In this analysis, with samples of 1970, 1975, 1980, and 1985 Korean Population and Housing Census, trends in the effects of ascriptive factors, such as father's education, father's occupation, and a place of residence, on access to university and college education were examined. Two different models-linear regression model and logistic regression model-were employed and compared in order to distinguish between changes in the distribution of educational opportunities across social groups and changes in the degree of the association of ascriptive factors with the allocation of educational opportunities to individuals. The effects of the place of residence on educational attainment were found to diminish over time in both linear and logistic regression analyses. By comparison, while the effects of father's education and father's occupation have declined over time in the linear regression analysis. they did not indicate any downward trend in the logistic anaysis. Korea is the case where the persistence of family background effects on the allocation of university and college education has offset the increasing diffusion of educational opportunities to youths from disadvantaged families. While differentials in the absolute years of schooling between youth from advantaged and disadvantaged families have decreased and the proportion of youths from disadvantaged families among university entrants has increased, the comparative benefits of those youths from advantaged families have been persistent over time end even increased according to some evidence.