Data from 787 international undergraduate students at an urban university in the United States were used to demonstrate the importance of separating a sample into meaningful subgroups in order to demonstrate the ability of an English language assessment to predict the first-year grade point average (GPA). For example, when all students were pooled…

Correlations between predicted averages and actual grades in four universities were compared with those based on current data. A loss on the average of .10 occured in the accuracy of the correlation. The author recommends that prediction studies should not be done every year but on an as-needed basis. (Author)

Examined the reliability, construct validity, and predictive validity of the Non-Cognitive Questionnaire (NCQ) for different racial groups, using a sample of 1,975 Black and White college students. Results showed the NCQ significantly predicted persistence for Black students and predicted grades for both groups. (JAC)

The reliability and the predictive and concurrent validity of the MATAP were investigated with the implicit goal of improving the prediction of course grades in the College of Fine and Applied Arts. It was found that reliability and validity coefficients were low, and it was suggested that the scoring system was a source of error variance. (MS)