The impact of revisions in the content of the Scholastic Assessment Test (SAT) and changes in the score scale on the predictive validity of the SAT were examined. Predictions of freshman grade-point average (FGPA) for the entering class of 1994 (who had taken the old SAT) were compared with predictions for the class of 1995 (who had taken the new…

Whether a set of predictor variables could be identified from pre-enrollment and post-enrollment data that would differentiate students who advance to a major in engineering from those who do not was studied with students at Auburn University, Alabama. Also studied was whether predictors could be isolated that would identify students likely to…

The analytically derived expected asymptotic standard errors (SEs) of maximum likelihood (ML) item estimates can be predicted by a mathematical function without examinees' responses to test items. The empirically determined SEs of marginal maximum likelihood estimation/Bayesian item estimates can be obtained when the same set of items is…

The purpose of this investigation is to study the differential characteristics of regressions of collegiate marks on aptitude test scores and secondary school marks in various demographic groups at an Ivy League college. (Author)

In a study to determine whether learning disabilities can be predicted from kindergarten teachers' anecdotal reports, 37 boys later diagnosed as learning disabled, were selected and their kindergarten class identified. (Author/MC)

Reports the results of an investigation of the accuracy of self-evaluation on test performance, in which junior high school students were asked to predict their scores before and after taking each unit test. Some students learned to improve their evaluations of self-performance on cognitive tasks without formal instruction. (JR)

In a nonmetric multiple-cue probability learning task involving 2 binary cue dimensions, it was found that Ss can learn to use configural or pattern information (a) when only the configural information is relevant, and in addition to the configural information, one or both of the cue dimensions are relevant. (Author/RK)

Previous puzzling inconsistencies in results of experimental comparisions of paired-associate anticipation with recall (study-test) methods are indicated by Izawa's 1971 data to reflect longer inter- and/or intratrial intervals in experiments showing marked superiority of the recall method. (Editor)

In light of the importance given to the consideration of individual difference variables in the identification of client-treatment compatability, thhe present study examined the locus of control variable as a predictor of individual preference for type of group counseling. (Author)

The classic view of traits as dispositions was examined and a number of ambiguities noted. When clarified, implication for predicting social behaviors from personality variables were derived. (Editor)

The purpose of this study was to compare five predictor models, including two least-square procedures, two probability weighting (semi-Bayesian) methods, and a Bayesian model developed by Lindley. (See also TM 501 088, TM 501 089, and TM 501 090) (Author/NE)

The present study evaluates the relationship between social learning and developmental theories in the prediction of changes in cognitive behavior. (Author)

The purpose of this research was to determine whether unfair test discrimination was occurring for yet another employee population - one which might be described as culturally disadvantaged, and to examine the evidence for test bias using multiple regression procedures. (Author/RK)