Exemption from College Freshman Composition: The College Level Examination Program (CLEP) General Examination in English Composition and the American College Testing Program (ACT) English Usage Test

Abstract

Problem. This study investigated exemption from freshman composition. Its two primary purposes were, first, to examine the validity and cutting scores of the oriqinal CLEP General Examination in English Composition (GE:E), a test used nationwide to grant exemption; and, second, to measure the impact freshman composition or exemption from it has on GPA. A comprehensive review of the literature on the GE:E revealed that some researchers had confidenoe in the validity of the GE:E and cutting scores based on its norms whereas others expressed doubt and suspicion. The literature on the ACT Enqlish usage test (ACT:E) was reviewed to provide a basis for comparing the efficacy of the GE:E. The results of correlation studies on the test scores and GPA's for the two were typical for such tests and showed them to be similar in validity. The limits and deficiencies of correlation studies and their role in test choice and use are discussed. Procedure. To provide additional, unique information on the GE:E, two small (N=22) but homogeneous samples were drawn from the same population, fall quarter 1975 Mankato State University (MSU) freshmen. The samples were not representative of the MSU or ACT and CLEP populations. Sample A students' CLEP-GE:E soores had exempted them from English 101 Composition I. Sample B students took the course. The two samples were matched in gender, major, and composition ability as measured by the ACT:E. The two criteria used to determine the validity of the GE:E and the appropriateness of its cutting scores were GPA and the ACT:E. Pearson product-moment correlation coefficients were used to test the null hypotheses that stated there were no re1ationships (p<.05) between ACT:E scores and GPA's and GE:E scores and GPA's for Sample A. The same procedure was used to test the null hypotheses for the relationships between ACT:E scores and GPA'S for Sample B. Analysis of covariance was used to test the null hypotheses that stated there were no differences between the mean GPA's of the two samples. In addition, Sample A students were mailed a questionnaire that asked for their opinions on exemption and the GE:E. Sample B's questionnaire asked them their opinion of freshman composition. Findings. The correlations of test scores with GPA's revealed extremely low correlations of GE:E scores with GPA and low to moderate correlations of ACT:E scores with GPA. Because the correlations for the GPA for the ACT:E were greater than those for the original GE:E, it had greater predictive and content validity than the GE:E for these samples. Two of the mean GPA's for the samples were not statistically different but two others were statistically higher for Sample A. For Sample A the GE:E had predictive validity and the cutting scores were appropriate. Composition course work had no measurable impaot on GPA for Sample B. In their questionnaire responses Sample A expressed satisfaction in exemption and doubt in the GE:E. In their questionnaire responses Sample B was supportive of freshman composition with qualifications. Conclusions. If the same serious oharqes that were made against the original GE:E in the review of the literature can be made against the revised edition, it could be rejected for lacking content and predictive validity and for not being normed properly. Although both Sample A and Sample B were generally supportive of freshman composition, the content of the course and the quality of the instruction need to be scrutinized. Reoommendation. This researoher recommends that the MSU English department conduct a thorough validation study of the ACT:E. If it has validity, a multiple-regression equation based on high sohool GPA and ACT:E scores could be developed to predict MSU composition grades. This equation and a faculty constructed and graded essay examination could be used to grant or waive credit in Composition I and II. If the ACT:E is invalid, a thorough study of another standardized test of composition ability might be conducted.