ESRA 2017 Programme

Tuesday 18th July      Wednesday 19th July      Thursday 20th July      Friday 21th July     

     ESRA Conference App

Wednesday 19th July, 16:00 - 17:30 Room: F2 105

Analyzing Social Mechanisms in Life-Course Research 2

Chair Dr Tilo Beckers (Heinrich Heine University Düsseldorf )
Coordinator 1Dr Dominik Becker (Eberhard Karls Universität Tübingen)
Coordinator 2Dr Nicole Hiekel (University of Cologne)

Session Details

This session links mechanism-based explanations to longitudinal survey research and builds on a recently initiated debate aiming at improving theoretically grounded and empirically sound explanations in life-course research. We aim to bring together pioneer work in the field of life course research that comprises empirical applications using advanced statistical methods to unravel why various life course patterns and outcomes exist by studying how they came about. A key principle of the life course approach is the linkage between contextual characteristics and situations on the one hand (situational mechanisms), and the linkage between situations and action or attitude formation on the other hand (action formation mechanisms). Life course researchers have developed powerful statistical techniques to examine panel data, yet mechanism-based explanations underlying the life course process of interest are rare.

The last decade has seen a growing interest in the concept of social mechanisms (Demeulenaere 2011; Tranow/Beckers/Becker 2016*). The debate focuses on the question of which principles define a satisfactory way of doing social sciences. Most advocates of the social mechanism approach agree that social phenomena should be explained by opening up the black box of explanation and making explicit the causal “cogs and wheels” (Elster 1989) by which these social phenomena are brought into existence.

Life course research provokes questions about social processes and mechanisms, i.e. the identification of patterns and trajectories of stability and change but also of catalysts, drivers and the “wheelworks” (Shanahan/Elder 1997; Keijer/Nagel/Liefbroer 2016). Within the life course approach, analyses and explanations may engage in the identification of i) social mechanisms related to the opportunity structure such as institutional arrangements (e.g., provision of family- or job-related subsidies), ii) belief-driven mechanisms (e.g. rational imitation or self-fulfilling prophecies), and iii) desire-driven mechanisms (e.g educational or job-related preference formation; dissonance reduction).

We invite the submission of abstracts on (emergent) research projects with statistically advanced empirical applications that specify and link social mechanisms to life course patterns and outcomes. Submitted abstracts should include a research question, theoretical and tested social mechanism(s), research design, data and technique of analysis of social mechanisms and (preliminary) results.

* English language Special Issue & Introduction in Analyse und Kritik 38(1): 1-30 (

Paper Details

1. Parenthood and Well-being: The Early Years
Dr Gerrit Bauer (LMU Munich)
Dr Thorsten Kneip (Max Planck Institute for Social Law and Social Policy )
Professor Josef Brüderl (LMU Munich)

Recently, the effect of fertility on parents' happiness has garnered much attention in scientific papers as well as in the media. We focus on the effects of first births on life satisfaction and make the following contributions to the literature:

Analysing data from 7 waves of the German Family Panel (pairfam), we estimate separate impact functions (distributed fixed-effects) for women and men and account for the age of the first child in 3-month intervals. This allows us to estimate the time-varying effect in more detail than does previous research which groups children’s age in broader categories.
We put particular emphasis on controlling for related dynamic features of the life course (or course of relationship, respectively) to isolate the actual effect of parenthood.

Overall, we find a positive effect of a first child on happiness. The effect is stronger for women and lasts until the child is 6-9 months old. Both men and women show positive anticipation effects 6 months before childbirth.

At least two alternative theoretical explanations have been put forth in the literature for explaining why parental well-being might change in response to childbirth and why this effect might not be time-constant but vary with a child’s age. Firstly, a perspective introduced by psychologist Kahneman (1999) and predominant in the empirical economics literature argues that “each individual […] tends to restore well-being to a predetermined set point after each change in circumstances”. Secondly, economic considerations about fertility and the family and the value of children approach argue that benefits and costs associated with childbirth indeed vary with a child’s age. For instance, younger children need more care than older children and are thus associated with higher opportunity costs, but older children cause higher direct costs than younger children. Additionally, transfers by the welfare state (i.e. child and family allowances or compensations) depend on a child’s age. Younger and older children should thus affect the economic situation of parents in different ways.

This calls for a disentangling of potential mechanisms, which might produce the overall pattern. We discuss numerous potential mediators and put them to empirical testing. Besides income, education and health, which already have received attention in previous studies, we also considered stress measures (e.g. average hours of sleep) and frequency of sexual intercourse. These variables could potentially explain why the effect of children on happiness varies with the child’s age.

The moderating impact of costs of children (e.g. more stress, less sex, lower income) we find is weak. Women and men would, by trend, be happier if children did not reduce sleep, income and the satisfaction with sexual intercourse.

2. Mechanisms of the transition to adulthood in cross-national comparison: an application of Hidden Markov Models
Miss Yu Han (Netherlands Interdisciplinary Demographic Institute, The Hague (NIDI/KNAW),)
Professor Aart Liefbroer (Netherlands Interdisciplinary Demographic Institute, The Hague (NIDI/KNAW),)
Professor Cees Elzinga (Netherlands Interdisciplinary Demographic Institute, The Hague (NIDI/KNAW),)

Recent theories about social and demographic change, such as individualization and the second Demographic Transition, suggest a type of late, protracted and complex pathway to adulthood. Over the past years, qualified support has been found for the emergence of a new pattern of transition to adulthood in most European countries. The transition to adulthood is a complex process of a series of interlinked events Life courses are greatly varying sequences of roughly the same life courses events, the complexity being caused by the fact that these sequences consist of correlated events and spells and these correlations seem to depend on gender, social class, cohort and cohort-related macro events. Our previous work demonstrated the application of a first order Hidden Markov model to uncover the mechanisms of transition to adulthood and the roles played by gender and education level of the birth cohort between 1956 and 1965 in France. Methodologically, the Hidden Markov model largely reduces the multi-channel sequence data into life state (hidden state) based transition sequences. Substantively, our result suggested a fertility and partnership driven pathway of transition to adulthood, while covariates played different roles in each life state. To further test the applicability of Hidden Markov models and to deepen our understanding of the transition differences between European countries, we expanded in the Hidden Markov modelling to a cross-national comparison context. Theoretically, we argue that different European countries are at different stages of second Demographic Transition at a given cohort and different transition mechanisms can be detected by Hidden Markov models. Therefore, this study adopts a life course approach using Hidden Markov models to quantify the transition to adulthood in different European countries. We will test hypothesis on social class- (parental SES, education, etc.) and gender related background variables in life state (hidden state) using 18 European/Western countries of respondents born between year 1956 and 1965 in Gender and Generation Survey (GGS), which consists full annual monthly life course sequence data of leaving parental home, partnership history and fertility history between age 15 to 40.

3. Equity and divorce. New findings for Western Germany and the United States.
Professor GOSTA ESPING ANDERSEN (Universitat Pompeu Fabra)

Like Cooke’s earlier studies (2006) we focus on West Germany and the United States not least because they have represented quite different degrees of adaptation to the new role of women, both attitudinally and in couple practices. Our study IS advantaged by the availability of newer recent data, thus permitting us to better identify recent change in light of the clear progress towards more gender egalitarianism in both countries. This , in turn, alloweS us to identify a clear turnabout in couple dynamics – especially for the German case.
In address to our first hypothesis -- the display of traditional gender identities can help stabilize relationships -- we find that, overall, traditional couples are more stable than equitable dual earner couples in West Germany; but not so in United States, where ‘double-shift’ women face a higher divorce risk, and where traditional couples do not enjoy a larger stability premium than do equitable dual earner couples.
As to our second hypothesis -- that equity is increasingly centrally important for conjugal stability within dual-earner partnerships -- we observe that the stability premium related to traditional gender roles in the domestic sphere has declined in both countries and that equitable dual earner couples have increased their advantage in the United States. In other words, the adoption of a traditional division of paid and unpaid work was once the best insurance against divorce, but this is no longer the case, be it in West Germany or the United States.
Another major finding is that the divorce risk among equitable dual earner couples has clearly declined, relatively speaking, within the younger cohorts. Indeed, these couples are now the single most stable in the U.S. Put differently, gender egalitarianism appears now to be key to marital stability in the United States; the same, albeit still in a more embryonic form, appears now to obtain also West Germany.
In this sense, our findings suggest that the explanatory power of the gender construction thesis is waning and possibly disappearing entirely as societies eventually adopt gender egalitarianism more broadly.
In this study we have performed a number of additional checks that should help ensure that our results are valid. An interesting challenge for future research would be to identify more precisely whether value shifts and changing partnership dynamics are being driven by specific social strata (such as, for example, the higher educated) and the extent to which there is ever-greater convergence across the social strata.