In this study I investigated the change in effects of family background and study experience
variables on middle school students' achievement level in Seoul between 1988 and 2007 by
comparing respective data. For this purpose I established the research model and analysed data using
the HLM statistical model, especially growth modelling. In this process I controlled related variables
which could affect the students' achievement results, for example gender; SES; attitude toward school
education; private tutoring experiences; educational aspiration level; anti-school behaviors; school level
variables, including mean SES of each school; and etc.
The results of this analyses are as follows. First, in the case of the 1988 data, SES variables
explained 8.9% of the achievement variance. And in the case of the 2007 data, SES variables
explained 10.1% of it. The ratio of explaining variance of study experience variables was 17.9% in
the case of the 1988 data but 4.7% in the case of the 2007 data. As a result, during last 20 years
the effect size of SES variables on achievement level has grown, but study experience variable's
effect has decreased gradually.
Second, the effect size of father's education level variable in 2007 was 0.175, and this value is
twice the 1988 data result. In the case of private tutoring experience, in the 2007 data the effect
size was 0.138 and in the 1988 data it was -0.072. But the educational aspiration level's effect of
the 2008 data was smaller than that of the 1988 year data.
According to the results of this study, the effect of SES on achievement level in 2007 was higher
than it was in 1988. But the effects of student's socio-psychological variables(attitude toward school
education, educational aspiration level and anti-school behaviors) was less important than in 1988. So
I can conclude that over the last twenty years the Korean education system has moved in a less
egalitarian direction. But more data is needed in order to find out the reasons behind this trend.