Factor analysis of comprehensive examinations

In working toward new, integrated curricula, dental schools have often failed to concurrently revise their evaluative techniques to be more compatible with the objectives of the curriculum. One question of major importance is whether the schools are totally aware of their students' capabilities or are graduating students proficient in some fields while lacking in others. Is the student's overall or total score satisfactory even though deficiencies may exist in certain areas? This latter situation is a real possibility when the simple sum of subtest scores is used as the primary means in assessing students. Using factor analysis, the scores on the subtests of eight dental school comprehensive examinations (total of 1,023 subjects, on first-, second-, and third-year examinations given between December 1970 and May 1972) at the University of Pittsburgh School of Dental Medicine were examined. The results, culled from the correlation matrices, Chi-square analysis, and factor matrices, indicated that the subtests did not form a homogenous unit (one, general factor), and hence that use of the total score is questionable.