Due to a concern regarding the validity of Chicago public elementary schools standardized test scores on the Iowa Tests of Basic Skills, a test audit was conducted. The audit was limited to 40 schools. Two schools were selected from each of the 20 administrative districts, with each school testing one seventh- and one eighth-grade class. These schools did not constitute a representative sample but were chosen because of possible anomalies. Other schools were retested to lend geographical balance to the sample and to provide an estimate of score change in a retest situation. There were a total of 23 suspect and 17 comparison schools. Nineteen of the 80 classrooms had median score declines that met or exceeded the set cutpoints. Of these classrooms, 17 were in suspect schools. It was suspected that exceeding the publishers' time limits was the most common way of increasing scores. Also, instances of altered answer sheets were detected using a computer program. However, district superintendents' investigations revealed no irregularities that would account for the score drops. Tentative plans for the spring 1985 testing provide for an increased emphasis on test and answer sheet security, adherence to time limits, and inservice training for testers. (PN)