The result indicated that the initial, unadjusted scores of girl schools are significantly higher than the coeducational schools as the literature predicted. The advantage of boy schools is not significant as expected. What these preliminary analysis show is that the difference in the average reading achievement of girls who attend single-sex as against
coeducational schools is more apparent than real. When adequate control is exercised for the social and ethnic background, parental involvement and investment and school climate, the initial advantage of girl schools still exist. However, when
school intake ability enter into the final model, the significant differences between girl-schools and coeducational schools disappear.
It can be argued that the popular belief of girls' better academic performance at single-sex schools is not sustained by the Hong Kong data reported in this paper. The results support the arguments of March (1989, 1992), Bell (1989), Young (1994) and Young & Fraser (1992) that comparisons made after adequate control for background and ability factors show no evidence of the academic superiority of either type of school in terms of their academic achievement.
In sum, it is not a question of 'whether single-sex education is preferable or more beneficial for women than is
co-education; rather, the concern is how each of these settings interacts with learning variables to influence
the creation of gender habitus and achievement behaviour. It appears that co-educational schools are characterized by masculine models of education, which determine the status hierarchy of subjects and pedagogical techniques are still essential factors in affecting the socialization of gender habitus, which are noteworthy to be explored in future studies.

Keywords

• Gender difference
• Single-sex and co-education Schools

(30 min. Conference Paper, English)