The purpose of the study was to examine test bias and the "non-effects" of schooling. Teachers were given a list of words selected from standardized vocabulary tests and asked to indicate the words they had taught. The words were classified by the grade level at which they were first introduced. Ninety-five third-grade students in four schools were given two 17-item subtests: subtests of items the students did and did not have an opportunity to learn. The difference in the scores on the tests was used as an index of school effects of student vocabulary. Achievement test bias (when equated with content validity) was present. Students scored higher on the subtest containing items they had had an opportunity to learn. (Author/RC)