Abstracts - families and households strand

Strand organisers: Dr. Wendy Sigle-Rushton, Ben Wilson, London School of Economics & Political Science

Abstracts are listed in the order they are scheduled for presentation. Please refer to the programme for timings.

The changes of family formation patterns in post-socialistic Europe during socio-economic transformation
Tomasz Chaberko, Pawel Kretowicz, Institute of Geography and Spatial Management, Jagiellonian University

The main goal of this study is to examine family formation patterns over time and space in selected countries of Central and Eastern Europe. These countries include: Poland, East Germany (former German Democratic Republic), Czech Republic, Slovakia and Hungary. As our research draws from the concept of Europe's Second Demographic Transition developed by D. Van de Kaa (1987), we attempt to emphasize post-communistic European distinctiveness and regional disparities along with their causes. This distinctiveness is reflected by e.g. fast and sudden pace of changes (during one generation), the considerable role of non-economic determinants (such as relaxation of religion values, changes in mentality) and high reliance on social benefits. The latter is evidenced by the increasing proportion of non-married couples who become entitled to single parent benefits.

Email: tomekchaberko@gmail.com|

The heterogeneity of relationship patterns within and across countries: an examination of the United States and 14 countries in Europe
Brienna Perelli-Harris, Mark Lyons-Amos, University of Southampton

The timing, duration, and repetition of cohabiting and marital unions have become very complex. This variety of relationship types poses a challenge for understanding new family forms and comparing them across countries. Most studies characterize and classify countries based on average behaviours, for example average length of cohabitation or mean age at union formation. In some studies, the predominant form of cohabitation is used to signify the type of cohabitation practiced in a country, for example "alternative to single," or "alternative to marriage." Surprisingly, few studies have examined the multiple types of cohabitation that coexist within countries. Yet multiple forms of relationships can coexist simultaneously within countries, indicating that cohabitation has complex meanings in any country. In this paper we use latent class growth analysis to compare the heterogeneity of union types across 14 countries in Europe and the United States. The latent class growth curves show how individuals can change their relationship status over the lifecourse, between the ages of 15-45. It also shows the predominant age pattern and duration of different relationship states. Statistical tests optimize the number of latent classes within and across countries and calculate the per cent of the population that falls into each class. This allows us to examine significant minorities that practice different types of relationship patterns, as well as the timing of events across the lifecourse. Taken as a whole, these analyses will provide insights into the emergence of new relationship forms and how similar they are across countries.

Email: Dr. Brienna Perelli-Harris: B.G.Perelli-Harris@soton.ac.uk|

Long-term trends of men\'s co-residence with children in England and Wales
Ursula Henz, London School of Economics

The past decade has seen a lively interest in the study of fatherhood. This interest was partly driven by concerns about an erosion of fatherhood in light of the increasing numbers of childless men and fathers with reduced or no contact with their children. These trends are also the basis for claims of a decreasing family involvement of men in the United States. This study is guided by the question whether Britain experiences similar long-term trends of men's family involvement. It examines age-specific shares of fathers living with their children for England and Wales by analyzing the National Statistics Longitudinal Study (LS) and the Labour-Force Survey (LFS). It presents period comparisons for ten-year intervals from 1971 to 2010 as well as cohort comparisons for ten-year birth cohorts from 1930-39 to 1970-79 and will examine variations in the observed patterns between different socio-demographic groups, i.e. by social class, education and ethnicity. All analyses are also carried out for women to see whether declining shares of adults living with their child is particular to men.

Email: U.Henz@lse.ac.uk|

Educational differentials in the effect of economic context on first union formation
Jorik Vergauwen, Karel Neels, Department of Sociology, University of Antwerp

The recent economic recession has fostered research into the impact of economic context on union and family formation. Literature suggests that particularly young adults are very vulnerable as entry into the labour market is delayed under adverse economic conditions. With career maturity being hampered by adverse economic conditions, union formation is delayed given the lack of financial and social long-term prospects. In this paper we consider the effect of aggregate-level economic context on entry into unmarried cohabitation and entry into marriage. We additionally investigate how these effects are differentiated in terms of educational attainment. Since marriage is mostly characterized by high commitment and extensive investments, a deferring impact of economic uncertainty on first marriage is hypothesized for all educational levels. Adversely, non-marital cohabitation is often considered as a strategy to deal with economic uncertainty. The effect of economic downturn is therefore assumed to be smaller or positive for entry into unmarried cohabitation. This coping strategy is considered to be more frequent among the lower educated. The analyses use union histories for eight European countries drawn from the Generations and Gender Survey. Multilevel discrete-time event-history models are used to assess the impact of economic context on entry into first unions between 1970 and 2005. Harmonized unemployment rates provided by the OECD are included as a macro-level indicator of economic uncertainty. Our results suggest that first marriage is particularly hampered by economic uncertainty in younger age groups. A similar negative effect of economic uncertainty is not found for entry into unmarried cohabitation among the higher educated, whereas lower educated are found more likely to enter unmarried cohabitation under adverse economic conditions.

Email: Jorik Vergauwen: jorik.vergauwen@ua.ac.be|

Early childhood education and care in England: deprivation and quality of services Ludovica Gambaro, Kitty Stewart, Jane Waldfogel, London School of Economics

This paper uses English administrative data on enrolment in early childhood education and care (ECEC) services to examine how variations in children's background interact with quality of services.  We show that children from more deprived areas are enrolled mainly in publicly provided services. When examining differences in staff qualification level and inspection outcomes, the study reveals that children from disadvantaged areas attend services staffed by more highly trained workers, but they are nonetheless less likely than their more advantaged peers to be in a setting whose inspection outcome indicates high quality.

Email: Ludovica Gambaro: L.F.Gambaro@lse.ac.uk|

What is the role of grandparents in enabling first-time lone mothers to participate and stay in employment in the UK?
Shireen Kanji, University of Basel

Raising employment rates of lone mothers has been a key policy target of recent years in the UK, as it has been in a number of other countries. Despite a significant increase in lone mothers' employment overall, a very low proportion of women who start out as lone mothers work after the birth of their first child. The question addressed in this research is how do these mothers, operating under intense constraints, manage to support work and care. The focus of the analysis is the contribution of help from grandparents in raising the employment participation of both lone and partnered mothers. The data for the analysis come from the Millennium Cohort Study, a nationally representative survey of families with children born in the UK around the time of the millennium. The paper analyses how grandparents' care, conceptualised as arising from personal support in the form of childcare, gifts and regular financial assistance, raises the participation of lone and partnered mothers. I measure the effect of grandparents' care using a biprobit estimation technique, in which I instrument grandparents' care in the work participation probit equation with other variables that are assumed to be correlated with grandparent care but not with mother\'s participation in the labour market. The analysis investigates how grandparents assist lone mothers to persist in what is often very precarious employment. The results show that grandparent care not only raises lone mothers' participation significantly but is even more important in enabling the few lone mothers who persist in work in being able to do this.

Email: Dr. Shireen Kanji: shireen.kanji@unibas.ch|

Trust and fertility dynamics
Arnstein Aassve, Francesco C. Billari, Bocconi University; Léa Pessin, Pompeu Fabra University

In this paper we argue and show that the different fertility trends in advanced societies are in part driven by differences in trust. The argument builds around the ideas that 1) trust is associated with a benign assessment of society and with economic development and 2) trust implies that individuals and couples are willing to outsource traditional family activities to other individuals outside their own family. Trust is therefore seen as a catalyser for the process of increased female labour force participation, the diffusion of childcare facilities, and hence a halt to the continuing fertility decline. Support of this hypothesis is drawn from the World Values Survey and European Values Survey. We present evidence both from country-level fixed effect estimation and from a series of multi-level analyses. We find that trust by itself is positively associated with fertility during the recent decades. Moreover, trust has an important effect, interacting with women’s education. In particular, as higher education for women has expanded, which traditionally is seen as a robust predictor for lower fertility, trust is a precondition for achieving higher fertility.

Email: Professor Francesco Billari: francesco.billari@uniboccini.it|

Uptake of antidepressant medications in couples after divorce: Convergence or divergence in couples
Christiaan Monden, University of Oxford; Pekka Martikainen, University of Helsinki;Niina Metsa-Simola, University of Helsinki; Saska Saarioja, University of Helsinki

On average divorce seems to have negative health effects on individuals. At the couple level little is known about the pattern of divorce effects. Does a divorce affect both partners equally? Or is most often only one partner negatively affected but not the other? Or are divorce effects purely individual? Using Finnish registry data we examine these questions by comparing longitudinal changes in antidepressant use in 4558 couples divorcing over the follow-up period to 108637 intact couples. Antidepressant use increases during the process of divorce and the effect for the two partners appears to be independent. Among continuously married couples, however, there is a clear pattern of convergence: If one partner starts to use antidepressants this increases the likelihood of medication use in the other partner. This suggests that the effects of divorce are mostly individual.

Email: Dr. Christiaan Monden: christiaan.monden@nuffield.ox.ac.uk|

Education and marital dissolution: does marital satisfaction explain the gradient?
Diederick Boertien, University of Pompeu-Fabra; Juho Härkönen, Stockholm University

The relationship between education and divorce has recently reversed in several countries with the less educated men and women being currently more likely to divorce. This carries potential implications for class and gender inequalities in family life, well-being, and children's life chances. However, little is known about why the least educated currently have lower family stability. Two main explanations can be identified in the literature. First, those with less education could have lower marital satisfaction. Second, those with less education could have lower thresholds to divorce. Relationship dissolution means that a person also gives up access to their partners' resources. If spouses possess many resources, the drop in marital satisfaction required to motivate a person to divorce might therefore be higher. Empirical evidence so far has not provided evidence that the educational gradient in divorce is explained by variation in marital satisfaction or variation in exit thresholds by educational groups. We use data from the British Household Panel Survey to analyze the trends in marital satisfaction among educational groups and the risks of divorce in these groups given satisfaction levels. We find that marital satisfaction trajectories are practically identical for people with distinct levels of education. The hypothesis that lower educated people have worse relationships therefore did not hold. Instead the educational gradient in divorce in Britain seems to be explained by differences in exit thresholds from relationships. Additional analysis will be presented to investigate what explains variation in exit thresholds by education.

Email: Diederick Boertien: diederik.boertien@upf.edu|

Does culture affect divorce decisions? Evidence from European immigrants in the US
Delia Furtado, University of Connecticut; Miriam Marcén, University of Zaragoza; Almuneda Sevilla, Queen Mary London

This paper explores the role of culture in determining divorce decisions by examining country of origin differences in divorce rates of immigrants in the United States. Because childhood-arriving immigrants are all exposed to a common set of US laws and institutions, we interpret relationships between their divorce tendencies and home country divorce rates as evidence of the effect of culture. Our results are robust in controlling for several home country variables including average church attendance and GDP. Moreover, specifications with country of origin fixed-effects suggest that divorce probabilities are especially low for immigrants from countries with low divorce rates who reside amidst a large number of co-ethnics. Supplemental analyses indicate that divorce culture has a stronger impact on the divorce decisions of females than of males pointing to a potentially gendered nature of divorce taboos.

Email: Almuneda Sevilla: almudena.sevilla@gmail.com|

Welfare entry and exit after separation among Australian mothers
Anna Zhu, Bruce Bradbury, Social Policy Research Centre, University of New South Wales

This paper uses Australian welfare program data to examine the impact of separation on welfare entry and the patterns of subsequent welfare exit for nearly 55,000 low and middle income mothers. These patterns are differentiated across mothers with different demographic characteristics such as age, number of children, birthplace, marital status, state and the socioeconomic status of their neighbourhood. We differentiate exits from welfare associated with repartnering from those due to increases in the mothers' own income. We find that after separation, mothers rapidly increase their likelihood of receiving welfare by around 60 percentage points. After this, their income support receipt steadily decreases, by about 7 percentage points per annum. The increase in income support receipt is lower among women who had higher rates of welfare receipt prior to separation, including younger mothers, those with four or more children, those in de facto marriages, the overseas-born, and those living in more disadvantaged areas. However, we find that these mothers are more likely to exit from welfare at a slower pace, suggesting that they are less equipped to escape the impoverishment of separation. This paper also includes a duration model summarising the multivariate impact of the different correlates on rates of exit from welfare. This model will contribute to previous research by controlling for pre-separation patterns of welfare, differentiating exits from welfare associated with re-partnering from those due to increases in the mothers' own income and disaggregating by demographic and socio-economic circumstance of mothers.

Email: Anna Zhu: anna.zhu@unsw.edu.au|

Is the boomerang generation of young adults a real phenomenon? Returning home in young adulthood in the UK
Juliet Stone, Ann Berrington, Jane Falkingham, ESRC Centre for Population Chage, University of Southampton

Young adults in the UK have tended to leave the parental home earlier than their European peers but are also more likely to return. There has been discussion in the British media about the relative importance of postponement in first leaving and increased returns to the parental home. The British Household Panel Study includes individuals from 5500 households interviewed annually from 1991 to 2008. We pool data for individuals aged 16-34 with data from two consecutive waves who are living away from their parents in the first wave (t0), then calculate the proportion returning one year later (t1). We use individual-level and contextual variables in binary logistic regression analyses to investigate determinants of returning. We additionally use a cohort-based approach including respondents aged 16-17 years and living with their parents at baseline. We follow them for five years to investigate patterns of leaving and returning to the parental home. The paired-years analysis shows men are more likely to return home than women but over time, women have become more likely to return. 'Turning-points' such as union dissolution and becoming unemployed show strong, positive associations with returning home. The cohort-based approach indicates young adults living in a family with two natural parents are least likely to leave the parental home. For those with step-families or living with a lone parent, we see more complex patterns of leaving and returning, indicative of more chaotic pathways out of the parental home.

Email: Dr. Juliet Stone: j.stone@soton.ac.uk|

Kin support in the context of childbearing postponement
Alice Goisis, London School of Economics

The past decades have been characterized by a considerable postponement of childbearing behaviours. Given that women who postpone childbearing come, on average, from higher educational and social strata, children of older mothers are expected to benefit from this process. However, evidence presented by the "weathering" hypothesis literature suggests that ethnic minority children do not necessarily benefit from their mothers' older age at (first) birth, suggesting that the process and outcome of postponement could be seen as qualitatively different for subgroups of the population. In light of these differences, it is important to consider whether any other aspect of the family sphere, in addition to/or rather than socioeconomic status, alters as maternal age at (first) birth increases. For example, knowledge of whether patterns of extended family ties change with increasing maternal age at birth remains elusive. This is an important oversight as diminished participation in extended kin networks could have negative consequences for ethnic minority mothers and their children as it could be associated with more stressful and difficult pregnancies and experiences of child rearing. This paper intends to investigate whether in the U.K. patterns of social support change with increasing maternal age at birth and whether they change differently among ethnic groups. This will be done by using the Millennium Cohort Study and employing a Latent Class Analysis approach.

Email: A.Goisis@lse.ac.uk|

Birth order and occupational choice
Feifei Bu and Maria Iacovou, ISER, University of Essex

The puzzle of birth order has intrigued scientists for a long time, with the earliest scholarly discussion dating back to the 1870s. Since then researchers across several disciplines have examined the effects of birth order on various outcomes, including nutritional status, IQ, educational performance, earnings, personality, attitudes, and delinquency. However, until recently, there has been no empirical study of the relationship between birth order and occupational status or interest (an outcome which is arguably related to both education and personality). The current study is aimed to fill this academic gap. Our data are taken from the British Household Panel Survey (BHPS) and National Child Development Study (NCDS). Based on Holland's (1973/1997) RIASEC occupational scheme, we have found that in general first-borns are more likely to work in "investigative" occupations; while later-borns are over-represented in highly-educated occupations. Similar patterns are also found in the occupational interest of adolescents. The results are robust to both cross-family comparisons and sibling matching methods, and for both the data sets we consider.

Email: Feifei Bu: fbu@essex.ac.uk|

Family form as cultural assimilation: Variations of the extended household by ethnicity and immigration generational status
Berkay Özcan, London School of Economics & Tim F. Liao, University of Illinois at Urbana-Champaign

A century-long decline in extended households has been followed by a recent rise in the US, which is attributed to immigration. Yet, no researchers have studied the changing trends as a process of immigrant assimilation. Conceptualizing the change in household composition as a cultural assimilation, we study the process of acculturating one’s family structure over immigration generations. We analyze data from the 1960 and 1970 US censuses and the 1994, 1999, 2004, 2009 Current Population Surveys, via the hierarchical age-period-cohort model that properly estimates age, period, and cohort effects. We answer three research questions: (1) whether family structure is assimilated over three immigrant generations (2) whether certain immigrant groups in terms of origin experience a slower process in such assimilation (3) whether the tempo of such assimilation would be faster for immigrants with a higher degree of education. We can also obtain an estimated period trend, controlling the effects of immigration.

Email: Dr. Berkay Ozcan: B.Ozcan@lse.ac.uk|