The advantage of daughters in hypogamous families: parental heterogamy and educational outcomes among children of highly educated parents

ABSTRACT

Increasing female educational attainment across OECD countries is making hypogamy a widespread phenomenon. This trend provides an opportunity to re-examine the effects of educational assortative mating on children's educational outcomes. This research explores the effects of hypergamy, homogamy, and hypogamy on gender differences in children's expectation of university graduation and actual college graduation. For the first purpose, logistic regression with country fixed-effects is applied to individual-level data from PISA 2015; a similar analysis is carried out for the second purpose with data from the European Social Survey. Three characteristics make us expect higher female advantage among children of hypogamous couples: higher probability of mothers being the main family breadwinner; higher probability of gender value conflict, eventually leading to family breakup and the father's absence; and the possibility that the father's occupation discourages sons from pursuing higher education. A systematic female advantage is indeed found among children of hypogamous couples in terms of expectation of college graduation and actual college graduation. Among the possible mechanisms behind this female advantage, only the father's and the mother's occupation could be explored with the data at hand, but none of them explain this advantage.

KEYWORDS:

• Gender
• educational expectation
• college graduation
• hypogamy
• hypergamy
• homogamy

Acknowledgements

This paper profited from the comments received at the conference of the European Consortium for Sociological Research held at the University of Lausanne in September 2019. I am also grateful to Diederik Boertien and Daniel Oesch for their insightful comments and support. This research was carried out with the financial support of the Spanish Agency of Research (Agencia Estatal de Investigación, AEI), project number: FFI2016-CSO2016-80399-R (AEI/FEDER, EU).

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 Certainly, the normalisation of single-mother families can alleviate challenges for both daughters and sons in these families. We do not discuss here the effect of single motherhood on sons and daughters as a function of how much single motherhood goes against social norms, but rather its singular effect on boys as a function of father's absence. As single motherhood becomes more frequent in society, so might its subsequent effect on boys relative to girls.

2 Contrary to this line of reasoning, it could be argued that men who select themselves into hypogamous couples are likely to be more gender egalitarian than the norm in their social environment. At the end of the day, they accept female partners with an educational attainment higher than theirs. As an exploratory analysis, we carried out ordinal logistic regressions taking two statements included in the European Social Survey (ESS) that are explicitly aimed at capturing gender egalitarianism (‘Women should be prepared to cut down on paid work for the sake of family’ and ‘Men should have more right to a job than women’) as dependent variables. The disagreement with these statements was regressed to the interviewee's gender and type of couple he/she belongs to. As we can see in Table A3 (Annex), men in both hypergamous and hypogamous couples are less likely to disagree with non-egalitarian statements than men in homogamous couples (reference category). Women's attitudes are in line with men's only in the case of hypergamous couples: they are less likely to disagree with non-egalitarian statements than women in homogamous couples (reference category). This is not the case with hypogamous couples; here, women tend to disagree with non-egalitarian statements even more than their peers in homogamous couples (reference category). In sum, the divergence in gender egalitarianism values (and thus the likelihood of a conflict around gender egalitarian values) seems higher in hypogamous couples than in the other two types of parental couples.

3 The control includes three categories: native, second-generation immigrant (if the student was born in the OECD country included in the analysis but her/his parents were born abroad), and first-generation immigrant (if she/he was also born abroad).

4 PISA 2012 does provide information on the relative with whom the interviewee is living (father, mother, both), but the question about his/her own educational expectations was not formulated. PISA 2009 has information on both family structure and the student's educational expectations, but the latter is only available for a reduced number of countries (14). Although PISA 2015 also lacks information on family structure, it has information on the student's educational expectations for a relatively large number of countries.

5 The individuals interviewed for the ESS may have problems recalling their father's and mother's educational and occupational attainment when they were 14 years old. Certainly, the number of missing values in the initial variables for father's (10.1%) and mother's education (6.7%) is not negligible. Yet, the distribution of these missing values is concentrated among low occupational categories, which leads us to suspect that those missing values actually conceal low levels of either father's or mother's education (results available upon request). Even if the results of the analysis of college graduation (ESS) should be taken with more caution for this reason, we believe the core analysis with ESS data (for individuals with at least one college-graduated parent) is not as affected by the absence of information in the variables capturing father's and mother's education as it would be for the general sample.

6 An alternative analysis considering the detailed education of the member of the couple who is not a college graduate (father in the case of hypogamous couples; mother in the case of hypergamous couples) did not modify the effect of relative parental education reflected in Table 1. Results are available upon request.

7 Due to the gender wage gap, it is reasonable to assume that family resources in families in which the mother is a college graduate and the father is not (hypogamy) are lower than in families where the father is a college graduate and mother is not (hypergamy).

8 This average marginal effect of gender (14%) corresponds to an actual probability of expecting university graduation being slightly higher than 60% for daughters of hypogamous couples and below 50% for their sons. Results are available upon request.

9 Direct transmission of businesses from fathers to sons may also become a negative incentive for male offspring to raise expectations of university graduation if this family business does not require university training. Unfortunately, the information on parental employment in PISA 2015 does not distinguish across fathers’ employment status (employee, self-employed).

10 A robustness check with ISEI similar to the one carried out for educational expectation was not possible for college graduation, first, because ISEI is not readily available for the interviewee's father and/or mother in the ESS and, second, because the father's and mother's occupations are registered with ISCO 1-digit codes, which makes the computation of ISEI scores almost impossible.

11 Ten-year birth cohorts were created from 1929 to 1997. The lower bound of this period is due to the fact that the presence of a number and variation in terms of types of parental couples is negligible for those born before 1920. The upper bound, however, was established in 1997 because there is no individual with at least 23 years of age born after 1997 in the cumulative sample. Given the scarcity in the number of cases, the first cohort exceptionally includes individuals born between 1920 and 1939.

Additional information

Funding

This research was carried out with the financial support of the Spanish Agency of Research (Agencia Estatal de Investigación, AEI), project number: FFI2016-CSO2016-80399-R (AEI/FEDER, EU); Spanish Ministry of Economy and Competitiveness (Spanish Research Agency).

Notes on contributors

Luis Ortiz-Gervasi

Luis Ortiz-Gervasi is Associate Professor of Sociology at Universitat Pompeu Fabra (Barcelona). Before joining the Department of Political Sciences & Sociology of this university, he was Visiting Research Fellow at the Industrial Relations Research Unit (University of Warwick, UK) and Lecturer at the London School of Economics & Political Science (LSE) and Universidad de Salamanca. His current research interests deal with educational mismatch, social and gender inequalities in the transition from school to work and inequality of educational opportunities in access and progression within higher education. He has been principal research of a research project on inequalities in access and progression within higher education, funded by the Spanish Research Agency (CSO-2016-80399-R) and he is currently starting a research project on gender and social determinants of sustainable employment in times of crisis (PID2020-119800RB-I00), funded by the same institution.