The problem under investigation was to determine the differential effects of three types of treatment on students' test-taking skills. The study also investigated whether test-taking instruction would result in significant mean test score gains on a standardized reading test for white and black, high and low socioeconomic status (SES) third-grade children. The sample consisted of 275 white and black, high and low SES third-graders from six elementary public schools in a midwestern urban school system. Generally, the results indicated that the test-taking skills of some students can be significantly increased through systematic instruction. The extent of increase appeared to depend on such variables as SES, race, and sex. The data indicated that the apparent gains in test-taking skills did not transfer in terms of expected equivalent levels of performance on the Gates reading test. (Author/RB)