Achievement tests that are maximally sensitive to effective instruction in reading and mathematics for grades 2 and 5 were developed and refined. Important considerations regarding the tests' validity were: its coverage of instructional content (opportunity to learn), and its reactivity to instruction. Student ability must be minimally related to change in performance, and when the opportunity to learn is high, that factor must be maximally related to performance in order to have a test that is useful in studies of teaching effectiveness. An ability by opportunity to learn matrix was constructed for every item included in the test batteries. The data for each subtest are included, as well as correlations between the subtests. Sample test items are included in Appendix A. (Author/GDC)