Educational Expectations and Gender in Contemporary Japanese High Schools:Analyzing Interaction Effects between School Type and Educational Level
スポンサーリンク
概要
- 論文の詳細を見る
This article examines factors determining high school studentsʼ educational expectations in modern Japan. Previous research studying educational attainment or expectations reports that womenʼs educational decisions are not differentiated by meritocratic factors, i.e. academic achievement or the school type they belong to as strongly as men. Some researches argue that recently, however, the educational decision process of women is becoming closer to that of men due to the increasing importance of influences from the role of meritocratic factors. They furthermore state that declining direct effects of social background on education also contribute to the increment of commonality between menʼs and womenʼs educational decision processes. Educational expansion is one important factor leading to these changes, since rising numbers of women enrolling in university makes womenʼs subsequent educational choices after graduation from high school similar to those of men. In this article, I discuss in greater detail the factors causing changes in the educational decision process over time, and state that a shift in womenʼs educational choices from short-term college to university, which occurs in academic high school, and womenʼs advance to the general academic high school, play an important role. As a consequence of these processes, I predict that high school type effectively differentiates current female studentsʼ educational expectations, and when we focus on a particular educational level, a similar structure to men is then observed. Concretely, school type appears to be an important determinant of womenʼs educational expectations when we examine whether it has an influence on womenʼs plans of going to university, but not when on going on to other forms of higher education.<BR><BR>The PISA 2003 data set was analyzed to verify the validity of above prediction. The respondents for this research were collected by a two-stage random sampling design, and were nationally representative of 15 years old students in Japanese high school at that time. In the analysis, I use the statistical method known as the cumulative logit model with partial proportional odds assumption. The difference between this method and the ordinary proportional odds version, or the ordered logit regression model, is that this model allows the effects of more than one predictor vary with the ordinal category of an outcome variable: in other words, we can estimate the interaction effects between a predictor and a threshold term; whereas the ordinary version assumes that the effects of all predictors are same for each logit for different cumulative probabilities. Thus, we can examine whether the effects of a particular predictor vary depending on a different category of outcomes by using this applied method. Results show that school type has a strong influence on studentsʼ expectations of going to university for both women and men, but this influence is remarkably weak for that of going to short-term college or other higher education for women, as predicted. In addition, the direct effects of social background on educational expectations are highly significant for both men and women, whereas those effects are slightly stronger for women than for men.