The Long Shadow: The Relationship of Family Background to Skills and Adult Outcomes

• In Event: The Impact of Skills on Life Chances and Social Mobility for Adults Around the World

Abstract

In many education policy studies, educational attainment is used as a proxy for skills. PIAAC survey results offer a valuable opportunity to study, in a global comparative context, the relationship between attainment and skills, as well as the relationship of other variables to both attainment and skills. The purpose of this presentation is to report on a study that examined the strength of the relationships of respondents’ parents’ education to respondents’ developed skills and other adult outcomes, holding their educational attainment constant. Here, parental education functions as a proxy for respondents’ home background. The focus centers on findings that if, for a given level of attainment (e.g. bachelor’s degree) there is a systematic relationship between parental education and an outcome (e.g. numeracy score), then that is an indication that education has not fully succeeded in compensating for differences in home backgrounds. From a policy perspective, interest lies in comparing (i) patterns within a jurisdiction across levels of attainment and (ii) within levels of attainment across jurisdictions.
The study utilized data from PIAAC 2012, focusing on respondents aged 25-55. Parental education was classified into three ordered levels. Variables studied included cognitive outcomes (literacy, numeracy, problem solving in technology-rich environments) and work outcomes (employment, wages, training opportunities). Auxiliary characteristics included respondents’ gender, age class, etc. Methods employed were cross-tabulations, ordinary regression and logistic regression. Analyses were conducted for all OECD jurisdictions that participated in PIAAC. Results demonstrated substantial differences in outcome levels among jurisdictions, but generally similar patterns in relationships. For example, parental education was systematically related to most outcomes in each attainment stratum, with the strongest relationships typically found in the lower attainment strata – suggesting that greater amounts of education do contribute to reducing inequality of outcomes. Generally, the above relationships were not much diminished by taking account of other respondent characteristics. Comparisons among jurisdictions, in conjunction with other relevant information regarding both social and educational contexts, can yield insights useful to both educators and policy makers. For example, it appears that the U.S. higher education system is less successful than the systems in other countries in compensating for differences in home backgrounds. Such finding can suggest directions for further research using national data bases.

Author

• Henry I. Braun, Boston College