Abstract
In this paper we identify the lines along which social ties between high-school teenagers are primarily formed. To this end, we introduce interaction weights between pupils in the same school class that are a function of exogenous individual background characteristics, such as gender, ethnicity, and having older siblings. The resulting model with endogenous interactions and school-specific fixed effects is estimated using data from the Dutch National School Youth Survey (NSYS), a survey in which, in principle, all pupils in a sampled class are interviewed. By combining the 1992, 1996, 1999, and 2001 NSYS data, we are able to identify trends in social relationships of teenagers. We find that the roles that gender and ethnicity play in how teenagers interact varies strongly across different types of behaviour. For example, 'going out' shows strong within-ethnicity interactions, while expenditures on cell phones and on clothing exhibit mainly between-girls interactions. Having older siblings has a minor effect on within-school-class social interactions. There is weak evidence of decreased ethnic segregation within school classes during the decade considered.
Original language | English |
---|---|
Pages (from-to) | 373-391 |
Number of pages | 19 |
Journal | Oxford Review of Economic Policy |
Volume | 21 |
Issue number | 3 |
DOIs | |
Publication status | Published - 2005 |
Keywords
- MOBILITY EXPERIMENT
- NEIGHBORHOOD
- EDUCATION
- BEHAVIOR
- GHETTOS