The scores of handicapped students taking tests such as the Scholastic Aptitude Test (SAT) or the Graduate Record Examinations are flagged so that admissions officers will be aware that they were achieved under special circumstances. A series of studies was initiated to determine whether special administrations of such tests are comparable to standard administrations, in which case flagging would no longer be necessary. The studies looked at comparability data for test takers with hearing impairments, visual impairments, physical handicaps, and learning disabilities. Comparability between standard and nonstandard test forms was found to be high, particularly with respect to characteristics as reliability, factor structure, and differential item difficulty. Analysis of the tests' predictive validity with regard to academic performance found that there was little over- or under-prediction for the great majority of handicapped students. The SAT did, however, substantially overpredict college performance for learning-disabled students, and this overprediction was exacerbated by time extensions during test administrations. The need for flagging test scores may be eliminated by establishing comparable timing criteria for special test administrations or by rescaling nonstandard test administrations according to how handicapped students performed in school. The comparability study also examined admissions decisions, test content, and testing accommodations. (JDD)