Comparison of the Validity of Two Temperament Scales in Predicting College Achievement

Abstract

THE 32-item temperament scale devided by Gough (4) has been shown to be moderately valid predicting college Quality Point Average (1, 5), but its inter correlations with other scales suggest that it is an impure measure of several independent traits. It correlates positively with the ACE test, and negatively with Taylor's Manifest Anxiety scale, while the ACE and MAS scales are re latively unrelated (5). In addition, both the ACE and Hr scales correlate with QPA while the MAS does not(l). At least two traits appear to be measured by these tests: an achievement related trait that is measured well by the ACE, moderately by the Hr, and not measured by the MAS, and a second achievement unrelated trait that is measured well by the MAS, moderately by the Hr, and is not measured by the ACE to any appreciable extent. The first trait is probably the common verbal factor found in most achievement tests, but the second trait is not necessarily the anxiety trait presumably measured by the MAS since we do not know whether the MAS is a unidimensional test. The new Need Achievement scale in Edwards Personal Preference Schedule (3) provides another tool for identifying personality traits that may be related to college achievement. Presumably scores on the Need Achievement scale should be positively related to college achievement, but it is not known whether this scale (a) is valid in predicting QPA, or (b) measures the same tempera ment traits measured by the Hr scale. The present study was designed to provide further information on the traits measured by the Hr and Need Achieve ment scales and the relationship of the traits measured to college achievement.