The purpose of the study was to examine standardized college entrance examination test scores to determine if these may be appropriately used to predict long-range academic success.
The design of the study was based on the procedure developed by Gulliksen and Wilks (1950) in which errors of estimate, slopes, and intercepts for the regression of a criterion variable on a single predictor, or on a best weighted combination of predictors, were compared sequentially. The finding of a significant difference between groups in the standard error of estimate would mean that the groups differ with respect to amount of error in predicting average college grades. If the standard errors of estimate did not differ significantly, the slopes of the regression lines were then tested. If they were found not to differ, the intercepts were compared, and failure to find a significant difference in intercepts indicated that the same regression equation may be used to predict grades for both groups.
Level of Degree
Individual, Family, and Community Education
First Committee Member (Chair)
Ronald E. Blood
Second Committee Member
Third Committee Member
Richard E. Lawrence
Scott, Carmen Casillas. "Long-Term Predictive Validity of College Admission Tests for Anglo, Black, and Mexican American Students." (1976). https://digitalrepository.unm.edu/educ_ifce_etds/89