This paper analyzes research findings about the role of values and beliefs in aptitude testing. Indications are found that the ongoing controversy over the lack of ethnic, racial, and sex fairness in aptitude testing is rooted in the tests themselves. Test designers, it is argued, are influenced by their socioeconomic position (they are usually members of the majority group), but they are also swayed by a desire for statistical elegance and scientific methodology. Recommendations are made: test makers must accommodate social differences, however difficult to quantify; and the scientific community should be more receptive and less defensive toward criticism about ability testing and show sensitivity toward the problems created by imperfect instruments. These aspects of aptitude testing receive particular attention: the significance of large group differences in scores found among different ethnic groups; the non-objective nature of many tests; the insensitivity of test makers to cultural differences; the use of test scores for job selection and admission to college; exaggerations about the usefulness of tests; reports that test limitations favor minorities; and the deliberately quantitative, empirical, and pragmatic nature of test methodology. (KH)