Predicting the Timing of Social Transitions from Personal, Social and Socio-Economic Resources of German Adolescents Predicting the Timing of Social Transitions from Personal, Social and Socio-Economic Resources of German Adolescents Jana Gläßer, Wolfgang Lauterbach, Fred Berger Abstract: Social transitions are characterized by an increased heterogeneity in Western societies. Following the life course perspective, individual agency becomes central in shaping one’s life course. This article examines social transitions of ado- lescents using individual resource theory to explain differences of the timing of fi ve transitions in partnership and family formation: the fi rst sexual experience, the fi rst intimate relationship, the fi rst cohabitation, the fi rst marriage, and the birth of the fi rst child. Since little is so far known about how individual characteristics interact and infl uence the social transition to adulthood, we focus on the varying impacts of personal, social and socio-economic resources across the social life course. We use longitudinal data from the German LifE-Study, which focuses on the birth cohort of individuals born between 1965 and 1967. Using event history analysis, we fi nd that the timing of the fi rst sexual experience and fi rst partnership transitions are mainly infl uenced by personal and social ressources, whereas socio-economic resources offer better explanations for the timing of entering marriage and parenthood. Most striking are the different explanatory models for women and men. Keywords: Social transitions · Partnership and family formation · Resources · Life course · Youth 1 Introduction On the way to adulthood most people experience social transitions: they have sex- ual experiences, form romantic relationships, begin living with a partner in a shared household, marry and start a family of their own (Meulemann 1999). In Western societies, the sequence and timing of these transitions is characterized by an in- creased heterogeneity (Dommermuth 2008; Ross et al. 2009). Even, there is consid- erable variance regarding the timing of these transitions: Some people experience transitions early, others rather late. For example, the beginning of the fi rst romantic Comparative Population Studies Vol. 43 (2018): 157-186 (Date of release: 12.12.2018) Federal Institute for Population Research 2018 URL: www.comparativepopulationstudies.de DOI: 10.12765/CPoS-2018-11en URN: urn:nbn:de:bib-cpos-2018-11en1 • Jana Gläßer, Wolfgang Lauterbach, Fred Berger158 relationship varies from age 17 to age 23 for individuals born in 1965. The young- est 25 percent of women born between 1965 and 1969 were younger than 22 years old when they married, the oldest 25 percent entered marriage being older than 30 years of age. Among men of the same birth cohort, the interquartile range is even 30 years (Scherger 2007: 161). There is also a signifi cant range regarding the length of time between the fi rst sexual contact and the birth of the fi rst child. While most young adults experience their fi rst sexual contacts before the age of 20 (Bode/Heßling 2015), the average ages at marriage and at birth of the fi rst child do not occur until they reach their fourth decade (marriage: women 31 years, men: 34 years; birth of fi rst child: wom- en: 30 years (Federal Statistical Offi ce of Germany 2016b: 36, 54)). Thus, social transitions on the way to middle adulthood stretch over a long period of time over the life course (Mayer 2009). At the same time, the transitions are very age-heterog- enous: Some pass the mentioned transitions quite quickly at an early age, others to the contrary much more slowly at an older age. What we also fi nd is that men and women experience these transitions at different points in time, women being more quickly than men. Transitions that occur during the life course are determined by institutional con- texts (the educational system, labour market or legal requirements) as well as indi- vidual resources (Konietzka 2010). Social norms, however, have largely ceased to be effective in orienting and planning individual life stages (Beck 1999). Young adults themselves have become central in planning and organizing their transition to adult- hood, embedded in institutional contexts (Beck 1986; Mayer/Müller 2009). They are themselves responsible for beginning romantic partnerships, for the number of relationships over their life course, and whether and when to start a family. The individual is the “active agent”: “Individuals construct their own life course through choices and actions they take within the opportunities and constraints of history and social circumstance” (Elder et al. 2004). Thus, individual agency becomes cen- tral in shaping one’s life course and making decisions to pass through certain life stages. Adolescents and young adults decide, based on their own resources, when the time is “right” for these transitions (Neyer 1999). This explains the empirically “mixed age picture” with regard to social transitions. In this paper, we present a resource perspective of social transitions to middle adulthood in Germany for the birth cohort 1965-1967. The objective of this contribu- tion is to analyse the varying explanatory power of individual resources, which we differentiate between personal, social and socio-economic resources. A number of studies have focused on the latter type of resources and showed how partnership and family formation in Germany is infl uenced by, e.g., the educational attainment (e.g. Kreyenfeld/Konietzka 2008; Schaeper et al. 2013), labour force participation (Huinink 1989; Seyda 2003) and educational homogamy (Bauer/Jacob 2010; Bloss- feld/Timm 1997). However, little is known about the question of how individual char- acteristics interact and infl uence the social transition to adulthood. How do resourc- es infl uence sexual experiences, partnership formation, cohabitation, marriage and the birth of the fi rst child? Particularly the varying effects of personal, social and Predicting the Timing of Social Transitions ... • 159 socio-economic resources on these transitions are unknown. Does the infl uence of some factors change over time? And does the infl uence differ for women and men? To address these questions, we outline the changes of social transitions since the latter half of the twentieth century (Section 2.1.). We proceed with the differ- ences of the transitions analysed in this study (Section 2.2.). Next, we offer a model that represents the changing infl uences of individual resources on a sequence of fi ve social transitions in the second, third and fourth decade of life (Section 2.3.). Next, we describe the data, measurements and analytical procedures employed (Section 3). Finally, we present the results (Section 4) and discuss our central fi nd- ings (Section 5). 2 Theoretical considerations and state of research 2.1 Social transitions and their timing over the life course Becoming socially independent from the parental home and forming an own social identity as well as entering into relationships are major steps on the way to middle adulthood. On this way, adolescents and young adults are faced with decisions re- garding the timing of the fi rst sexual contact, the fi rst partnership, cohabitation or marriage. This process is understood as a “pathway through the age-differentiated life span, to social patterns in the timing, duration, spacing, and order of events […]” (Elder 1978: 21). This pathway is divided into individual points in time, that represent life stages marked by transitions (Elder et al. 2004). Transitions such as entering marriage or parenthood indicate a change from one status to another (e.g. from unmarried to married). Since the late 1960s, however, the transition to adulthood has changed in most Western societies, which also applies to Germany. Whereas in the 1950s and 1960s, leaving the parental home was directly linked with entering marriage, usually fol- lowed by the birth of the fi rst child, nowadays these events have been increasingly decoupled. For example, the share of nonmarital births has more than doubled in the last 25 years in Germany, from 15 percent of all births in 1990 to 35 percent in 2015. For fi rst births, the percentage is even higher reaching 44 percent (Federal Statistical Offi ce of Germany 2016a). Furthermore, Hayford et al. (2014) show for U.S. women throughout the second half of the 20th century that the link between marriage and parenthood has weakened: Women who married in the 1970s or later show more variation in the timing of birth of their fi rst child compared to women of earlier birth cohorts who had children more immediately after marriage. Not least because adolescents and young adults remain longer within the educational sys- tem, transitions to adulthood have decoupled and postponed. Thus, events which occurred within a short time span, are now considered over a longer stretch of time (Mayer 2009). Moreover, it becomes increasingly unclear which events defi ne the status of adulthood, since there is an increasing number of individuals who (perma- nently) remain unmarried and/or childless. In Germany, for instance, the percentage of permanently childless women amounts to around 21 percent in the 1967 cohort, • Jana Gläßer, Wolfgang Lauterbach, Fred Berger160 comparable to the cohort in our analysis. In the 1937 cohort, the share was around 11 percent (Federal Statistical Offi ce of Germany 2017). Therefore, we suggest no longer to speak of becoming an adult but rather only of status transitions that occur during the life course within the fi rst four decades (see also Konietzka 2010). 2.2 Differences of social transitions over the life course The social transitions examined here are: the fi rst sexual contact, the fi rst romantic relationship, the fi rst cohabitation, the fi rst marriage and the birth of the fi rst child. These transitions differ in terms of the investments, commitments and liabilities re- quired in friendships and partnerships. The fi rst sexual contact and the fi rst relation- ship are usually experienced in youth or young adulthood and are associated with rather low investments in a relationship. By marriage and especially by the birth of the fi rst child, partnerships become highly committed and institutionalized. An intermediate stage is cohabiting with a partner (Lauterbach 2006). Cohabita- tion has become more and more relevant, since the number of cohabitating un- ions considerably increased since the 1970s (1972: 137,000, 1996: 1,382,000, 2014: 2,920,000 (Federal Statistical Offi ce of Germany 2002; Krack-Roberg et al. 2016)). This living arrangement is especially popular among young adults between the ages of 18 and 35, and represents the beginning of partnership and family formation (Peuckert 2012). Cohabitation resembles marriage, since it involves an intimate and time-intensive relationship with a partner, a joint household and a merger of per- sonal economies (Thornton et al. 1995). One main difference between cohabitation and marriage can be seen in the absence of a formally recognized status (Nock 1995). Cohabitation is more binding than fi rst partnership experiences but does not have the same obligations and institutionalized status as a marriage or especially the birth of a child. Partnerships can break up without any legal consequences, whereas being married or having children makes partners legally obligated to one another – even after divorce. Moreover, unlike the fi rst sexual and partnership experiences, economic security is necessary for establishing a shared household and even more so marriage and parenthood. Even if normative regulations regarding the timing and sequencing of status passages have generally decreased, fi nancial independence from the paren- tal home or a certain economic status is still highly signifi cant especially for entering into marriage and starting a family. Following family economics, entering long-term relationships and starting a family is not desirable as long as individuals are in edu- cation (institution effect), because achieving a certain educational level requires a lot of time that cannot be invested in an intimate relationship or family. Furthermore, there are normative expectations that married men and women have to be econom- ically independent (Brüderl/Diekmann 1997; Huinink 2000). Therefore, “fi nishing one’s education (…) counts as one of the important prerequisites for entering into marriage and parenthood” (Blossfeld/Huinink 1991). For 58 percent of participants of the Population Policy Acceptance Study 2003 a low income was the reason for postponing marriage (BiB 2009). Several studies have shown the infl uence of the level of education, which is itself strongly connected with the timing of economic in- Predicting the Timing of Social Transitions ... • 161 dependence, on delays in marriage and family formations (Blossfeld/Huinink 1991; Dommermuth 2008; Kreyenfeld 2015; Kreyenfeld/Konietzka 2015; Schaeper et al. 2013). Thus, individuals want to achieve economic security before they establish long-term intimate relationships (Huinink 2000). Consequently, we expect that with increasing investments and commitments in relationships and thus an increasing instutionalization of relationships, the individual’s educational and economic situa- tion becomes more important for the timing of social transitions. 2.3 Resource model From a life course perspective, we argue, that different resources become impor- tant at different stages in the fi rst four decades: the personal, social and socio-eco- nomic resources (Diewald/Mayer 2009). Thus, the pathway to middle adulthood is a co-determined process of individual action, based on a framework of institutional settings and individual resources (Asendorpf/Neyer 2012). Resources formed and stabilised during childhood and youth serve as guiding scripts to master the stages to middle adulthood (Huinink/Feldhaus 2009). The resource model of life management assumes that the (social) transitions (t 1 - t 5 ) in the fi rst four decades of the life course, represented here by the fi ve transition markers, are infl uenced by early life conditions and resources. We argue that social transitions are infl uenced by personal, social and socio-economic resources formed in youth, and are embedded in institutional settings and historical opportunities. Figure 1 illustrates this assumption and provides examples for indicators of the dif- ferent resources which will be described in the next section. However, passing through the social transitions and reaching middle adulthood is different for male and female adolescents. As studies show, men experience vari- ous transitions later than women (Bode/Heßling 2015; Dommermuth 2008; Grün- heid 2017; Marbach/Tölke 1996; Müller et al. 1999), and they also have a higher risk of never experiencing specifi c transitions, since they more often remain permanent- ly single or childless (Schmitt/Winkelmann 2016; Sprenger 2014). Therefore, we as- sume that personal, social and socio-economic resources have different meanings for male and female young adults in terms of the fi ve social transitions under study. 2.4 Personal resources Personal resources include cognitive, emotional and social competences which characterize people’s individual agency (Raithel 2011), or rather their coping capac- ity (Hurrelmann/Quenzel 2012). Personal resources comprise both cognitive skills and the perceived capacity for self-regulation (emotion control, impulse control, as well as temper and personality) (Petermann/Schmidt 2006). Research fi ndings have proven the link between personal resources, including the personality factors neuroticism and extraversion, such as shyness, self-control or sociability, and partnership characteristics. For example, partnership status dif- fers by personality: Singles are often much shyer than persons who live in a steady partnership. People committed to a partner show higher levels of extraversion and • Jana Gläßer, Wolfgang Lauterbach, Fred Berger162 F ig . 1: R e so u rc e m o d e l o f lif e m an a g e m e n t S o u rc e : O w n p re se n ta ti o n , b as e d o n F e n d e t a l. 2 0 0 9 Predicting the Timing of Social Transitions ... • 163 lower levels of neuroticism than singles (Neyer 1999). Having a fi rst relationship is linked with decreases in neuroticism (Lehnart et al. 2010; Neyer/Lehnart 2007). Sociability is associated with an increased probability of fi nding a partner (Neyer/ Lehnart 2007), and high extraversion and low neuroticism are linked to a high num- ber of children (Jokela et al. 2011) and a probability of encountering the fi rst roman- tic experiences between the ages of 23 and 25 (but not at other ages analysed in the study) (Wagner et al. 2015). Additionally, Stavrova and Ehlebracht (2015) show the connection between extraversion and an increased probability of entering a romantic relationship. Furthermore, the infl uence of childhood and adolescent characteristics and re- sources has been revealed in longitudinal studies which also point to gender differ- ences: It has been found that men’s shy behaviour in childhood is related to delayed entry into marriage and fatherhood. In contrast, shy girls are more likely to adopt a traditional female role, but in this case shyness is not associated with the timing of marriage and parenthood (Caspi et al. 1988). Similarly, a German longitudinal study shows that personality types in childhood predict the timing of entering the fi rst stable romantic relationship for men but not for women. Men with personali- ties characterized as resilient enter into their fi rst romantic partnerships earlier than men described as “over-“ or “under-controlled” (Dennissen et al. 2008). The timing of the fi rst sexual relationship is also associated with personal resources. Individu- als characterized as “under-controlled” in childhood fi nd it hard to maintain social contacts, and experience their fi rst sexual contacts earlier than those characterized as “over-controlled” or “resilient”; gender differences were not tested (Atkins/Hart 2008). Another longitudinal study demonstrates that women with a low self-control in childhood tend to enter motherhood earlier (Kokko et al. 2009). A further important personal resource which becomes salient in youth is the self-perceived physical attractiveness. Adolescents develop a specifi c idea of their own appearance and attractiveness (in comparison to others (see Fend 2005)). With respect to partner selection and behaviour in social interactions, these concepts of one’s own appearance are especially relevant in Western cultures. This applies in particular to young women (Smolak 2004), who learn that they have to pay attention to their appearance, how to judge their body image and how it is judged by others (Brown/Gilligan 1994; Müller 2006). Girls are more worried about body fat and in- vest more in their body than boys and are more likely to act according to their body satisfaction (see Smolak 2004 for a review). Research has also shown that the body image is related to the perceived rela- tionship quality for both, men and women (Ambwani/Strauss 2007; van den Brink et al. 2018). This association is mediated by greater sexual satisfaction of those sat- isfi ed with their bodies (van den Brink et al. 2018). However, it is yet unknown how body image infl uences the timing of partnership transition markers. Overall, personal resources seem to infl uence social transitions from the fi rst sexual experience to the fi rst birth. Regarding gender differences, we fi nd no clear picture from the literature as we described above. But, due to gender-specifi c so- cialization processes which expect young girls more than young boys to control their emotional and behavioural reactions in social situations (Davis 1995; Weis et • Jana Gläßer, Wolfgang Lauterbach, Fred Berger164 al. 2013), and the fact that women put more importance on their body image (Smo- lak 2004), we expect personal resources to genereally be more important for fe- males than for males. 2.5 Social resources Theories on the explanatory power of resources designate networks in which ado- lescents are integrated as social resources that they can draw on. These are primar- ily relationships to parents, but also to peers inside and outside of school (Hurrel- mann 2002; Oerter/Dreher 2002). In this article, we focus on two indicators of social resources: the peer group integration and the experience of parental divorce. Particularly the integration into peer groups is a decisive factor infl uencing the transitions to middle adulthood. In peer groups, adolescents learn to form, to main- tain and also to terminate social relationships. The peer group is a place where adolescents test their ability to engage in important relationships (McCormick et al. 2011). Thus, the peer group serves as an early model for romantic relationships in emerging adulthood (Reitz et al. 2014). We therefore expect that a strong peer inte- gration – and tied with this a testing of one’s own ability to form and maintain rela- tionships – should speed up the transition to adulthood. Furthermore, the infl uence of peer groups in the cohort analysed in this paper (individuals born in the middle of the 1960s) should be stronger for young men than for young women, because girls are raised to be more emotionally attached to the parents and stay longer within the protected environment of the family (Naudascher 1978; White 1994; Goldscheider/ Goldscheider 1999) in contrast to boys of whom it is expected to be more emotion- ally independent and self-suffi cient earlier (Shulman/Seiffge-Krenke 1996; Becker- Stoll et al. 2000). A well-known fi nding is that children who experience their parents’ divorce have themselves a higher risk of divorce (e.g. Amato/Patterson 2016; Berger 2009; Diekmann/Engelhardt 1995; Diekmann/Schmidheiny 2013; Feldhaus/Heintz-Martin 2015). Therefore, it is plausible that experiencing parental divorce also has effects on the timing of other partnership events. According to the economic deprivation theory, a parental divorce infl uences the economic situation of the parents and pos- sibly leads to an earlier entry of adolescents into the labour market as well as to an earlier age at fi rst marriage, less commitment to intimate partners, and a higher prevalence of union dissolutions. In addition, socialization theory supposes that pa- rental divorce is connected to a reduced control behaviour of parents and hence leads to earlier sexual contacts and earlier partnership formation (Berger 2009; Diekmann/Engelhardt 1995). In sum, the theory proposes that a parental divorce is linked to an acceleration in the childrens’ own partnership and family formation pro- cess. As Raab (2017) shows for Germany, experiencing an alternative family struc- ture (not living with both biological parents) up to the age of 18 is associated with earlier cohabitation and parenthood. Regarding gender differences, previous fi nd- ings for Germany are rather ambivalent (Berger 2009; Diekmann/Engelhardt 1995; Engelhardt 1998; Schulz 2009). Predicting the Timing of Social Transitions ... • 165 2.6 Socio-economic resources Socio-economic factors also have a powerful infl uence on the transitions to mid- dle adulthood. Educational attainment, for example, infl uences the transitions from adolescence to adulthood, as especially one’s educational level is considered to be a deciding factor in the timing of marriage and the birth of the fi rst child. Accord- ing to economic theory, higher educated people tend to delay marriage and family formation into higher ages, the effects being stronger for women than for men. Highly educated women tend to delay marriage and motherhood, while for men a u-shaped effect is observed (Bauer/Jacob 2010; Diekmann 1990; Dommermuth 2008; Huinink/Kreyenfeld 2004; Lauterbach 2004). Women with high educational levels tend to gain less from marriage, since they are economically independent. In contrast, highly educated men have higher income prospects and are therefore attractive marriage partners. So far, there has been insuffi cient longitudinal examination of the infl uence of the education level on the timing of the fi rst cohabitation: Do people with a univer- sity degree who stay longer within the educational system delay cohabitation with a partner? Uncertainty in one’s life situation might have a negative effect on the timing of cohabitation. However, cohabitation requires less fi nancial commitment and has lower opportunity costs than marriage (Berrington/Diamond 2000). Thus, the effect of the level of education should be stronger for marriage than for cohabitation. Mül- ler et al. (1999) found gender-specifi c effects of the educational level on the timing of cohabitation and marriage. While highly educated men enter into marriage more quickly, highly educated women cohabit earlier. The most important socio-economic resource for adolescents and young adults is, however, economic independence. From a historical point of view, the underlying normative concept indicates that young men must become economically independ- ent before they can responsibly form a shared household, marry, and have their fi rst child (Huinink 1995). Historically, individual fi nancial independence is of less impor- tance for women. According to the traditional female life course pattern, economic independence from the parental home was achieved by establishing a household with a marriage partner, rather than by becoming gainfully employed. However, the old male breadwinner model has been shaken (Blossfeld/Drobnic 2001; Janssens 1997; Lauterbach 2006). High educational levels and gainful employment at the be- ginning of the working life, have increased women’s fi nancial independence. Today, fi nancial independence is not only expected for men but also for women before entering into long-term intimate relationships and starting a family (Huinink 2000). One can therefore conclude that the later young adults achieve fi nancial independ- ence, the later they start a joint household and enter into marriage and parenthood. In addition, socio-economic resources from the family of origin can also be seen as resources for adolescents. It has been shown that adolescents growing up in families with lower socio-economic backgrounds tend to enter into their fi rst re- lationships, cohabitation and marriage earlier than their counterparts from an ad- vantaged background (Berrington/Diamond 2000). Social background may directly affect the timing of these transitions through class-specifi c norms and perceptions • Jana Gläßer, Wolfgang Lauterbach, Fred Berger166 about the “right” timing of cohabitation or marriage. Indirectly, the social back- ground can have a mediating effect through one’s own educational qualifi cations (Berrington/Diamond 2000; Blossfeld/Huinink 1991). 2.7 Hypotheses In sum, theoretical considerations and previous fi ndings suggest that personal, so- cial and socio-economic resources do infl uence social transitions experienced by adolescents. Nevertheless a systematic, concurrent examination of different transi- tion markers and their prediction by personal, social and socio-economic resources is still missing. Thus, the varying explanatory power of these resources over the life course is unclear. Hence, our fi rst research question reads as follows: Question 1: How does the explanatory power of personal, social and socio-eco- nomic resources for different social transitions vary over the life course? As described above, the transitions studied here differ in terms of investments and commitments in partnerships. Initial sexual and romantic experiences are usu- ally experienced in different living situations than the fi rst cohabitation and espe- cially the fi rst marriage and the birth of the fi rst child, which all have different de- mands on individual resources. Thus, our fi rst hypothesis assumes the following: Hypothesis 1: Due to different investments and commitments in partnerships, personal and social resources should have a strong effect on early transitions and thus on the fi rst sexual and partnership experiences. In contrast, with increasing institutionalization of relationships (cohabitation, fi rst marriage and birth of the fi rst child) socio-economic resources should become more important. Additionally, our literature review showed that men and women experience dif- ferent transitions at different points in time and previous studies found that the importance of resources is gender-specifi c. Thus, the following research question and hypotheses were addressed: Question 2: Are there gender differences in the varying infl uences of personal, social and socio-economic resources? Hypothesis 2: The impact of personal resources should be stronger for women than for men. Hypothesis 3: The impact of social resources (peer group integration) should be stronger for men than for women. Hypothesis 4: Economic independence infl uences both, men and women. The later young adults achieve fi nancial independence, the later they will enter their fi rst cohabitating union, marriage and especially parenthood. Hypothesis 5: The higher the educational level, the later young adults will marry, especially in the case of highly educated women. Predicting the Timing of Social Transitions ... • 167 3 Method 3.1 Data and sample The analyses conducted to test the assumptions above are based on data from the German LifE Study (Pathways from Late Childhood to Adulthood). The study is one of a small number of studies worldwide, providing information on adolescents’ life courses from age 12 to the age of 45 (1979-2012). Thus, a range of more than 30 years is covered by the dataset. The LifE Study is a sample of men and women in Germany born between 1965 and 1967, which was fi rst surveyed from 1979 to 1983, and again in 2002 and 2012 (Lauterbach et al. 2016). Initially, the study was conducted as a school survey (adolescence study) com- prising 12-year-old adolescents who were interviewed annually until the age of 16. The study focused on adolescents growing up in both urban and rural environ- ments, as well as in different school contexts. Thus, an urban area of former “West” Germany (Frankfurt am Main) and two adjacent rural areas (Odenwald, Bergstraße) with different types of schools were chosen. The study was, therefore, conducted in a “quasi-experimental design”. The research focused on adolescents coping with developmental tasks and on the development of personality. Analysing these devel- opments required using a longitudinal design. The fi rst data collection was carried out in 1979. 1,830 children in the 6th grade, which is equivalent to the age of 12, were interviewed. The survey – conducted as a panel study – was retaken fi ve times in one-year intervals. The data collection was completed when the adolescents at- tended the 10th grade. Panel mortality and non-response were controlled by two parallel studies. Data from parents and teachers were also collected twice in 1980 and 1982. In 2002 and 2012, data of the former students were collected again, investigating the conditions for a productive life development in young and middle adulthood. Life course data are now available for 1,359 persons. External validation in 2012 us- ing two German representative datasets (the Socio-Economic Panel (SOEP) and the microcensus) proved that the LifE data are reliable, especially regarding family and partnership characteristics. However, the LifE data show a lower divorce rate for women and a higher number of families with two children (Lauterbach et al. 2016). An important component of the LifE Study are retrospectively sampled partner- ship biographies from the ages of 16 to 45. Only partnerships that lasted longer than six months were taken into account. Subjects were also asked whether their part- nerships led to a shared home and marriage. Thus, we have data on the partnership and family formation process over a period of almost thirty years. Data on personality characteristics as well as on socio-economic and cultural characteristics of the contexts of origin are also available. For the analyses, we draw on data from 1983, when subjects were 16 years old. When information was lacking, data from 1982 and 1981 were imputed. Only men and women with full information on the dependent as well as inde- pendent variables were included in our study. Thus, after restricting the sample to persons with full information on partnership and family transitions, the case number • Jana Gläßer, Wolfgang Lauterbach, Fred Berger168 reduced to 1,008 (74 percent). Dropping cases with incomplete data on childhood and youth resources further reduced the sample to 695 cases (51 percent),1 leav- ing 388 women and 307 men in the sample. Robustness checks showed that even though the sample includes more women than men (50 percent baseline, 56 per- cent analysis sample), less respondents who experienced parental divorce during childhood and adolescence (12 percent baseline sample, 9 percent analysis sample) and more individuals with a university degree (23 percent baseline sample, 26 per- cent analysis sample), there exist no systematic differences between the baseline sample and the analysis sample. 3.2 Measures Dependent variables Five events were investigated as factors infl uencing the social transitions: fi rst sex- ual contact, age of fi rst partnership, fi rst cohabitation, fi rst marriage and birth of the fi rst child. All variables are measured in years. Information on the partnership and family history were collected in 2002 and 2012 retrospectively from age 16 to age 45. If events had not taken place until the age of 45, the respective case was treated as right censored. Information about the fi rst sexual contact was only collected in 2002, when subjects were approximately 35 years of age. Therefore, analyses on this topic are limited up to this age. A Personal resources We used three indicators to measure personal resources: emotional stability, out- goingness and body image (A in Fig. 1). The scale emotional stability is comparable to the Big Five factor of neuroticism and asks how individuals deal with their feelings and emotions, which is an important aspect of neuroticism. The scale was opera- tionalized by eight items from the youth study2 (e.g. “I sometimes get upset about every little thing”). The possible responses are “true” vs. “not true”. High values of the emotional stability scale indicate a high emotional stability. The internal consist- ency of the items was satisfying (α=0.79). Second, we used the scale outgoingness, which is comparable to the Big Five characteristic of extraversion. This scale was also operationalized by eight items (e.g. “I enjoy making new acquaintances”). The variables are dichotomous as well (“true” vs. “not true”). High values of the outgoingness scale indicate a high interest in establishing and maintaining contacts. Cronbach’s alpha for this scale was 0.65. 1 Even though some missing information from age 16 could be replaced by information from the two previous waves, the share of missing values among the independent variables was still 8 to 15 percent, meaning that of the 1,008 individuals with complete social life course information another 313 cases had to be dropped by listwise deletion. 2 All items are presented in the Appendix. For a detailed documentation of all scales and items of the youth study, see Fend/Prester (1986). Predicting the Timing of Social Transitions ... • 169 The third personal resource used is body image which refers to a person’s gen- eralized perception of his or her physical attractiveness. A person’s own perception and evaluation of his or her appearance is signifi cant for this study because it is as- sociated in Western culture with greater social success and especially with better chances on the “partner market” (Fend 2005: 239). The scale used here includes six items (e.g. “Compared with others, I really look very good.”), coded the same way as emotional stability. High values point to a positive body image. Cronbach’s alpha for this scale was 0.71. B Social resources In order to measure social resources, we used two indicators: integration in a peer group and the experience of parental divorce during childhood or youth (B in Fig. 1). The scale peer group integration consists of eight items (e.g. “Do you belong to a group of youths who meet regularly or often and have a sense of belonging together? We do not mean a club or association”) and shows the extent to which pupils were integrated into networks of social relationships with peers. Cronbach’s alpha was 0.66. The experience of parental divorce in childhood and youth is used as a second indicator of social resources. It is assumed that experiencing a parental divorce leads to an earlier acquisition of adult roles and an acceleration of the timing of so- cial transitons. The dummy-coded variable indicates whether a parental divorce or separation had been experienced up to the age of 16 or not. C Socio-economic resources In this study, the highest achieved individual level of education was included as socio-economic resource. Due to the small number of respondents with a low ed- ucational level, low and medium qualifi cations had to be combined into one cat- egory. Two different variables were included in our analyses. Since the events fi rst sexual contact and fi rst partnership were usually experienced from ages 16 to 20, the highest educational attainment for these events was indicated by the type of school attended (lower secondary school track (Hauptschule and Realschule) (0) vs. higher secondary school track (Gymnasium) (1)).3 For the other three further events (cohabitation, marriage, fi rst child), the highest individual level of education was indicated by a variable containing the two categories “vocational training” (0) and “higher educational or technical college degree” (1). In the LifE Study, low educa- tional levels such as “no vocational training” were under-represented (Lauterbach et al. 2016). In order not to violate the Cox model proportionality assumption (see Sec- tion 3.3), pupils from Hauptschule and Realschule were combined into one group, and pupils with no vocational qualifi cation (N=14) were not included at all in the 3 Depending on the level of courses attended, pupils from integrated comprehensive schools (in- tegrierte Gesamtschule) were assigned to the three different types of schools. This procedure is appropriate as it appears that pupils largely remain at the level of their courses. Change occurs above all with pupils assigned to the Realschule. • Jana Gläßer, Wolfgang Lauterbach, Fred Berger170 analysis. Because the level of education violates the proportional hazard assump- tion (see Section 3.3), this variable was included as time-varying variable. In 2002 respondents were asked when they fi rst earned enough money to fend for themselves. This variable was included in the analyses as timing of economic independence (coded as a time-dependent variable). The study focuses on adolescents’ own resources. To control for effects of so- cial origin, we included the socio-economic background of the family of origin as robustness check in the regression models. Therefore, we used the HISEI (“High- est International Socio-Economic Index of Occupational Status”) as a measure for the parental socio-economic status. The HISEI has a metric range from 16 (janitor, cleaning personnel) to 90 (medical doctor, judge). For the analyses, the HISEI was coded into three categories (16-40=low, 41-65=middle, 66-90=high). 3.3 Methods of analysis To study the timing of the fi ve partnership status transitions, event history analysis was applied. On a descriptive level, survival curves are presented based on the Kaplan-Meier estimator. To measure the infl uence of personal, social and socio-economic resources on the timing of different status changes, Cox models were estimated. Effects in the Cox models can be read as ß-coeffi cients. The hazard ratio (hz) represents the an- tilogarithm of the ß-coeffi cient (hz=exp(ß)) and (hz-1)*100 can be interpreted as the percentage effect of a covariate on the occurrence rate of a respective event (Weick 1993: 97). A hazard ratio of 1.50 can thus be interpreted as an increase in the prob- ability of the occurrence by 50 percent. We can thus speak of an acceleration of the event. In contrast, a hazard ratio of 0.75 signifi es a reduction in the probability of the occurrence of a specifi c event by 25 percent ((0.75-1)*100). Thus, one can speak of a deceleration of the transition. Variables that do not change their value over time (e.g. sex, social background) are considered as time-independent variables. Covariates that change their values over time are time-dependent variables. Thus, the effects of these variables are not constant, but rather unfold (increase or decrease) with the appearance of a specifi c event. In the study, the commencement of economic independence is modelled as a time-dependent covariate. This is done by episode splitting, whereby the data are transferred to a person-period format. The observation period is divided into an- nual intervals from the fi rst occurrence4 of the event until the end of the period of observation in 2012. The Cox model assumes that the hazard function for different levels of a co- variate is proportional over time, which means that effect sizes of covariates are 4 For the fi rst sexual experience, the age of 12 was chosen to exclude involuntary sexual experi- ences. The age of the fi rst partnership was set at 16 years based on questionnaire items. The same holds for age at fi rst cohabitation. The possible start of the fi rst marriage was set at 18 years (legal marriage age), fi rst birth at 12 years. Predicting the Timing of Social Transitions ... • 171 constant over time. This proportionality assumption was tested both graphically as well as by using the zero slope test developed by Therneau and Grambsch (2000). If the proportionality assumption is violated, this can be corrected by introducing interaction terms of the relevant variables with time (Blossfeld 2010: 1005; Windzio 2013: 112). In the following analyses, the effect of educational attainment changes over time and thus is modelled as time-varying variable. To correct for violation of the proportionality assumption, we included the main effect as well as an interac- tion term in the model. For example, a main effect of 0.05 means that the effect was negative immediately after the beginning of the observation: Highly educated peo- ple severely delay the event. However, a coeffi cient of the interaction term of 1.10 would suggest that the two educational levels under consideration converge over time by a factor of 1.10 each year, thus 10 percent a year. Modelling not only time-independent covariates but also time-dependent vari- ables and time-varying effects is a special advantage of the Cox model. A further ad- vantage of event history analysis is that so-called “right censored cases” (subjects who have not yet experienced the respective event by age 45) can be included in the analyses as well (Blossfeld 2010). Based on suspected differences between women and men, separate models are estimated. In addition, all scales that display random metrics were z-transformed in order to compare effect sizes in the Cox regressions 4 Findings In a fi rst step, we present descriptive fi ndings, followed by Cox regression models to estimate the impact of time-independent and time-varying resources on the tim- ing of transitions. The descriptive results in Table 1 show for each status change that more men than women had not yet experienced the respective event by age 35 or 45 respec- tively. While the share of censored cases for fi rst sexual contact is less than 2 per- cent, it increases with subsequent events. In total, 2.3 percent of women did not have a partnership by the age of 45. For men, this share is nearly the same with 2.9 percent. The shares of unmarried women and men are also similar (17.0 per- cent/18.2 percent). Likewise, up to the age of 45, slightly more men (22.2 percent) than women (19.6 percent) remained childless, which is comparable with represent- ative data for Germany (reported in Section 2.1.) Examining the average age at the events in Table 1, it is apparent that they oc- cur earlier among women than among men. The t-tests show that these differences are all statistically signifi cant, even though the difference in the age of fi rst sexual contact between men and women is rather low. Differences in timing can be observed in the age differences between men and women as well as in the age range from 16 to 45 years across almost the entire ob- servation period. Figure 2 graphically shows the transition rates for the fi ve events separately for men and women based on Kaplan-Meier estimators (for better read- ability, only from ages 16 to 35). The survival distributions of all events differ signifi - cantly between men and women (see log-rank test in Table 1). • Jana Gläßer, Wolfgang Lauterbach, Fred Berger172 Until the age of 18, the proportion of young women who experienced their fi rst relationship is slightly higher than the proportion of those who had their fi rst sexual intercourse. Hence, especially young women seem to make their fi rst experiences within a partnership. After the age of 18, the situation is reversed and the proportion of women who had their fi rst sexual contact is higher than the proportion of those who ever had a partnership. On average, the fi rst cohabitation takes place later. More than three-quarters of women had lived together with a partner at the age of 25. Up to the age of 35, 6 percent of women had never entered a cohabitation. These three transitions (fi rst sexual experience, fi rst partnership, fi rst cohabita- tion) are mainly experienced during late adolescence and young adulthood. The situation is different, however, in the case of the fi rst marriage and the birth of the fi rst child. Not surprisingly, these events mainly occur towards the end of the third decade of life. On average, the curves show a traditional pattern: Women fi rst get married and then have a child. In contrast to women, the proportion of men who had their fi rst sexual experi- ences is always higher than the proportion of men who had a partnership that lasted six months or longer. Furthermore, for men, moving in with a partner for the fi rst time takes place later in life. 60 percent had lived together with a partner at the age of 25. Similar to women, men also marry before they have children. Altogether, women seem to experience the investigated transitions earlier than men. The reason can be seen in the earlier physical maturation of girls who thus enter into relationships more quickly (Wendt 2010). An additional explanation for Tab. 1: First sexual contact, partnership, cohabitation, marriage, and fi rst child. Average age (and standard deviation) and shares of men and women without experiences Average Age No experience, Log-rank in % test Transition Women Men t (df) N Women Men Chi² First sexual 18.2 18.7 -2.0* (383, 301) 1.3 2.0 4.6* experience (2.8) (3.5) (682) (5) (6) First partnership 18.6 20.2 -5.1*** (379, 298) 2.3 2.9 21.9*** (3.7) (4.7) (675) (9) (9) First cohabitation 21.1 23.9 -7.5*** (364, 288) 6.2 6.2 22.0*** (4.1) (5.3) (650) (24) (19) First marriage 27.2 29.2 -4.8*** (322, 251) 17.0 18.2 6.5* (5.0) (5.2) (571) (66) (56) First birth of child 28.8 30.9 -5.3*** (312, 239) 19.6 22.2 8.5*** (4.7) (4.8) (549) (76) (68) Level of signifi cance: *** p<.001, ** p<.01, * p<.05 Source: LifE 2012, own calculations Predicting the Timing of Social Transitions ... • 173 Fig. 2: Percentages of women and men, who experienced the fi ve transitions until the age of 35 Source: LifE 2012, own calculations 0 20 40 60 80 100 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 age first intercourse first relationship first cohabitation first marriage first birth in percent Women Men 0 20 40 60 80 100 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 age first intercourse first relationship first cohabitation first marriage first birth in percent • Jana Gläßer, Wolfgang Lauterbach, Fred Berger174 the gender differences might be that women often choose older partners while men select younger ones (Wendt 2010). Moreover, it seems that for young women fi rst sexual experiences are more often associated with a relationship than in the case of men. In the German youth report on sexuality, women reported more often than men that their fi rst sexual partner was well known or a very good friend of them (64 percent vs. 58 percent). 11 percent of men reported that their fi rst sexual part- ner was not well known to them or was a casual acquaintance (women: 4 percent; Federal Centre for Health Education 2010). Next, using Cox regression, we examined the timing of the fi ve partnership and family transitions. Table 2 summarizes the results for the complete models. It turns out that personal and social resources infl uence the timing of early transi- tions such as the fi rst sexual experience, the fi rst partnership and the fi rst cohabita- tion with a partner. A high degree of emotional stability increases the probability of these events by 11 percent to 19 percent per standard deviation. A high degree of outgoingness, a positive body image and a high degree of peer group integra- tion accelerate the timing of the fi rst sexual experience and the fi rst partnership by 9 percent to 16 percent per standard deviation. The experience of a parental divorce Tab. 2: Cox regression of timing of fi rst sexual experience, fi rst partnership, fi rst cohabitation, fi rst marriage and family formation (Hazard Ratios) First sexual First First coha- First First birth experience partnership bitation marriage Men (ref. women) 0.88 0.72*** 0.76** 0.91 0.83 A1: Personal resources Emotional stabilityz 1.19*** 1.15*** 1.11* 1.01 1.04 Outgoingnessz 1.09* 1.09* 1.07 1.09 1.01 Body imagez 1.12** 1.09* 1.05 1.00 0.99 A2: Social resources Peer group integrationz 1.16*** 1.13** 1.05 0.98 1.05 Parental divorce 1.83* 1.31 1.18 1.05 1.26 B: Socio-economic resources Individual level of education: low/medium (ref.) high 0.47 0.84 0.13*** 0.07*** 0.02*** Economic independenceb 1.08 1.09 1.54*** 5.23*** 2.91*** Education*timec 1.04 1.01 1.09*** 1.09*** 1.13*** N subjects 695 695 695 695 695 N events 684 677 652 573 551 Chi² 79.15*** 64.81*** 85.64*** 89.10*** 70.80*** Level of signifi cance: * p < 0.05, ** p < 0.01, *** p < 0.001 z z-standardized continuous covariates; b time-dependent covariates; c time-varying effect Robustness check: controlled for parental socio-economic status (HISEI) Source: LifE 2012, own calculations Predicting the Timing of Social Transitions ... • 175 or separation in childhood or youth appears to have a strong infl uence on the timing of the fi rst sexual experience (acceleration rate of 83 percent). In contrast, the tim- ing of the fi rst marriage and the birth of the fi rst child are not signifi cantly infl uenced by personal or social resources. Socio-economic resources show no signifi cant infl uences on the timing of the fi rst sexual experience and the fi rst partnership. However, the fi rst cohabitation, fi rst marriage and the birth of the fi rst child can signifi cantly be predicted by young adults’ education level and the timing of their economic independence. Educational attainment is modelled as a time-varying variable. This means that the transition rate for highly educated men and women is at fi rst much lower than for men and women with a vocational degree (Hazard Ratio: 0.13). But, the difference in the haz- ard ratio declines by 9 percent (factor of 1.09) each year. This pattern is repeated for fi rst marriage and the birth of the fi rst child. Highly educated people delay marriage and childbirth but converge with lower educated people over time. Furthermore, achieving economic independence leads to a clear acceleration of the fi rst cohabi- tation (increase of 54 percent), fi rst marriage (423 percent) and the timing of family formation (191 percent). Overall, results show that the timing of the fi rst sexual experience and partner- ship is infl uenced by personal and social resources. In contrast, the timing of the fi rst marriage and the birth of the fi rst child are infl uenced by socio-economic re- sources. Entering the fi rst cohabitation is infl uenced by personal and social as well as socio-economic resources. Table 3 summarizes the results for women. We see that personal resources sig- nifi cantly infl uence the timing of women’s fi rst sexual experience, their fi rst partner- ship and their fi rst cohabitation. A high emotional stability increases the probability of the fi rst sexual experience and fi rst partnership by 26 percent and 17 percent respectively. An increase of one standard deviation on the outgoingness scale leads to an increase in the probability of the fi rst sexual experience, fi rst partnership and fi rst cohabitation by 12 percent to 19 percent. For women, a favourable body im- age accelerates the timing of the sexual debut and the fi rst partnership: The higher the evaluation of their own appearance, the more likely women are to begin these events (acceleration of 10 percent to 12 percent per standard deviation). The timing of the fi rst marriage and the birth of the fi rst child are not signifi cantly infl uenced by personal resources. Further social resources do not have a strong infl uence on any transitions studied. Only the fi rst sexual experience is strongly infl uenced by divorce experiences in childhood and youth (acceleration rate of 52 percent per standard deviation). For women, personal resources seem to exert a greater infl u- ence than social resources on the timing of social transitions. Socio-economic resources have no signifi cant infl uences on the timing of the fi rst sexual experience and the fi rst partnership. For the timing of the fi rst cohabita- tion, only women’s economic independence is signifi cant (acceleration of 57 per- cent). Major infl uences on the timing of the fi rst marriage and the birth of the fi rst child are exerted by the level of education and economic independence. Since edu- cational attainment is modelled as a time-varying effect, at fi rst, the marriage rate for highly educated women is much lower than for women with a vocational degree • Jana Gläßer, Wolfgang Lauterbach, Fred Berger176 (Hazard Ratio: 0.05). But, the effect declines by 10 percent each year (factor of 1.10). This pattern is repeated for the timing of the fi rst birth. Women with a university de- gree have their fi rst child later but converge with lower educated women over time. Table 4 presents the results for men. Here, the highest level of education is also modelled as time-varying effect. In contrast to women, personal resources do not appear to have a strong infl uence on the timing of partnership events. Only body image has a strong infl uence on the timing of men’s fi rst sexual experiences: The more attractive they feel, the earlier they experience their fi rst sexual contact (ac- celeration of 17 percent per standard deviation). While for women social resources have relatively little infl uence, men are strong- ly infl uenced by their integration in a peer group as well as by the experience of a parental divorce in childhood or youth. A high peer group integration accelerates the sexual debut, the fi rst relationship and the fi rst cohabitation by 20 percent to 30 percent per standard deviation. Moreover, the timing of these three events is strongly accelerated (74 percent to 140 percent) by the experience of a parental di- vorce in childhood or adolescence. Tab. 3: Cox regression of timing of fi rst sexual experience, fi rst partnership, fi rst cohabitation, fi rst marriage and family formation for women (Hazard Ratios) First sexual First First coha- First First birth experience partnership bitation marriage A1: Personal resources Emotional stabilityz 1.26*** 1.17** 1.10 0.97 1.07 Outgoingnessz 1.19** 1.12* 1.14* 1.09 1.07 Body imagez 1.10* 1.12* 1.04 0.99 0.96 A2: Social resources Peer group integrationz 1.06 1.08 0.96 0.94 1.02 Parental divorce 1.52* 1.11 0.87 0.82 1.21 B: Socio-economic resources Individual level of education: low/medium (ref.) high 0.30 0.89 0.29 0.05** 0.02** Economic independenceb 0.95 0.97 1.57** 4.31*** 1.97* Education*timec 1.07 1.01 1.05 1.10** 1.12** N subjects 388 388 388 388 388 N events 383 379 364 322 312 Chi² 44.91*** 22.42* 35.29*** 53.54*** 32.41*** Level of signifi cance: * p < 0.05, ** p < 0.01, *** p < 0.001 z – z-standardized continuous covariates; b time-dependent covariates; c time-varying effect Robustness check: controlled for parental socio-economic status (HISEI) Source: LifE 2012, own calculations Predicting the Timing of Social Transitions ... • 177 With respect to socio-economic resources, results show that achieving econom- ic independence leads to a profound acceleration of the fi rst cohabitation, the fi rst marriage and the birth of the fi rst child (acceleration of 45 to 506 percent). Further- more, highly educated men experience these events later. However, their timing converges over the course of time with that of men who have a vocational degree. 5 Discussion The purpose of this contribution was, fi rst, to determine how personal, social and socio-economic resources infl uence the timing of fi ve transitions on the way to mid- dle adulthood and, second, whether there are varying infl uences of personal, social and socio-economic resources among women and men. For this purpose, we ap- plied a resource model of life management, assuming that people pass different transitions drawing back on these resources. In line with Hypothesis 1, we found that personal and social resources predomi- nantly infl uence the timing of events that occur in late youth or early adulthood, Tab. 4: Cox-regression for timing of fi rst sexual experience, fi rst partnership, fi rst cohabitation, fi rst marriage and family formation for men (Hazard Ratios) First sexual First First coha- First First birth experience partnership bitation marriage A1: Personal resources Emotional stabilityz 1.12 1.10 1.11 1.06 1.00 Outgoingnessz 1.01 1.01 0.99 1.09 0.94 Body imagez 1.17* 1.06 1.07 1.00 1.05 A2: Social resources Peer group integrationz 1.30*** 1.30*** 1.20** 0.99 1.10 Divorce experience 2.40*** 1.74*** 1.89*** 1.48 1.33 B: Socio-economic resources Individual level of education: low/medium (ref.) high 0.51 0.69 0.11*** 0.16* 0.03*** Economic independenceb 1.14 1.23 1.45* 6.06*** 5.25** Education*timec 1.03 1.02 1.09*** 1.06* 1.12*** N subjects 307 307 307 307 307 N events 301 298 288 251 239 Chi² 66.52*** 44.86*** 47.99*** 33.27*** 30.00*** Level of signifi cance: * p < 0.05, ** p < 0.01, *** p < 0.001 z z-standardized continuous covariates; b time-dependent covariates; c time-varying effect Robustness check: controlled for parental socio-economic status (HISEI) Source: LifE 2012, own calculations • Jana Gläßer, Wolfgang Lauterbach, Fred Berger178 such as the fi rst sexual intercourse and the fi rst long-term relationship. At this stage of partnership formation processes individuals obviously draw on their own per- sonal and social resources, whereas socio-economic resources are less relevant. However, with increasing institutionalization of the partnership, i.e. the sharing of a household, marriage and the birth of the fi rst child, socio-economic resources be- come dominant. At these stages of partnership and family formation, commitments, responsibilities, and economic investments increase. As the descriptive fi ndings showed, the fi rst sexual experiences and fi rst romantic relationships occur in late adolescence and early adulthood, whereas marriage and childbirth take place later in life. Thereby, partnerships in youth do not only differ from those in adulthood in terms of timing and duration but also concerning the role, investment and expecta- tions placed on them (Walper et al. 2008). With marriage and parenthood, mutual support, stable emotional bonds and fi nancial security become more important. Overall, it turns out that in terms of personal, social and socio-economic re- sources the social life course from late adolescence to middle adulthood consists of two phases: The fi rst phase is predominantly infl uenced by personal and social re- sources, while the second phase is determined by socio-economic resources. While early events like the fi rst sexual contact and fi rst partnership more strongly depend on personal and social resources, educational attainment and especially economic security become more important with increasing institutionalization of partnerships (marriage and birth of the fi rst child). The fi rst cohabitation seems to play an in- termediate role in the process of partnership and family formation. The timing is infl uenced by personal and social resources as well as socio-economic resources. As very important result we found different effects on the timing of transitions among men and women. While women are strongly infl uenced by personal resourc- es (Hypothesis 2), such as emotional stability, high outgoingness, and self-perceived appearance, male adolescents are strongly infl uenced by their integration in a peer group and the parental support system in their family of origin (Hypothesis 3). A parental divorce proved to accelerate the transition to adulthood for men but not for women (except for the timing of the fi rst sexual experience). The results indicate that those men tend to begin their own romantic relationships at an earlier age than men with (structurally) intact families. It can be supposed that young men are par- ticularly infl uenced by the lack of (mostly) father fi gures, because the father as male role model is absent. Not only the absence of the role model, but also the reduction in parental control caused by a divorce which may lead to an earlier orientiation to peers and partnerships (Diefenbach 1999), seems to affect men in particular. In ad- dition, integration in a social network seems to be much more important for men than for women. Young men, who are part of a network are more likely to carry out transitions earlier. Peer group integration is not related to the timing of transitions among women. Overall, young men seem to be much more closely orientated to others while young women tend to be more interested in close female friends (Fend 2005; Steinberg 2010). Furthermore, the greater importance of the peer group for men compared to women can be explained by the different meaning of socialization agents for men and women. It can be argued that girls are socialized to a greater extent within the family; boys, however, are more oriented to further agents of so- Predicting the Timing of Social Transitions ... • 179 cialization besides the family (Erzinger 2009), such as the peer group which has a greater impact on men’s romantic development than on girls’. In line with our expectations (Hypothesis 4), economic independence acceler- ates the timing of social transitions for both men and women: Not only men are supposed to achieve economic independence before marriage and having children, but also women. Changes in women’s life course structure are obvious. Women and men postpone events like marriage and having children in times of economic uncer- tainty. Thus, economic independence seems to enhance the chances of partnership formation for both men and women (Brose 2008). In common with previous studies (Bauer/Jacob 2010; Dommermuth 2008) and in line with Hypothesis 5, our results show that highly educated women and highly educated men tend to delay marriage and parenthood. However, the effect of edu- cational attainment is stronger for women than for men. Furthermore, the retarding effect of the educational level is stronger for family formation than for marriage, for both men and women. This fi nding indicates that the most important transition marker in a partnership history is no longer marriage but the birth of the fi rst child. However, the infl uence of the educational level is stronger for men than for women. Moreover, the individual educational level only infl uences men’s timing of a cohabi- tation but not women’s. Contrary to our expectations, cohabitation is even more strongly infl uenced by men’s educational attainment than marriage. This indicates the increased acceptance and importance of cohabitation in the last decades. In summary, earlier social transitions appear to be determined by personal and social resources, but we found different mechanisms for men and for women. Later, however, when a partnership becomes institutionalized, the educational attainment and economic situation more strongly infl uence the timing of relevant passages in the process of partnership and family formation. A fi nding that holds true for men and women alike. References Amato, Paul R.; Patterson, Sarah E. 2016: The Intergenerational Transmission of Union Instability in Early Adulthood. In: Journal of Marriage and Family 79,3: 723-738 [doi: 10.1111/jomf.12384]. Ambwani, Suman; Strauss, Jaine 2007: Love Thyself Before Loving Others? A Qualita- tive and Quantitative Analysis of Gender Differences in Body Image and Romantic Love. In: Sex Roles 56,1-2: 13-21 [doi: 10.1007/s11199-006-9143-7]. Asendorpf, Jens B.; Neyer, Franz J. 2012: Psychologie Der Persönlichkeit. 5. Aufl age. Berlin: Springer. Atkins, Robert; Hart, Daniel 2008: The Under-Controlled Do It First: Childhood Person- ality and Sexual Debut. In: Research in Nursing & Health 31,6: 626-639 [doi: 10.1002/ nur.20288]. Bauer, Gerrit; Jacob, Marita 2010: Fertilitätsentscheidungen im Partnerschaftskontext. In: KZfSS Kölner Zeitschrift für Soziologie und Sozialpsychologie 62,1: 31-60 [doi: 10.1007/s11577-010-0089-y]. • Jana Gläßer, Wolfgang Lauterbach, Fred Berger180 Beck, Ulrich 1999: Die Zukunft Der Arbeit Oder Die Politische Ökonomie Der Unsicher- heit. In: Berliner Journal Für Soziologie 9,4: 467-479. Beck, Ulrich 1986: Risikogesellschaft. Auf Dem Weg in Eine Andere Moderne. Frankfurt am Main: Suhrkamp. Becker-Stoll, Fabienne et al. 2000: Autonomie und Verbundenheit bei Jugendlichen und jungen Erwachsenen. In: ZSE: Zeitschrift für Soziologie der Erziehung und Sozialisa- tion 20,4: 345-361. Berger, Fred 2009: Intergenerationale Transmission von Scheidung – Vermittlungspro- zesse Und Scheidungsbarrieren. In: Fend, Helmut; Berger, Fred; Grob, Urs (Eds.): Le- bensverläufe, Lebensbewältigung, Lebensglück. Ergebnisse Der LifE-Studie. Wiesba- den: VS Verlag für Sozialwissenschaften: 267-303. Berrington, Ann; Diamond, Ian 2000: Marriage or Cohabitation: A Competing Risks Analysis of First-Partnership Formation among the 1958 British Birth Cohort. In: Jour- nal of the Royal Statistical Society: Series A (Statistics in Society) 163,2: 127. BiB 2009: 1973-2008. 35 Jahre Bevölkerungswissenschaftliche Forschung Am BiB. Ein Öffentlicher Tätigkeitsbericht. Wiesbaden [https://www.bib.bund.de/Publikati- on/2009/pdf/35-Jahre-bevoelkerungswissenschaftliche-Forschung-am-BiB.pdf?_ _ blob=publicationFile&v=5, 11.4.2018]. Blossfeld, Hans-Peter 2010: Survival-und Ereignisanalyse. In: Best, Henning; Wolf, Chri- stof (Eds.): Handbuch der sozialwissenschaftlichen Datenanalyse. Wiesbaden: VS Ver- lag für Sozialwissenschaften: 995-1016 [doi: 10.1007/978-3-531-92038-2_37]. Blossfeld, Hans-Peter; Drobnic, Sonja (Eds.) 2001: Careers of Couples in Contemporary Society: From Male Breadwinner to Dual-Earner Families. Oxford/New York: Oxford University Press. Blossfeld, Hans-Peter; Huinink, Johannes 1991: Human Capital Investments or Norms of Role Transition? How Women’s Schooling and Career Affect the Process of Family Formation. In: American Journal of Sociology 97,1: 143-168 [doi: 10.1086/229743]. Blossfeld, Hans-Peter; Timm, Andreas 1997: Das Bildungssystem als Heiratsmarkt: eine Längsschnittanalyse der Wahl von Heiratspartnern im Lebenslauf. Arbeitspapier / Sfb 186. Bremen: Universität Bremen. Bode, Heidrun; Heßling, Angelika 2015: Forschung Sexualaufklärung: Jugendsexualität 2015. Repräsentative Wiederholungsbefragung. Die Perspektive Der 14- Bis 25 Jähri- gen. Köln: Bundeszentrale für gesundheitliche Aufklärung. Brose, Nicole 2008: Entscheidung Unter Unsicherheit – Familiengründung und -erweite- rung Im Erwerbsverlauf. In: KZfSS Kölner Zeitschrift Für Soziologie Und Sozialpsycho- logie 60,1: 34-56 [doi: 10.1007/s11577-008-0002-0]. Brown, Lyn Mikel; Gilligan, Carol 1994: Die Verlorene Stimme: Wendepunkte in Der Ent- wicklung von Mädchen und Frauen. Frankfurt/Main [u.a.]: Campus. Brüderl, Josef; Diekmann, Andreas 1997: Education and Marriage * A Comparative Study. Caspi, Avshalom; Elder, Glen H.; Bem, Daryl J. 1988: Moving Away from the World: Life- course patterns of shy children. In: Developmental Psychology 24,6: 824-831 [doi: 10.1037/0012-1649.24.6.824]. Davis, Teresa L. 1995: Gender Differences in Masking Negative Emotions: Abil- ity or Motivation? In: Developmental Psychology 31,4: 660-667 [doi: 10.1037/0012- 1649.31.4.660]. Predicting the Timing of Social Transitions ... • 181 Dennissen, Jaap J.A.; Asendorpf, Jens B.; van Aken, Marcel A.G. 2008: Childhood Per- sonality Predicts Long-Term Trajectories of Shyness and Aggressiveness in the Con- text of Demographic Transitions in Emerging Adulthood. In: Journal of Personality 76,1: 67-99 [doi: 10.1111/j.1467-6494.2007.00480.x]. Diefenbach, Heike 1999: Intergenerationale Scheidungstransmission in Deutschland: Relevanz und Erklärungsansätze. In: Hradil, Stefan et al. (Hrsg.): Differenz und Inte- gration: 117-121. Diekmann, Andreas 1990: Der Einfl uß Schulischer Bildung Und Die Auswirkungen Der Bildungsexpansion Auf Das Heiratsverhalten. In: Zeitschrift Für Soziologie 19,4: 265- 277 [doi: 10.1515/zfsoz-1990-0403]. Diekmann, Andreas; Engelhardt, Henriette 1995: Die Soziale Vererbung Des Schei- dungsrisikos. Eine Empirische Untersuchung Der Transmissionshypothese Mit Dem Deutschen Familiensurvey. In: Zeitschrift Für Soziologie 24,3: 215-228. Diekmann, Andreas; Schmidheiny, Kurt 2013: The Intergenerational Transmission of Di- vorce: A Fifteen-Country Study with the Fertility and Family Survey. In: Comparative Sociology 12,2: 211-235 [doi: 10.1163/15691330-12341261]. Diewald, Martin; Mayer, Karl Ulrich 2009: The Sociology of the Life Course and Life Span Psychology: Integrated Paradigm or Complementing Pathways? In: Advances in Life Course Research, Linked Lives and Self-Regulation. Lifespan – Life Course: Is It Really the Same? 14,1-2: 5-14 [doi: 10.1016/j.alcr.2009.03.001]. Dommermuth, Lars 2008: Wege Ins Erwachsenenalter in Europa. Forschung Gesell- schaft. 1. Aufl . Wiesbaden: VS Verl. für Sozialwissenschaften. Elder, Glen H. 1978: Family history and the life course. In: Hareven, Tamara K.: Transi- tions: The Family and the Life Course in historical perspective. New York: Academic Press: 17-64. Elder, Glen H.; Johnson, Monica Kirkpatrick; Crosnoe, Robert 2004: The Emergence and Development of Life Course Theory. In: Mortimer, Jeylan T.; Shanahan, Michael J. (Eds.): Handbook of the Life Course, Handbooks of Sociology and Social Research. New York, NY: Springer: 3-19 [doi: 10.1007/978-0-306-48247-2_1]. Engelhardt, Henriette 1998: Zur Dynamik von Ehescheidungen. Duncker & Humblot. Erzinger, Andrea 2009: Langzeitfolgen familiärer Beziehungserfahrungen im Jugendal- ter für partnerschaftliche Beziehungen im Erwachsenenalter – Wird die Qualität der Partnerbeziehung über die Generationen „vererbt“? In: Fend, Helmut; Berger, Fred; Grob, Urs (Eds.): Lebensverläufe, Lebensbewältigung, Lebensglück. Ergebnisse Der LifE-Studie. Wiesbaden: VS Verlag für Sozialwissenschaften. Federal Centre for Health Education 2010: Jugendsexualität 2010. Repräsentative Wie- derholungsbefragung von 14- Bis 17-Jährigen Und Ihren Eltern. Köln: BZgA [http:// publikationen.sexualaufklaerung.de/index.php?docid=1944, 05.06.2017]. Federal Statistical Offi ce of Germany 2017: Kinderlosigkeit, Geburten und Familien. Er- gebnisse Des Mikrozensus 2016. Wiesbaden: Statistisches Bundesamt. Federal Statistical Offi ce of Germany 2016a: Bei Jeder Dritten Geburt sind Eltern nicht Verheiratet. Pressemitteilung Vom 19. Dezember 2016 – 461/16. Wiesbaden: Statisti- sches Bundesamt. Federal Statistical Offi ce of Germany 2016b: Statistisches Jahrbuch. Deutschland Und Internationales 2016. Wiesbaden: Statistisches Bundesamt. Federal Statistical Offi ce of Germany 2002: Wirtschaft Und Statistik 1-2002. Wiesbaden: Statistisches Bundesamt. • Jana Gläßer, Wolfgang Lauterbach, Fred Berger182 Feldhaus, Michael; Heintz-Martin, Valerie 2015: Long-Term Effects of Parental Separa- tion: Impacts of Parental Separation during Childhood on the Timing and the Risk of Cohabitation, Marriage, and Divorce in Adulthood. In: Advances in Life Course Re- search 26: 22-31 [doi: 10.1016/j.alcr.2015.07.003]. Fend, Helmut 2005: Entwicklungspsychologie des Jugendalters. Nachdr. der 3., durch- ges. Aufl . 2003. Wiesbaden: VS, Verl. für Sozialwiss. Fend, Helmut; Berger, Fred; Grob, Urs 2009: 1527 „Lebensgeschichten“ von der späten Kindheit ins Erwachsenenalter – Konzept und Durchführung der LifEStudie. In: Fend, Helmut; Berger, Fred; Grob, Urs (Eds.): Lebensverläufe, Lebensbewältigung, Lebens- glück. VS Verlag für Sozialwissenschaften: 9-34 [doi: 10.1007/978-3-531-91547-0_1]. Fend, Helmut; Prester, Hans-Georg 1986: Dokumentation Der Skalen Des Projekts „Ent- wicklung Im Jugendalter“. Universität Konstanz: Sozialwissenschaftliche Fakultät. Goldscheider, Frances; Goldscheider, Calvin 1999: The Changing Transition to Adult- hood: Leaving and Returning Home. Thousand Oaks, California: Sage Publications [doi: 10.4135/9781452234014]. Grünheid, Evelyn 2017: Wandel Der Lebensformen in Deutschland. BiB Working Paper 2-2017. Wiebaden: Bundesinstitut für Bevölkerungsforschung. Hayford, Sarah R.; Guzzo, Karen Benjamin; Smock, Pamela J. 2014: The Decoupling of Marriage and Parenthood? Trends in the Timing of Marital First Births, 1945-2002. In: Journal of Marriage and the Family 76,3: 520-538. Huinink, Johannes 2000: Bildung und Familienentwicklung im Lebensverlauf. In: Zeitschrift für Erziehungswissenschaft 3,2: 209-227 [doi: 10.1007/s11618-000-0022-7]. Huinink, Johannes 1995: Warum noch Familie. Zur Attraktivität von Partnerschaft und Elternschaft in unserer Gesellschaft. Frankfurt am Main: Campus. Huinink, Johannes 1989: Ausbildung, Erwerbsbeteiligung von Frauen und Familienbil- dung im Kohortenvergleich. In: Wagner, Gert; Ott, Notburga; Hoffmann-Nowotny, Hans-Joachim (Eds.): Familienbildung und Erwerbstätigkeit im demographischen Wandel: 136-158 [doi: 10.1007/978-3-642-83989-4_9]. Huinink, Johannes; Feldhaus, Michael 2009: Family Research from the Life Course Per- spective. In: International Sociology 24,3: 299-324 [doi: 10.1177/0268580909102910]. Huinink, Johannes; Kreyenfeld, Michaela 2004: Family Formation in Times of Social and Economic Change: An Analysis of the 1971 East German Cohort. In: MPIDR Working Paper WP 2004-13 [http://www.demogr.mpg.de/de/projekte_publikationen/ publikationen_1904/mpidr_working_papers/family _formation_in_times_of_social_ and_economic_change_an_analysis_of_the_1971_east_ german_cohort_1626.htm, 21.02.2017]. Hurrelmann, Klaus 2002: Einführung in Die Sozialisationstheorie. Beltz-Studium: Kultur und Gesellschaft. 8., vollst. überarb. Aufl . Weinheim [u.a.]: Beltz. Hurrelmann, Klaus; Quenzel, Gudrun 2012: Lebensphase Jugend. Grundlagentexte So- ziologie. 11., vollst. überarb. Aufl . Weinheim: Beltz Juventa. Janssens, Angélique 1997: The Rise and Decline of the Male Breadwinner Family? An Overview of the Debate. In: International Review of Social History 42,S5: 1-23 [doi: 10.1017/S0020859000114774]. Jokela, Markus et al. 2011: Reproductive Behavior and Personality Traits of the Five Fac- tor Model. In: European Journal of Personality 25,6: 487-500 [doi: 10.1002/per.822]. Kokko, Katja; Pulkkinen, Lea; Mesiäinen, Päivi 2009: Timing of Parenthood in Relation to Other Life Transitions and Adult Social Functioning. In: International Journal of Behav- ioral Development 33,4: 356-365 [doi: 10.1177/0165025409103873]. Predicting the Timing of Social Transitions ... • 183 Konietzka, Dirk 2010: Zeiten Des Übergangs. 1. Aufl . Wiesbaden: VS Verlag für Sozial- wissenschaften. Krack-Roberg, Elle et al. 2016: Familie, Lebensformen und Kinder. In: Statistisches Bun- desamt (Ed.): Datenreport 2016. Ein Sozialbericht Für Die Bundesrepublik Deutsch- land. Bonn: 42-77. Kreyenfeld, Michaela 2015: Economic Uncertainty and Fertility. In: KZfSS Kölner Zeitschrift Für Soziologie Und Sozialpsychologie 67,S1: 59-80 [doi: 10.1007/s11577- 015-0325-6]. Kreyenfeld, Michaela; Konietzka, Dirk 2015: Sozialstruktur Und Lebensform. In: Hill, Paul B.; Kopp, Johannes (Eds.): Handbuch Familiensoziologie. Wiesbaden: Springer VS: 345-373. Kreyenfeld, Michaela; Konietzka, Dirk 2008: Education and Fertility in Germany. In: Hamm, Ingrid; Seitz, Helmut; Werding, Martin (Eds.): Demographic Change in Ger- many. Berlin/Heidelberg: Springer: 165-187 [doi: 10.1007/978-3-540-68137-3_7]. Lauterbach, Wolfgang 2006: Partner Ja, Single Nein, Oder Umgekehrt? Wege Der Sozialen Verselbstständigung und Die Dauer von Partnerschaften bis ins frühe Erwachsenen- alter. In: Lettke, Frank; Lange, Andreas (Eds.): Generationen Und Familie. Frankfurt am Main: Suhrkamp: 161-188. Lauterbach, Wolfgang 2004: Die multilokale Mehrgenerationenfamilie. Volume 13. Fami- lie und Gesellschaft. Würzburg: Ergon-Verl. Lauterbach, Wolfgang; Fend, Helmut; Gläßer, Jana 2016: LifE – Die Studie. Potsdam: Universitätsverlag. Lehnart, Judith; Neyer, Franz J.; Eccles, Jacquelynne 2010: Long-Term Effects of Social Investment: The Case of Partnering in Young Adulthood. In: Journal of Personality 78,2: 639-670 [doi: 10.1111/j.1467-6494.2010.00629.x]. Marbach, Jan H.; Tölke, Angelika 1996: Junge Erwachsene. Wandel Im Partnerschafts- verhalten und die Bedeutung Sozialer Netzwerke. In: Behrens, Johann; Voges, Wolf- gang (Eds.): Kritische Übergänge. Statuspassagen Und Sozialpolitische Institutionali- sierung. Frankfurt am Main: 114-154. Mayer, Karl Ulrich 2009: New Directions in Life Course Research. In: Annual Review of Sociology 35,1: 413-433 [doi: 10.1146/annurev.soc.34.040507.134619]. Mayer, Karl Ulrich; Müller, Walter 2009: Lebensverläufe Im Wohlfahrtsstaat. In: Soziale Ungleichheit: Klassische Texte Zur Sozialstrukturanalyse. Frankfurt a. M. [u.a.]: Cam- pus: 427-446. McCormick, Christopher M.; Kuo, Sally I-Chun; Masten, Ann S. 2011: Developmental Tasks across the Lifespan. In: Fingerman, Karen L. et al. (Eds.): Handbook of Lifespan Development. New York: Springer: 117-140. Meulemann, Heiner 1999: Stichwort: Lebenslauf, Biographie Und Bildung. In: Zeitschrift Für Erziehungswissenschaft 2,3: 305-324. Müller, Annette 2006: Die sexuelle Sozialisation in der weiblichen Adoleszenz. Münster u.a.: Waxmann Verlag. Müller, Rolf; Sommer, Thorsten; Timm, Andreas 1999: Nichteheliche Lebensgemein- schaft oder Ehe? Einfl üsse auf die Wahl der Partnerschaftsform beim ersten Zusam- menzug im Lebenslauf. In: Zeitschrift für Bevölkerungswissenschaft 24,4: 449-472. Naudascher, Brigitte 1978: Jugend und Peer Group. Die pädagogische Bedeutung der Gleichaltrigen im Alter von zwölf bis sechzehn Jahren. Bad Heilbrunn/Obb.: pedocs. • Jana Gläßer, Wolfgang Lauterbach, Fred Berger184 Neyer, Franz J. 1999: Die Persönlichkeit junger Erwachsener in Verschiedenen Lebens- formen. In: Kölner Zeitschrift für Soziologie und Sozialpsychologie 51,3: 491-508. Neyer, Franz J.; Lehnart, Judith 2007: Relationships Matter in Personality Development: Evidence From an 8-Year Longitudinal Study Across Young Adulthood. In: Journal of Personality 75,3: 535-568 [doi: 10.1111/j.1467-6494.2007.00448.x]. Nock, Steven L. 1995: A Comparison of Marriages and Cohabiting Relationships. In: Journal of Family Issues 16,1: 53-76 [doi: 10.1177/019251395016001004]. Oerter, Rolf; Dreher, Eva 2002: Jugendalter. In: Oerter, Rolf; Montada, Leo (Eds.): Ent- wicklungspsychologie. Weinheim: Beltz: 258-318. Petermann, Franz; Schmidt, Martin 2006: Ressourcen – Ein Grundbegriff Der Entwick- lungspsychologie Und Entwicklungspsychopathologie. In: Kindheit Und Entwicklung 15,2: 118-127 [doi: 10.1026/0942-5403.15.2.118]. Peuckert, Rüdiger 2012: Familienformen Im Sozialen Wandel. Wiesbaden: VS Verl. für Sozialwiss [doi: 10.1007/978-3-531-19031-0]. Raab, Marcel 2017: Childhood Family Structure and Early Family Formation in East and West Germany. In: Journal of Marriage and Family 79,1: 110-130 [doi: 10.1111/ jomf.12333]. Raithel, Jürgen 2011: Jugendliches Risikoverhalten. Eine Einführung. 2. überarbeitete Aufl age. Wiesbaden: VS Verlag für Sozialwissenschaften [doi: 10.1007/978-3-531- 94066-3]. Reitz, Anne K. et al. 2014: How Peers Make a Difference: The Role of Peer Groups and Peer Relationships in Personality Development. In: European Journal of Personality 28,3: 279-288 [doi: 10.1002/per.1965]. Ross, Andy et al. 2009: Family and Nonfamily Role Confi gurations in Two British Cohorts. In: Journal of Marriage and Family 71,1: 1-14 [doi: 10.1111/j.1741-3737.2008.00576.x]. Schaeper, Hildegard; Grotheer, Michael; Brandt, Gesche 2013: Familiengründung von Hochschulabsolventinnen. Eine empirische Untersuchung verschiedener Examensko- horten. In: Konietzka, Dirk; Kreyenfeld, Michaela (Eds.): Ein Leben ohne Kinder. Aus- maß, Strukturen und Ursachen von Kinderlosigkeit. Wiesbaden: Springer VS: 47-80. Scherger, Simone 2007: Destandardisierung, Differenzierung, Individualisierung: West- deutsche Lebensläufe im Wandel. Wiesbaden: VS Verlag für Sozialwissenschaften [doi: 10.1007/978-3-531-90712-3]. Schmitt, Christian; Winkelmann, Ulrike 2016: Wer Bleibt Kinderlos? Was sozialstruktu- relle Daten über Kinderlosigkeit bei Frauen und Männern verraten. In: Feministische Studien 23,1: 9-23 [doi: 10.1515/fs-2005-0103]. Schulz, Sonja 2009: Intergenerationale Scheidungstransmission und Aufwachsen in Stieffamilien: gibt es den Transmissionseffekt auch bei Stiefkindern? In: Zeitschrift für Familienforschung 21,1: 5-29. Seyda, Susanne 2003: Frauenerwerbstätigkeit und Geburtenverhalten. In: IW-Trends – Vierteljahresschrift Zur Empirischen Wirtschaftsforschung 30,2: 26-36. Shulman, Shmuel; Seiffge-Krenke, Inge 1996: Fathers and Adolescents: Developmental and Clinical Perspectives. London/New York: Routledge. Smolak, Linda 2004: Body Image in Children and Adolescents: Where Do We Go from Here? In: Body Image 1,1: 15-28 [doi: 10.1016/S1740-1445(03)00008-1]. Sprenger, Robin 2014: Konzept der Partnerlosigkeit aus kommunikationswissenschaftli- cher Perspektive. In: Männliche Absolute Beginner. Wiesbaden: Springer VS: 117-142 [doi: 10.1007/978-3-658-05924-8_5]. Predicting the Timing of Social Transitions ... • 185 Stavrova, Olga; Ehlebracht, Daniel 2015: A Longitudinal Analysis of Romantic Relation- ship Formation: The Effect of Prosocial Behavior. In: Social Psychological and Person- ality Science 6,5: 521-527 [doi: 10.1177/1948550614568867]. Steinberg, Laurence 2010: Adolescence. 9 edition. New York: McGraw-Hill Humanities/ Social Sciences/Languages. Therneau, Terry M.; Grambsch, Patricia M. 2006: Modeling Survival Data: Extending the Cox Model. New York: Springer. Thornton, Arland; Axinn, William G.; Teachman, Jay D. 1995: The Infl uence of School Enrollment and Accumulation on Cohabitation and Marriage in Early Adulthood. In: American Sociological Review 60,5: 762-774 [doi: 10.2307/2096321]. van den Brink, Femke et al. 2018: Relationships between Body Image, Sexual Satisfac- tion, and Relationship Quality in Romantic Couples. In: Journal of Family Psychology 32,4: 466-474 [doi: 10.1037/fam0000407]. Wagner, Jenny et al. 2015: The First Partnership Experience and Personality Develop- ment: A Propensity Score Matching Study in Young Adulthood. In: Social Psychologi- cal and Personality Science 6,4: 455-463 [doi: 10.1177/1948550614566092]. Walper, Sabine et al. 2008: Paarbeziehungen im Jugend- und Erwachsenenalter: Ent- wicklungsbedingte Besonderheiten und Einfl üsse der Herkunfsfamilie. In: Feldhaus, Michael; Huinink, Johannes (Eds.): Neuere Entwicklungen in der Beziehungs- und Familienforschung. Vorstudien zum Beziehungs- Und Familienpanel. Würzburg: Er- gon-Verl.: 115-150. Weick, Stefan 1993: Determinanten des Auszugs aus der Elterlichen Wohnung. In: Diek- mann, Andreas; Weick, Stefan (Eds.): Der Familienzyklus als sozialer Prozeß. Bevölke- rungssoziologische Untersuchungen mit den Methoden der Ereignisanalyse, Sozial- wissenschaftliche Schriften. Berlin: Duncker & Humblot: 86-108. Weis, Mirjam; Heikamp, Tobias; Trommsdorff, Gisela 2013: Gender Differences in School Achievement: The Role of Self-Regulation. In: Frontiers in Psychology 4 [doi: 10.3389/fpsyg.2013.00442]. Wendt, Eva-Verena 2010: Paarbeziehungen und Sexualverhalten heutiger Jugendlicher – Aus der Sicht der Sozialwissenschaften. Presented at the Tagung „Extreme – Ju- gendliteratur Ohne Tabus?“, Am 19.6.2010 in Tutzing. White, Lynn 1994: Coresidence and Leaving Home: Young Adults and Their Parents. In: Annual Review of Sociology 20,1: 81-102. [https://doi.org/10.1146/annurev. so.20.080194.000501]. Windzio, Michael 2013: Regressionsmodelle für Zustände und Ereignisse. Wiesbaden: Springer VS. Date of submission: 09.08.2017 Date of acceptance: 05.09.2018 Jana Gläßer (), Prof. Dr. Wolfgang Lauterbach. University of Potsdam. Potsdam, Germany. E-mail: glaesser@uni-potsdam.de, wolfgang.lauterbach@uni-potsdam.de URL: https://www.uni-potsdam.de/de/sozbildung/lehrstuhl/jana-glaesser.html https://www.uni-potsdam.de/de/sozbildung/lehrstuhl/wolfang-lauterbach.html Prof. Dr. Fred Berger. University of Innsbruck. Innsbruck, Austria. E-mail: fred.berger@uibk.ac.at URL: https://www.uibk.ac.at/iezw/mitarbeiterinnen/univ.-professorinnen/berger/index/ • Jana Gläßer, Wolfgang Lauterbach, Fred Berger186 N o te : Fo r d e ta ile d in fo rm at io n a b o u t si n g le s it e m s, s ca le c o n st ru ct io n a n d c h ar ac te ri st ic v al u e s, s e e F e n d /P re st e r (1 9 8 6 ). S o u rc e : O w n c al cu la ti o n S ca le It e m s A n sw e r C ro n b a ch ’s a lp h a E m o ti o n a l s ta b ili ty I so m e ti m e s g e t u p se t o v e r li tt le t h in g s. ( - ) T w o -p o in t re sp o n se s c a le : tr u e ( 2 ) o r n o t tr u e ( 1) , su m s co re 0 .7 9 O ft e n , I c a n n o t g e t o u t o f m y m o o d s. ( - ) I b e lo n g t o t h o se w h o s o m e ti m e s c a n n o t co n tr o l t h e ir a n g e r. ( - ) S o m e ti m e s I d o n ’t e v e n k n o w w h a t’ s w ro n g w it h m e . ( - ) T h e re a re d a y s w h e n e v e ry o n e a n n o y s m e . ( - ) S o m e ti m e s I a m v e ry s a d f o r n o r e a so n . ( - ) W h e n s o m e th in g a n n o y s m e , I so m e ti m e s lo se m y t e m p e r. ( - ) S o m e ti m e s I d o n ’t c a re a b o u t a n y th in g . ( - ) O u tg o in g n e ss I lik e b e in g a ro u n d p e o p le . T w o -p o in t re sp o n se s c a le : tr u e ( 2 ) o r n o t tr u e ( 1) , su m s co re 0 .6 5 I fe e l c o m fo rt a b le b e in g a lo n e . ( - ) I e n jo y m a ki n g n e w a cq u a in ta n ce s. I a lw a y s lo o k fo rw a rd t o m e e ti n g w it h m y f ri e n d s. I d o n ’t l ik e b e in g w it h a l o t o f p e o p le . ( - ) I e n jo y m e e ti n g n e w p e o p le . I d o n ’t m in d b e in g a lo n e f o r d a y s. ( - ) D u ri n g t h e s ch o o l b re a ks I p re fe r to b e a lo n e . ( - ) B o d y i m a g e S o m e ti m e s I e n v y c la ss m a te s w h o l o o k b e tt e r th a n m e . ( - ) T w o -p o in t re sp o n se s c a le : tr u e ( 2 ) o r n o t tr u e ( 1) , su m s co re 0 .7 1 If I l o o ke d b e tt e r, I w o u ld h a v e m o re f ri e n d s. ( - ) C o m p a re d t o o th e rs , I a m r e a lly g o o d l o o ki n g . I h a v e a lr e a d y b e e n t e a se d a b o u t m y a p p e a ra n ce . I h a v e a lr e a d y w is h e d I w o u ld l o o k co m p le te ly d if fe re n t. ( - ) T h e w a y I l o o k, I a m p o p u la r w it h g ir ls /b o y s. P e e r g ro u p in te g ra ti o n H o w m a n y f ri e n d s d o y o u h a v e? v a ri o u s re sp o n se c a te g o ri e s; v a lu e s w e re s ta n d a rd iz e d a n d c a lc u la te d t o a w e ig h te d su m s co re 0 .6 6 H o w m a n y o f y o u r fr ie n d s a re i n y o u r cl a ss ? H o w o ft e n d o y o u m e e t w it h y o u r fr ie n d s a ft e r sc h o o l? D o y o u h a v e t h e o p p o rt u n it y t o m e e t w it h y o u r fr ie n d s a ft e r sc h o o l? D o y o u b e lo n g t o a g ro u p o f y o u th s w h o m e e t re g u la rl y o r o ft e n a n d h a v e a se n se o f b e lo n g in g t o g e th e r? W e d o n o t m e a n a c lu b o r a ss o ci a ti o n . D o y o u s o m e ti m e s o r o ft e n f e e l l o n e ly ? ( - ) D o s ch o o lm a te s co m e t o y o u r p la ce s o m e ti m e s? D o y o u h a v e a b e st f ri e n d ? A p p e n d ix T a b . A 1: It e m s an d s ca le s Published by Prof. Dr. Norbert F. Schneider Federal Institute for Population Research D-65180 Wiesbaden / Germany 2018 Managing Editor Dr. Katrin Schiefer Copy Editor Dr. Evelyn Grünheid Dr. Katrin Schiefer Editorial Assistant Beatriz Feiler-Fuchs Wiebke Hamann Layout Beatriz Feiler-Fuchs E-mail: cpos@bib.bund.de Scientifi c Advisory Board Karsten Hank (Cologne) Michaela Kreyenfeld (Berlin) Marc Luy (Vienna) Peter Preisendörfer (Mainz) Nikola Sander (Wiesbaden) Zsolt Spéder (Budapest) Rainer Wehrhahn (Kiel) Comparative Population Studies www.comparativepopulationstudies.de ISSN: 1869-8980 (Print) – 1869-8999 (Internet) Board of Reviewers Martin Abraham (Erlangen) Laura Bernardi (Lausanne) Hansjörg Bucher (Bonn) Claudia Diehl (Konstanz) Andreas Diekmann (Zurich) Gabriele Doblhammer-Reiter (Rostock) Jürgen Dorbritz (Wiesbaden) Anette Eva Fasang (Berlin) E.-Jürgen Flöthmann (Bielefeld) Alexia Fürnkranz-Prskawetz (Vienna) Beat Fux (Salzburg) Joshua Goldstein (Berkeley) Sonja Haug (Regensburg) Hill Kulu (Liverpool) Aart C. Liefbroer (The Hague) Kurt Lüscher (Konstanz) Emma Lundholm (Umeå) Nadja Milewski (Rostock) Dimiter Philipov (Vienna) Roland Rau (Rostock) Tomáš Sobotka (Vienna) Jeroen Spijker (Barcelona) Olivier Thévenon (Paris) Helga de Valk (Brussels) Heike Trappe (Rostock) Michael Wagner (Cologne)