Two questions of scholarly and public policy interest concerning the well-d
ocumented racial difference in scores on achievement tests are How much of
the racial difference ("gap") can be attributed to social-class differences
between blacks and whites? and How much has the racial gap changed over th
e past 30 years? To address these questions, the authors analyzed evidence
from seven probability samples of national populations of adolescents from
1965 to 1996 and found that black-whiter differences in achievement are tar
ge and are decreasing slowly over time, About a third of the gap in test sc
ores is accounted for by racial differences in social class, a-nd although
this gap appears to have narrowed since 1965, the-rate at which it is narro
wing seems to have decreased since 1972. The two groups are becoming more e
qual at the bottom of the test-score distribution, but at the top, blacks a
re hugely underrepresented and are approaching parity with whites slowly, i
f at all.