The effects of students taking different levels of the same standardized achievement test were assessed by administering two levels of the same test to each student. The functional level of the test was taken by all students. The second level of testing was randomly assigned at the adjacent higher or lower level of the test. Functional level testing is the method of using students' instructional level rather than age or grade level to assign achievement test levels. Test reliability and stability of scores from one test level to another were studied. Content validity was also a major factor in determining which test was suitable for use. Testing appeared to be more reliable in high achievement schools than in low achievement schools. Three of the five tests evaluated were determined to be suitable for Chicago elementary schools. Two of the tests were found to be less suitable because it was difficult to determine an accurate means of student placement. (DWH)