The relationship between each of several dimensions of the expectancies of fifth through eighth grade black urban pupils and their school achievement was examined. While anticipated differentiation on racial items was not seen, two important factors were found: general personal efficacy and interpersonal control. (Author/JKS)

Measures of locus of control, intelligence, and academic achievement were administered to a sample of male juvenile delinquents. All intercorrelations among the three tests were positive and significant. However, when the effects of intelligence were partialed out, achievement did not vary significantly for different levels of locus of control.…

A relatively reliable and valid hierarchy of clusters of objects is plotted from the highest column entries, exclusively, of a matrix of interassociations between the objects. Having developed out of a loose definition of types, the method isolates both loose and highly definitive types, and all those in between. (Author/RC)

Previous knowledge of the difficulty index and the intercorrelations of the items will allow group results to be predicted and manipulated. A compound bionomial probability function of a testscore is established for which a computer program has been written. Three item selections and the appropriate probability distributions are given which give…

Compares the harmonic mean and Kramer unequal group forms of the Tukey test for various: (a) degrees of disparate group sizes, (b) numbers of groups, and (c) nominal significant levels. (RC)

Given a sample of responses to a pair of questionnaire items with interval scale values it is sometimes of interest to know the degree to which respondents select the same response for both items. The coefficient of equality measures the departure from independence in the direction of equality. (RC)

Shoemaker's conclusions related to the influence of various data base characteristics (reliability, variability of item difficulty indices, and degree of skewness in the normative distribution) on the standard error of a mean estimated via multiple matrix sampling procedures are examined. (Author/RC)

Examines the similarity between the coordinates which resulted when correlations were used as similarity measures and the factor loadings obtained by factor analyzing the same correlation matrix. Real data, a set of error free data, and some computer generated data containing deliberately introduced sampling error are analyzed. (RC)

It is shown that the rank-biserial correlation coefficient is a linear function of the U-statistic (Mann and Whitney), so that a test of group mean difference is equivalent to a test of zero correlation for the rank-biserial coefficient. (RC)

Compares various item option scoring methods with respect to coefficient alpha and a concurrent validity coefficient. Scoring methods compared were: formula scoring, a priori scoring, empirical scoring with an internal criterion, and two modifications of formula scoring. The empirically determined scoring system is seen as superior. (RC)

Joreskog's model of cogeneric tests is used to analyze agreement between raters. Raters are treated as measuring instruments. The model of cogeneric tests, of which classical parallelism and tau-equivalence are shown to be special cases, is applied to teachers' ratings of students' responses on open-end questions. (Author/RC)

A procedure is presented for the analysis of rating data with correlated intrajudge and uncorrelated interjudge measurement errors. Correlations between true scores on different rating dimensions, reliabilities for each judge on each dimension and correlations between intrajudge errors can be estimated given a minimum of three raters and two…

Systematically investigates the effects of test context on verbal analogy item difficulty, in terms of both simple percentage correct and easiness estimates from a parameter-invariant model (Rasch, 1960). (RC)

The diversity among interpretations of single items in personality questionnaires has been noted previously. Using adjectives from the Adjective Check List (ACL), the study sought evidence bearing on these questions: Does such diversity make the responses to an item not comparable across subjects? If so, what are the implications for scores based…

Component analysis provides an attractive alternative to factor analysis, since component scores are easily determined while factor scores can only be estimated. The correct method of determining component scores is presented as well as several illustrations of how commonly used incorrect methods distort the meaning of the component solution. (RC)