Abstract
The highly stratified high school system in contemporary Japan leads to a correspondence between the social structure and choice of high school. To explain this link, the author focused on the relative risk aversion hypothesis proposed by Richard Breen and John H. Goldthorpe (1997) and the downward educational mobility aversion hypothesis proposed by Toru Kikkawa (2006), and tested the validity of these hypotheses in the context of Japan using data obtained from second year students of Japanese senior high schools and their mothers in 2002 (N= 545). The analysis showed that fathers? occupations and parental education have direct effects on the rank of high school attended, even after controlling for economic resources and average grades. Moreover, the effects of average grade on the rank of high school attended differed between fathers? occupations and between levels of parental education, the pattern of which was expected by the hypotheses. Another analysis indicated that fathers? occupations and parental education affected the choice of the type of high school in a way that was predicted by the estimated occupational and educational returns to the type of high school. These analyses confirmed the validity of the relative risk aversion hypothesis and the downward educational mobility aversion hypothesis in explaining socio-economic differences in the choice of high school in Japan.
