This study deals with dropout in Japanese higher education. First, it reviews three basic models which have been developed in order to examine dropout in American higher education. Then, it shows various completion rates in colleges and universities which are calcurated in dividing the number of graduates by the number of freshman of four years ago. Finally, this study measures how much educational conditions and size of institution have effect on dropout.
Major findings can be summarized as follows; 1. The completion rate is higher in two-year institutions than in four-year institutions.
2. National universities show higher completion rate than other public universities, and private universities show the lowest rate.
3. Female students are more likely to graduate than male students.
4. The completion rates are higher in the study fields such as the humanities and education, and lower in the social sciences and engineering.
5. The result of regression analysis shows that the worse the educational condition (in other words, larger student/teacher ratio; larger average class size; larger part time/full time faculty ratio), the more the number of dropout.
6. The three variables of educational conditions being controlled, the larger the size of institution, the fewer the number of dropout.
7. The dropout rate, defined as the number of dropout divided by the number of the enrollment, tends to be lower in the institutions with larger student/teacher ratio and in the larger institutions.
8. The three variables of educational ocnditions being controlled, the larger the institutional size, the lower the dropout rate.
These results confirm both the “college impact model” and “chartering model” which has been developed by sociologist John Meyer.