RECENT RESEARCH IN COMPUTERIZED ADAPTNE TESTING OF MUSICAL APTITUDE WALTER VISPOEL University of Iowa Recent studies both within and outside of music have shown that computerized adaptive testing proce- dures can provide greater efficiency, reliability, and validity than conventional testing procedures. In this article, I focus primarily on procedures for testing musical aptitude. It is important to know the history and uses of conventional tests of musical aptitude as well as the differences between conventional and adaptive testing procedures. Recent studies comparing adaptive and conventional music tests yield further insights of value to music educators. History and Use of Conventional Tests of Musical Aptitude Musical aptitude traditionally is defined as poten- tial for learning music-related tasks. Formal tests of musical aptitude date back to the pioneering work of Carl E. Seashore at the University of Iowa and his Measures of Musical Talents (Seashore, 1919). Seashore assessed aural acuity in a variety of modes (pitch, duration, intensity, timbre, rhythm, and tonal memory), emphasizing the relationship between per- ceptual ability and musical achievement. His work sparked a sizable amount of subsequent research and inspired the development of several additional tests of musical aptitude (e.g., Bentley, 1966; Drake, 1957; Gordon, 1965; and Wing, 1961). The most notable of these is Gordon's Musical Aptitude Profile (1965), which several music testing experts acknowledge as the best musical aptitude test currently available (Buros, 1972). Gordon also has developed other tests for Walt" P. is proftssor in the Colkge of Education at Univrrsity of Iowa, Iowa City, IA 52242. SPRING/SUMMER 1993 young children (1978, 1982) and adults (1989), but the basic approach to musical aptitude testing remains unchanged, despite advancements in technology and testing theory. Since 1919, music educators have tended to use tests of musical aptitude more as research tools than as classroom assessment instruments. This may seem sur- prising, because music educators stand to gain much useful information from these tests. They could use the results to diagnose musical learning difficulties, detect discrepancies between achievement and apti- tude, and identify individuals who are likely to become excellent musicians. One reason why music educators have not used tests of musical aptitude more extensively has to do with the tests' shortcomings. Even the best commercially available tests are ineffi- cient and often show evidence of only low to moder- ate reliability and validity (Shuter-Dyson & Gabriel, 1981). Whellams (1971), for example, noted that stu- dents taking the Musical Aptitude Profile (MAP) waste 40 minutes of the 2.5 hours of administration time responding to items that do not discriminate among them with respect to the aptitudes tested. Efforts to shorten tests such as the MAP improved efficiency but lowered reliability and validity (Brown, 1969). These problems led some music educators to abandon available tests and to seek other methods of evaluation (Davies, 1978). Until the advent of com- puterized adaptive tests, however, few superior alterna- tives had emerged. Advantages of Adaptive Tests Adaptive tests improve efficiency, reliability, and validity because the administration of items is adjusted 39