This study compared the reliability and validity indexes of randomly parallel tests administered under inclusion, exclusion, and correction for guessing directions, using 54 graduate students. It also compared the criterion-referenced grading decisions based on the different scoring methods. (TJH)

This study investigated characteristics differentiating life-long learners from low participants in learning among 195 human service professionals in southern New England. Discriminant function analysis of seven extracted factors and three demographic variables indicated significant differences distinguishing the two groups. (TJH)

This study investigated the distributional properties of the standardized residual that is commonly used in the Rasch model calibration program to develop various indices of item and person fit. The power of the standardized residual to detect measurement disturbances is also addressed. (TJH)

Profile analyses within multivariate statistical applications use "levels" and the "parallelism" hypotheses to test group differences. Procedures for the use of these hypotheses within a multistage discriminant analysis are illustrated using an artificial data set, with emphasis on the geometrical representation of group profile configurations.…

The degree to which tests of divergent production may be confounded by verbal ability and loquacity; and relationships among divergent thinking, social competency, and the syntactic complexity of prose written by 61 fifth and sixth graders were assessed. Complex thinkers may appreciate the value of simplicity in facilitating communication. (TJH)

The effect of item context on differential item performance based on gender on mathematics test items was studied, using 404 male and 375 female adults. The analyses were based on a modified one-parameter item response theory methodology. Gender differences emerged; however, they may be due to chance. (TJH)

Classical test theory methods are derived to allow statistical evaluation of all KT(KT - 1)/2 unique, non-trivial difference scores that occur by comparing one or more variables (K) at two or more testings (T) for a single individual (or object of measurement). Future research needs are discussed. (Author)

Cohen's kappa statistic is frequently used to measure agreement between two observers using categorical polytomies. Cohen's statistic is: shown to be inherently multivariate in nature; expanded to analyze ordinal and interval data; and extended to over two observers. A non-asymptotic test of significance is provided for the generalized statistic.…

Alternate versions of the Western Personnel Test were administered via computer and in the standard paper and pencil format to two groups of adults (N=50). Results suggest that the two forms of administration yield comparable results, and that separate norms for the computer-administered version are not necessary. (TJH)

The construct validity of extraversion-introversion was explored, as measured by the Myers Briggs Type Indicator (MBTI) and the Eysenck Personality Inventory. Findings supported the complexity of extraversion-introversion. Two MBTI scales, Extraversion Introversion and Judging Perceiving, were factorially valid measures of impulsivity…

A visual analogue mood scale was developed that included an option indicating subjects didn't understand the meaning of an adjective in an item. The item content significantly affected responses, raising questions about the adequacy of recently proposed affective taxonomies that were based on restricted samples of emotion-descriptive adjectives.…

The use of the arc-sine transformation in analysis of variance can lead to difficult inference situations and pose problems in interpretation. It can also produce tests of noticeably lower power when the null hypothesis is false, and is not recommended as a standard tool. Simulated illustrations are provided. (Author/GDC)

Either positive or negative test instructions and two item types (attitude or personality) were administered to undergraduates. Students were more likely to respond true when receiving positive instructions. Students receiving attitude test instructions were not more likely to respond true than students receiving personality test instructions.…

To control for response bias in student ratings of college teachers, an index of rater error was used that was theoretically independent of actual performance. Partialing out the effects of this extraneous response bias enhanced validity, but partialing out overall effectiveness resulted in reduced convergent and discriminant validities.…

Confirmatory factor analysis was used to reanalyze correlational data from selected variables in Guilford's Aptitudes Research Project. Results indicated Guilford's model reproduced the original correlation matrix more closely than other models. Most of Guilford's tests indicated high loadings on their hypothesized factors. (GDC)