This is an interesting report because it was looking at specific teaching practices among children of different racial groups and seeing which practices were the most effective.
"Mr. Wenglinsky focused on 20 such practices in math, ranging from the time classes spent on instruction in the subject to the topics covered in class. He also took into account other characteristics that might lead to achievement differences, such as whether students were poor enough to qualify for federally subsidized lunches or whether their parents subscribed to newspapers and magazines.
Some classroom practices, Mr. Wenglinsky found, correlated with better test scores for all 4th graders, regardless of race or ethnicity. The more time students spent learning math, for example, the better they did on the tests. Students whose teachers emphasized geometry and routine mathematics exercises also scored higher than students in classes where those practices were emphasized less.
Like measurement, estimation, and data analysis, other practices seemed to benefit particular groups of students disproportionately. African-American students fared worse than either Hispanic or white students, for example, in classes where teachers gave frequent tests.
Mr. Wenglinsky said the differences he found were substantial. Black or Hispanic children in classrooms that emphasized some of the practices he identified as key for their groups scored one-quarter to one-third of a grade level higher than their peers in other classes. Taken together, he said, such practices could erase the achievement differences between minority and white students in the same schools.
We found the report at the Education Policy Analysis Archives.
Closing the Minority Achievement Gap