INTERGENERATIONAL TRANSMISSION OF EDUCATION ACROSS EUROPEAN COUNTRIES AND COHORTS

ABSTRACT

This article investigates the intergenerational transmission of education in 28 European countries. The main aim is to answer the following questions: (1) To what degree are parents' education and the educational attainment of their children related in different countries and cohorts, and (2) how can we explain the country-cohort variation in these effects by looking at contextual characteristics? In order to explain this variation, we focus on the degree of industrialisation, female labour force participation, the structure of the educational system and the political ideology of a country. Regarding industrialisation, we do not only take its size into account, but also the pace of the development. Multilevel estimates on 76,821 individuals nested in 250 country-cohort combinations from three waves of the European Social Survey (2002–2006) show that we cannot fully support the prediction of decreasing intergenerational transmission of education solely under the influence of industrialisation. While being an important factor that positively influences the general level of schooling for a certain cohort in a particular country, its decreasing effect of the parents' education on their children's schooling is complemented by the other interacting contextual factors, such as female labour force participation and the quality of the school system.

Keywords:

• intergenerational transmission of education
• education
• comparative research
• industrialisation
• social mobility

Notes

1. It is beyond the scope of this research to also look more into detail at the gender differences in educational attainment. However, it is for further research to investigate the educational trajectories of boys and girls in the context of a gendered labour market.

2. For more information on the ESS, please refer to http://www.europeansocialsurvey.org/ where the data and documentation is freely available to download.

3. We used the Cook Distance and the SDbeta's in order to estimate the influence on the total estimates and parameter estimates separately. By excluding Turkey, the total estimates became more stable.

4. We also tested models where there was a distinction made between the education of the father and the mother, but the effects were the same. Therefore we decided to opt for the more parsimonious model.

5. For more information on the Historical Statistics of the World Economy: 1-2006 AD dataset, please refer to http://www.ggdc.net/maddison/ where the data and documentation is freely available to download.

6. Although we did not hypothesise influence on the country level, we also ran models with three levels of analysis (individuals [level 1] nested in country-cohorts [level 2] and country-cohorts nested in countries [level 3]). This three-level analysis only showed slightly different results.