A1 Refereed original research article in a scientific journal
Parental education, class and income over early life course and children's achievement
Authors: Jani Erola, Sanni Jalonen, Hannu Lehti
Publisher: ELSEVIER
Publication year: 2016
Journal: Research in Social Stratification and Mobility
Journal acronym: RSSM
Volume: 44
First page : 33
Last page: 43
Number of pages: 11
ISSN: 0276-5624
eISSN: 1878-5654
DOI: https://doi.org/10.1016/j.rssm.2016.01.003
Web address : http://www.sciencedirect.com/science/article/pii/S0276562416300038
tVery few studies on intergenerational achievement consider the high correlation between separatemeasures of parental socioeconomic position and possible life course variation in their significance forchildren. We analyse how socioeconomic characteristics of mothers and fathers over children’s life courseexplain children’s occupational outcomes in adulthood. Using Finnish register data, we matched the occu-pational position (ISEI) of 29,282 children with information on parents’ education, occupational classand income when children are 0–4, 5–9, 10–14, 15–19, 20–24 and 25–29 years old. We fitted three-levelrandom effects linear regression models and decompose family-level variance of siblings’ ISEI by eachmeasure of parental status. We show that parental education explains family variation in siblings’ occu-pation most and income explains it least. Status characteristics of fathers together explain approximatelyhalf of children’s outcomes, and those of mothers explain slightly less. These explanations vary only alittle during children’s life course. We also find that independent, non-overlapping effects of observedparental indicators vary over time. Mothers’ education explains independently most in infancy, whereasthat of fathers in early adulthood. The influence of class alone is minor and time constant, but the effect ofincome alone is negligible over the entire follow-up. The independent effects are overall relatively small.The largest proportion of children’s outcomes explained by these parental measures is shared and cannotbe decomposed into independent effects. We conclude that bias due to ignoring life course variation instudies on intergenerational attainment is likely to be small.