl ict between work- and family-identities for mothers. Thus, under the GDR’s framework of high availability and societal acceptance of institutional childcare for children of any age, women were not forced to trade off work against children or vice versa. This is also mirrored in the descriptive fi nding that women (just like men) had very high preferences for both paid work and children. The study’s fi nding that the vast majority of respondents in the FRG had the opinion that children suffer if their mothers work supports the notion of social policy – in this case the conservative-corporatist policies supporting the male-breadwinner model and maternal childcare – simultaneously refl ecting societal gender norms, and functioning as a normative anchor (Gangl/Ziefl e 2015). That people in the FRG in the 1980s had more negative attitudes towards maternal employment than their neighbours in the GDR strengthens the post-reunifi cation evidence on East-West differences in gender norms (e.g. Pfau-Effi nger 2005; Bauernschuster/Rainer 2012; Ebner et al. 2020). Regression analyses showed that in the FRG, respondents who had high preferences for children were more likely to agree with maternal non- employment, but attitudes towards maternal employment did not depend on preferences for work. These fi ndings can be interpreted in line with preference theory (Hakim 2002) as well as identity formation theory (Bielby/Bielby 1989). Firstly, preference theory categorises women in the FRG as “home centred” and thus Gender Norms under Socialism and Capitalism • 171 expects them to prioritise family, which is also backed by the study’s fi nding that women in the FRG placed more importance on children and less importance on paid work than men (see also Beblo/Görges 2018). Secondly, identity formation theory (Bielby/Bielby 1989) argues that employment and child raising were confl icting identities for women in the FRG which provoked an either/or decision between either being childless and employed or having children and not being employed (at least as long as they could fi nancially afford not to be employed) (Braun et al. 1994; Becker-Schmidt 2010; Goldstein/Kreyenfeld 2011). Thus, if women in the FRG had a strong desire to work, they – due to the policy framework and societal gender norms – might have already ruled out becoming a mother, so that maternal employment was not something that was thought about. Taken together, the fi ndings for both former states suggest congruence between social policy and gender norms at the time, but under the totalitarian regime in the GDR there was greater heterogeneity in attitudes towards maternal employment than expected. This insight cannot be gained by solely observing maternal employment rates, and it confi rms previous social-historical research showing the dialectical relationship between the regime and society in the GDR (Wolle 1998; Lindenberger 2002; Bauerkämper 2005; Jarausch 2012). Moreover, it contributes to a more differentiated understanding of gender norms in a socialist East European state, and at the same time highlights the need to refl ect on the frequently applied Western perspective – which assumes very high attitudinal conformity – in researching former socialist regimes in Eastern Europe. The study is limited by, fi rstly, the fact that a direct comparison between the FRG and the GDR was not possible because the indicators for the (in-)dependent variables were measured using different scales and the items for the dependent variables were not identical in wording. Secondly, whereas the FRG sample was representative of the 18-37-year-old population, the sample in the GDR was recruited via employers in certain districts only and women were overrepresented. These data cannot claim to be representative of the GDR. Nevertheless, it is a unique source for historical attitudinal research in a socialist East European state. Thirdly, the “Women’s Study 1984” has to be treated with caution, as social science was strictly politically controlled. Nevertheless, it would be enlightening if further research could make use of the unexploited data potential of the ZIJ for more in- depth analyses of the GDR, especially as contemporary witnesses of the study processes can still be consulted. The Women’s Study 1984, for example, also measured respondents’ division of housework and childcare as well as the degree of agreement with gendered responsibilities. Future analyses of these aspects could provide a more comprehensive understanding of societal gender norms in the GDR, which were refl ected in both behaviour and attitudes. Acknowledgements This study was conducted within a project funded by the Deutsche Forschungs- gemeinschaft (DFG, German Research Foundation) – Project number 434947732. • Leonie Kleinschrot172 I appreciate the helpful comments by the guest editors, anonymous reviewers and Felix Berth which contributed to signifi cant improvements in the manuscript. References Adler, Marina A.; Brayfi eld, April 1997: Women’s Work Values in Unifi ed Germany: Regional Differences as Remnants of the Past. In: Work and Occupations 24,2: 245- 266. https://doi.org/10.1177/0730888497024002006 Adler, Marina A.; Lenz, Karl; Stöbel-Richter, Yve 2016: Germany. In: Adler, Marina A.; Lenz, Karl (Eds.): Father involvement in the early years: An international comparison of policy and practice. Bristol, UK: Policy Press: 61-96. Alwin, Duane F.; Scott, Jacqueline 1996: Attitude change: Its measurement and interpretation using longitudinal surveys. In: Taylor, Bridget; Thomson, Katarina (Eds.): Understanding Change in Social Attitudes. Aldershot: Dartmouth: 75-106. Arránz Becker, Oliver; Lois, Daniel; Nauck, Bernhard 2010: Differences in fertility patterns between East and West German women. Disentangling the roles of cultural background and of the transformation process. In: Comparative Population Studies 35,1: 7-34. https://doi.org/10.12765/CPoS-2010-02 Banaszak, Lee Ann 2006: The Gendering State and Citizens’ Attitudes toward Women’s Roles: State Policy, Employment, and Religion in Germany. In: Politics & Gender 2,1: 29-55. https://doi.org/10.1017/S1743923X06060016 Barth, Denise et al. 2020: Mothers in Eastern and Western Germany: Employment Rates and Attitudes Are Converging, Full-Time Employment is Not. In: DIW Weekly Report 10,38: 403-410. ht t ps://doi.org /10.18723 /diw_d w r:2020 - 3 8 -1 Bauerkämper, Arnd 2005: Die Sozialgeschichte der DDR. München: Oldenbourg. Bauernschuster, Stefan; Rainer, Helmut 2012: Political regimes and the family: how sex-role attitudes continue to differ in reunifi ed Germany. In: Journal of Population Economics 25,1: 5-27. https://doi.org/10.1007/s00148-011-0370-z Beblo, Miriam; Görges, Luise 2018: On the nature of nurture. The malleability of gender differences in work preferences. In: Journal of Economic Behavior & Organization 151: 19-41. https://doi.org/10.1016/j.jebo.2018.05.002 Becker-Schmidt, Regina 2010: Doppelte Vergesellschaftung von Frauen: Divergenzen und Brückenschläge zwischen Privat- und Erwerbsleben. In: Becker, Ruth; Kortendiek, Beate (Eds.): Handbuch Frauen- und Geschlechterforschung. Theorie, Methoden, Empirie. Geschlecht und Gesellschaft. Vol. 35. Wiesbaden: VS Verlag für Sozialwissenschaften: 65-74. https://doi.org/10.1007/978-3-322-99461-5_8 Berth, Felix 2019: Vom Argwohn zur Akzeptanz. Der Wandel der Einstellungen zu nicht-mütterlicher Kinderbetreuung in Deutschland. In: Diskurs Kindheits- und Jugendforschung 4: 446-459. https://doi.org/10.3224/diskurs.v14i0.02 Bertram, Hans 1997: Einleitung. In: Bertram, Hans (Ed.): Soziologie und Soziologen im Übergang. Beiträge zur Transformation der außeruniversitären soziologischen Forschung in Ostdeutschland. KSPW: Transformationsprozesse 23. Opladen: Leske und Budrich: 9-15. Bielby, Denise D. 1992: Commitment to work and family. In: Annual Review of Sociology 18,1: 281-302. https://doi.org/10.1146/annurev.so.18.080192.001433 Gender Norms under Socialism and Capitalism • 173 Bielby, William T.; Bielby, Denise D. 1989: Family Ties: Balancing Commitments to Work and Family in Dual Earner Households. In: American Sociological Review 54,5: 776- 789. https://doi.org/10.2307/2117753 Bolzendahl, Catherine I.; Myers, Daniel J. 2004: Feminist Attitudes and Support for Gender Equality. Opinion Change in Women and Men, 1974-1998. In: Social Forces 83,2: 759-789. https://doi.org/10.1353/sof.2005.0005 Braun, Michael; Scott, Jacqueline 2009: Changing public views of gender roles in seven nations, 1988-2002. In: Haller, Max; Jower, Roger; Smith, Tom W. (Eds.): The International Social Survey Programme 1984-2009. Charting the globe. London/New York: Routledge: 358-377. Braun, Michael; Scott, Jacqueline; Alwin,Duane F. 1994: Economic Necessity or Self- Actualization? Attitudes toward Women’s Labour-Force Participation in East and West Germany. In: European Sociological Review 10,1: 29-47. https://doi.org/10.1093/oxfordjournals.esr.a036314 Brislinger, Evelyn; Hausstein, Brigitte; Riedel, Eberhard 1998: Sozialwissenschaftliche Daten aus der DDR und den neuen Bundesländern – 1968 bis 1996. In: GESIS (Ed.): Materialien zur Erforschung der DDR-Gesellschaft: Quellen. Daten. Instrumente. Wiesbaden: VS Verlag für Sozialwissenschaften: 95-47. Brooks, Clem; Bolzendahl, Catherine 2004: The transformation of US gender role attitudes: cohort replacement, social-structural change, and ideological learning. In: Social Science Research 33,1: 106-133. https://doi.org/10.1016/S0049-089X(03)00041-3 Bundesminister für Jugend Familie, Frauen und Gesundheit (Ed.) 1986: Siebter Jugendbericht. Jugendhilfe und Familie – die Entwicklung familienunterstützender Leistungen der Jugendhilfe und ihre Perspektiven. Bundestags-Drucksache 10/6730. Duncan, Simon; Edwards, Rosalind 1997: Lone Mothers and Paid Work-Rational Economic Man or Gendered Moral Rationalities? In: Feminist Economics 3,2: 29-61. https://doi.org/10.1080/135457097338690 Ebner, Christian; Kühhirt, Michael; Lersch, Philipp 2020: Cohort Changes in the Level and Dispersion of Gender Ideology after German Reunifi cation: Results from a Natural Experiment. In: European Sociological Review 36,5: 814-828. https://doi.org/10.1093/esr/jcaa015 Friedrich, Walter 1991: Zur Einleitung: Fast 25 Jahre „Zentralinstitut für Jugendforschung“. In: Hennig, Werner; Friedrich, Walter (Eds.): Jugend in der DDR. Daten und Ergebnisse der Jugendforschung vor der Wende. Weinheim/München: Juventa Verlag: 11-26. Fritz, Wolfgang 2000: Kleines Lexikon der amtlichen Erwerbstätigenstatistik in der DDR. In: Historical Social Research Transition 2. https://doi.org/10.12759/hsr.trans.2.v01.2000 Gangl, Markus; Ziefl e, Andrea 2015: The making of a good woman: Extended parental leave entitlements and mothers’ work commitment in Germany. In: American Journal of Sociology 121,2: 511-563. https://doi.org/10.1086/682419 GENESIS online 2021: Total fertility rate (per woman): Germany, years, age groups of the women [https://www-genesis.destatis.de/genesis//online?operation=table&co de=12612-0009&bypass=true&levelindex=0&levelid=1639073184875, 18.05.2021]. Gerlach, Irene 2008: Wichtige Stationen bundesdeutscher Familienpolitik. In: bpb. Informationen zur politischen Bildung 301 [https://www.bpb.de/shop/zeitschriften/ izpb/8012/familie-und-familienpolitik/, 26.02.2021]. GESIS – Leibniz-Institut für Sozialwissenschaften 2002: ALLBUS/GGSS 1980-86 (Kumulierte Allgemeine Bevölkerungsumfrage der Sozialwissenschaften/Cumulated German General Social Survey 1980-86). https://doi.org/10.4232/1.1335 • Leonie Kleinschrot174 GESIS – Leibniz-Institut für Sozialwissenschaften 2021: Frequently Asked Questions about ALLBUS [https://www.gesis.org/allbus/allbus/allgemeine-informationen/faqs, 31.03.2021]. Goldstein, Joshua R.; Kreyenfeld, Michaela 2011: Has East Germany overtaken West Germany? Recent trends in order-specifi c fertility. In: Population and Development Review 37,3: 453-472. https://doi.org/10.1111/j.1728-4457.2011.00430.x Görges, Luise 2021: Der Einfl uss institutioneller Rahmenbedingungen auf die individuelle Einschätzung der Wichtigkeit von Familie. In: Sachverständigenkommission des Neunten Familienberichts (Ed.): Eltern sein in Deutschland. Materialien zum Neunten Familienbericht. München: DJI Verlag Deutsches Jugendinstitut: 205-235. https://doi.org/10.36189/DJI232021 Griese, Hartmut 1991: Ausblick: Das ZIJ und seine Forschung in der DDR. In: Friedrich, Walter; Griese, Hartmut (Eds.): Jugend und Jugendforschung in der DDR: Gesellschaftspolitische Situationen, Sozialisation und Mentalitätsentwicklung in den achtziger Jahren. Opladen: Leske + Budrich: 211-224. Grunow, Daniela; Veltkamp, Gerlieke 2016: Institutions as reference points for parents- to-be in European societies: a theoretical and analytical framework. In: Grunow, Daniela; Evertsson, Marie (Eds.): Couples’ transitions to parenthood: Analysing gender and work in Europe. Cheltenham England/Northampton, MA: Edward Elgar Publishing: 3-33. Gysi, Jutta; Meyer, Dagmar 1993: Leitbild: berufstätige Mutter. DDR-Frauen in Familie, Partnerschaft und Ehe. In: Helwig, Gisela; Nickel, Hildegard Maria (Eds.): Frauen in Deutschland 1945-1992. Berlin: Akademie Verlag: 139-65. Hadjar, Andreas; Berger, Jörg 2010: Persistent Educational Inequalities in West Germany, East Germany, and Switzerland: A Cohort Analysis with Respect to Social Origin and Gender. In: Zeitschrift für Soziologie 39,3: 182-201. https://doi.org/10.1515/zfsoz-2010-0302 Hakim, Catherine 2002: Lifestyle preferences as determinants of women’s differentiated labor market careers. In: Work and Occupations 29,4: 428-459. https://doi.org/10.1177/0730888402029004003 Helwig, Gisela 1993: Einleitung. In: Helwig, Gisela; Nickel, Hildegard Maria (Eds.): Frauen in Deutschland 1945-1992. Berlin: Akademie Verlag: 9-22. Jarausch, Konrad H. 2012: Realer Sozialismus als Fürsorgediktatur: zur begriffl ichen Einordnung der DDR. In: Historical Social Research, Supplement 24: 249-272. Kolinsky, Eva 1989: Women in West Germany. Life, work and politics. German studies series. Oxford: Berg. Kremer, Monique 2007: How welfare states care. Culture, gender and parenting in Europe. Amsterdam: Amsterdam Univ. Press. Lee, Kristen S.; Alwin, Duane F.; Tufi s, Paula A. 2007: Beliefs about Women’s Labour in the Reunifi ed Germany, 1991-2004. In: European Sociological Review 23,4: 487-503. https://dx.doi.org/10.1093/esr/jcm015 Leitner, Sigrid; Ostner, Ilona; Schmitt, Christoph 2008: Family Policies in Germany. In: Ostner, Ilona; Schmitt, Christoph (Eds.): Family Policies in the Context of Family Change. Wiesbaden: VS Verlag für Sozialwissenschaften: 175-202. Lindenberger, Thomas 2002: Herrschaft und Eigen-Sinn in der Diktatur. Das Alltagsleben der DDR und sein Platz in der Erinnerungskultur des vereinten Deutschlands. In: Aus Politik und Zeitgeschichte B 40/2000. Gender Norms under Socialism and Capitalism • 175 McDonald, Peter 2000: Gender Equity, Social Institutions and the Future of Fertility. In: Journal of Population Research 17,1: 1-16. https://doi.org/10.1007/BF03029445 Mood, Carina 2010: Logistic Regression: Why We Cannot Do What We Think We Can Do, and What We Can Do About It. In: European Sociological Review 26,1: 67-82. https://doi.org/10.1093/esr/jcp006 Nickel, Hildegard Maria 1998: Women and women’s policies in east and west Germany, 1945-1990. In: Kolinsky, Eva (Ed.): Social transformation and the family in post- Communist Germany. Basingstoke/Hampshire: Macmillan: 23-36. Oechsle, Mechtild 1998: Ungelöste Widersprüche: Leitbilder für die Lebensführung junger Frauen. In: Oechsle, Mechtild; Geissler, Birgit (Eds.): Die Ungleiche Gleichheit. Junge Frauen und der Wandel im Geschlechterverhältnis. Wiesbaden: VS Verlag für Sozialwissenschaften: 185-200. Ostner, Ilona 1993: Slow motion: Women, work and the family in Germany. In: Lewis, Jane (Ed.): Women and social policies in Europe. Work, family and the state. Aldershot: Elgar: 92-115. Panayotova, Evelina; Brayfi eld, April 1997: National context and gender ideology: Attitudes toward women’s employment in Hungary and the United States. In: Gender & Society 11,5: 627-655. https://doi.org/10.1177/089124397011005006 Pfau-Effi nger, Birgit 2004: Development of culture, welfare states and women’s employment in Europe. Aldershot: Ashgate. Pfau-Effi nger, Birgit 2005: Culture and Welfare State Policies: Refl ections on a Complex Interrelation. In: Journal of Social Policy 34,1: 3-20. https://doi.org/10.1017/S0047279404008232 Pfau-Effi nger, Birgit; Smidt, Maike 2011: Differences in women’s employment patterns and family policies: eastern and western Germany. In: Community, Work & Family 14,2: 217-232. https://doi.org/10.1080/13668803.2011.571401 Raven, Judith et al. 2011: An institutional embeddedness of welfare opinions? The link between public opinion and social policy in the Netherlands (1970-2004). In: Journal of Social Policy 40,2: 369-386. http://dx.doi.org/10.1017/S0047279410000577 Rosenfeld, Rachel A.; Trappe, Heike; Gornick, Janet C. 2004: Gender and Work in Germany. Before and After Reunifi cation. In: Annual Review of Sociology 30,1: 103- 124. https://doi.org/10.1146/annurev.soc.30.012703.110531 Schmidt, Hans-Dieter 1996: Erziehungsbedingungen in der DDR. Offi zielle Programme, individuelle Praxis und die Rolle der Pädagogischen Psychologie und Entwicklungs- psychologie. In: Trommsdorff, Gisela (Eds.): Sozialisation und Entwicklung von Kin- dern vor und nach der Vereinigung. Beiträge zu den Berichten zum sozialen und politi- schen Wandel in Ostdeutschland 4,1. Opladen: Leske + Budrich: 15-171. Schneider, Silke L. 2016: The Conceptualisation, Measurement, and Coding of Education in German and Cross-National Surveys. GESIS Survey Guidelines. Mannheim: GESIS. Schütze, Yvonne 1986: Die gute Mutter. Zur Geschichte des normativen Musters „Mutterliebe“. Bielefeld: Kleine. Sjöberg, Ola 2004: The Role of Family Policy Institutions in Explaining Gender-Role Attitudes: A Comparative Multilevel Analysis of Thirteen Industrialized Countries. In: Journal of European Social Policy 14,2: 107-123. https://doi.org/10.1177/0958928704042003 Staatliche Zentralverwaltung für Statistik (Ed.) 1986: Statistisches Jahrbuch 1985 der Deutschen Demokratischen Republik. Berlin: Staatsverlag der Deutschen Demokratischen Republik. • Leonie Kleinschrot176 Statistisches Bundesamt (Ed.) 1984: Statistisches Jahrbuch 1983 für die Bundesrepublik Deutschland. Stuttgart/Mainz: W. Kohlhammer. Svallfors, Stefan 2010: Policy feedback, generational replacement, and attitudes to state intervention: Eastern and Western Germany, 1990-2006. In: European Political Science Review 2,01: 119-135. https://doi.org/10.1017/S1755773909990257 Szydlik, Marc 1994: Incomes in Planned and a Market Economy: The Case of the German Democratic Republic and the ‘Former’ Federal Republic of Germany. In: European Sociological Review 10,3: 199-217. https://doi.org/10.1093/oxfordjournals.esr.a036333 Terwey, Michael; Baumann, Horst 2013: ALLBUS/GGSS: German General Social Survey – Cumulation 1980-2010. GESIS-Variable Reports. Trappe, Heike 1996: Work and Family in Women’s Lives in the German Democratic Republic. In: Work and Occupations 23,4: 354-377. https://doi.org/10.1177/0730888496023004003 Trappe, Heike; Pollmann-Schult, Matthias; Schmitt, Christian 2015: The Rise and Decline of the Male Breadwinner Model. Institutional Underpinnings and Future Expectations. In: European Sociological Review 31,2: 230-242. https://doi.org/10.1093/esr/jcv015 Weber, Hermann 2011: Die DDR 1945-1990. Oldenbourg Grundriss der Geschichte 20. München: Oldenbourg. Winkler, Gunnar (Ed.) 1990: Frauenreport ‚90. Im Auftrag der Beauftragten des Ministerrates für die Gleichstellung von Frauen und Männern. Berlin: Verlag Die Wirtschaft. Wolle, Stefan 1998: Die heile Welt der Diktatur. Alltag und Herrschaft in der DDR 1971- 1989. Berlin: Links. Wooldridge, Jeffrey M. 2010: Econometric analysis of cross section and panel data: MIT press. Zentralinstitut für Jugendforschung, Leipzig 1984: Women‘s Study 1984: GESIS Data Archive, Cologne. ZA6138 Data fi le Version 1.0.0. https://doi.org/10.4232/1.6138. Date of submission: 14.12.2021 Date of acceptance: 17.02.2023 Leonie Kleinschrot (). Deutsches Jugendinstitut München (DJI), Abteilung Kinder und Kinderbetreuung. Munich, Germany. E-mail: kleinschrot@dji.de URL: https://www.dji.de/en/about-us/employees/detailview/mitarbeiter/leonie- kleinschrot.html Gender Norms under Socialism and Capitalism • 177 A p p e n d ix T a b . A 1: C la ss ifi c a ti o n o f e d u ca ti o n C la ss ifi c a ti o n F R G G D R C A S M IN fo r a n a ly si s E d u ca ti o n V o ca ti o n a l E d u ca ti o n V o ca ti o n a l e q u iv a le n t Tr a in in g Tr a in in g L o w e d u ca ti o n N o s ch o o l- le av in g W it h /w it h o u t tr a in in g / U p t o 7 th g ra d e W it h /w it h o u t 1a , 1b , 1c q u a lifi c a ti o n in te rn sh ip (s e m i- )s ki lle d w o rk e r q u a lifi c a ti o n L o w e r se co n d a ry W it h o u t tr a in in g / 8 th /9 th g ra d e s ch o o l W it h o u t tr a in in g o r sc h o o l ( V o lk ss ch u le , in te rn sh ip le av in g sk ill e d w o rk e r H a u p ts ch u le ) q u a lifi c a ti o n 8 th g ra d e W it h a p p re n ti ce sh ip M id d le e d u ca ti o n L o w e r se co n d a ry W it h a p p re n ti ce sh ip 9 th g ra d e W it h a p p re n ti ce sh ip 2 a, 2 b sc h o o l S e co n d a ry s ch o o l W it h /w it h o u t 10 th g ra d e ( P O S W it h /w it h o u t (s e m i- ) (M it tl e re R e if e , a p p re n ti ce sh ip g ra d u a ti o n 1 ) o r sk ill e d w o rk e r R e a ls ch u le ) 11 th g ra d e q u a lifi c a ti o n H ig h e d u ca ti o n A b it u r W it h /w it h o u t 12 th g ra d e W it h o r w it h o u t 2 c, 3 a, 3 b a p p re n ti ce sh ip E O S 2 /A b it u r a p p re n ti ce sh ip U n iv e rs it y o f a p p lie d Te ch n ic a l c o lle g e sc ie n ce s M a st e r M a st e r U n iv e rs it y U n iv e rs it y 1 P O S = P o ly te ch n is ch e O b e rs ch u le . T h e te n -y e ar “ P o ly te ch n is ch e O b e rs ch u le ” w it h c o m p u ls o ry v o ca ti o n al le ss o n s w as in tr o d u ce d in t h e G D R in 1 9 5 9 a n d r e p la ce d t h e p re v io u s e ig h t- y e ar s ch o o ls ( S ch m id t 1 9 9 6 : 8 2 ). 2 E O S = E rw e it e rt e O b e rs ch u le . S o u rc e : O w n il lu st ra ti o n b as e d o n H a d ja r/ B e rg e r (2 0 1 0 ) • Leonie Kleinschrot178 Tab. A2: Agreement with the norm of full-time maternal employment, GDR 1984, average marginal effects, separate models Model (1a) (2a) (3a) (4a) Preferences for Work (ref.: low) Children (ref.: low) high 0.123** 0.116** 0.063 0.076 (0.098) (0.038) (0.039) (0.040) Sex (ref.: male) female -0.025 -0.033 (0.032) (0.032) Age 0.001 0.002 (0.004) (0.004) Family Status (ref.: single) married -0.008 -0.030 (0.042) (0.043) LAT -0.033 -0.042 (0.050) (0.050) cohabiting 0.016 0.010 (0.051) (0.051) Education (ref.: low) middle -0.002 -0.006 (0.046) (0.046) high 0.012 0.010 (0.055) (0.055) Employment Status (ref.: part-time) full-time 0.200*** 0.203*** (0.052) (0.052) Observations 1300 1300 1300 1300 Pseudo R2MF 0.006 0.018 0.002 0.014 Note: Standard errors in parentheses. * p < 0.05, ** p < 0.01, *** p < 0.001 Source: Women’s Study 1984, own calculations Gender Norms under Socialism and Capitalism • 179 Tab. A3: Agreement with the norm of maternal non-employment, FRG 1982, average marginal effects, separate models Model (1b) (2b) (3b) (4b) Preferences for Work (ref.: low) Children (ref.: low) high -0.024 -0.009 0.103 0.113* (0.038) (0.041) (0.043) (0.046) Sex (ref.: male) female -0.039 -0.041 (0.027) (0.027) Age 0.001 0.002 (0.003) (0.003) Family Status (ref.: single) married 0.019 -0.003 (0.030) (0.030) LAT 0.064 0.051 (0.073) (0.073) cohabiting -0.015 -0.037 (0.077) (0.079) Education (ref.: low) middle 0.035 0.035 (0.044) (0.043) high 0.003 0.002 (0.049) (0.048) Employment Status (ref.: part-time) not employed 0.127** 0.130** (0.050) (0.047) full-time 0.044 0.046 (0.050) (0.050) Observations 1009 1009 1009 1009 Pseudo R2MF 0.0004 0.017 0.007 0.024 Standard errors in parentheses. * p < 0.05, ** p < 0.01, *** p < 0.001 Source: GGSS 1982, own calculations • Leonie Kleinschrot180 Fig. A1: Average marginal effects of preferences for work with 95% CIs, GDR 1984 Source: Source: Women’s Study 1984, own calculations -0.2 0 0.2 male female Sex Effects on Pr(Agreement with Norm) Gender Norms under Socialism and Capitalism • 181 Fig. A2: Average marginal effects of preferences for children with 95% CIs, FRG 1982 Source: Source: GGSS 1982, own calculations 0 0.1 0.2 0.3 male female Sex Effects on Pr(Agreement with Norm) 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)