The Course Attitude Questionnaire (CAQ) has been administered 100,000 times in 3,500 classes at the Pennsylvania State University since 1964 with extensive use in modified form at other institutions. The purpose of this research was to investigate the reliability and internal validity of the CAQ and the temporal stability of its original factor structure. CAO responses were obtained from 5,297 Penn State students in 272 classes during a 1973 term. These data were item-analyzed and submitted to principal-components and common-factor analyses with orthogonal and oblique transformations. Coefficient alpha reliabilities were computed for the factor-derived scales. The large normative data base was also inspected. The factor structure was not replicated by a "fit' of the derived factors to the CAQ-keyed subscales. Most factors were highly correlated and their reliabilities appeared to be a function of redundant item content. A "general" dimension pervaded the instrument, accounting for much of the total and common-factor variance. A trend toward inflation and skewness in scores was noted for the 10-year period. The CAQ) subscales should be interpreted with caution and systematically reviewed for currant validity. The CAQ, originally included in the Appendix of this document, has been removed because it does not reproduce. For related document see ED 029508. (Author/RC)