Changes and constancies in the importance of 57 situations and conditions in four stages of Army life are presented and interpreted following Maslow's analysis of the functioning of human needs. Ratings of importance were transformed to z-scores and plotted. Three patterns were observed. It is suggested that: (1) items remain stable because other…

Research on the effectiveness of graphical displays for information acquisition and retention lacks a system for classifying graph information and generating test items to assess learning. The purpose of this study was to validate a system based on two types of information and three types of informational units. Results of an analysis of variance…

Test development efforts for constructing 35 of the tests included in a battery of 56 tests of possible cognitive abilities are described. Data were collected on 172 boys and 210 girls who had just completed the fifth grade. Item and total score statistics obtained for the 56 tests in the total battery are presented and discussed. (Author)

The Rasch model and other latent trait models encounter some difficulty when faced with an appreciable amount of missing data or omitting behavior. The present note assumes that some reasonable missing data model has been formulated which does not involve the parameters associated with the latent ability of interest. A maximum likelihood function…

A method of typological analysis was applied to computer-generated 96-item questionnaire data for 100 cases, under a variety of conditions to analyze both the item-level and score-level. The results showed a considerable advantage of score-level approach in the number, size, and replicability of clusters recovered. (DEP)

In revising the Metropolitan Achievement Tests, the publisher attempted to balance the presentation of sexes and to portray both sexes in less stereotypic roles. This study examined whether the revision conformed to these guidelines. Results indicate the present test content is significantly better balanced and less stereotypic than previous…

Many expectations concerning performance on educational and psychological tests are sex-stereotyped. Females, for example, are expected to do better on verbal tests, while males are expected to do better on mathematical tests. These expectations are being reevaluated in terms of the cultural influences that may underlie them. (Author)

Hoyt's analysis of variance procedure for estimating reliability assumes that the residual mean square estimates error variability. If, however, an individual's true score varies across items, it is argued that residual mean square estimates two components--error and interaction--and hence Winer's modification of Hoyt's formula, understimates the…

A procedure for obtaining Rasch model estimates of item difficulty and of ability is detailed. The procedure approximates the optimal but difficult to obtain "unconditional" estimates. (JKS)

ANALYSIS, a FORTRAN computer program developed as an aid to investigation of and instruction in the psychometric properties of test items, is described. The program utilizes both correlational and analysis of variance approaches to classical item analysis. (Author/JKS)

Option weighting is an alternative to increasing test length as a means of improving the reliability of a test. The effects on test reliability of option weighting procedures were compared in two empirical studies using four independent sets of items. Biserial weights were found to be superior. (Author/CTM)

Eight recent General Certificate of Education (GCE) examinations, containing mainly free-response questions, were investigated in terms of their marking reliability. The tests of 200 randomly selected candidates from each subject were re-marked by a senior GCE examiner, and these marks were compared with the marks awarded previously as a result of…

The reliabilities yielded by several differential option weighting scoring procedures were compared among themselves as well as against conventional testing. It was found that increases in reliability due to differential option weighting were a function of inter-item correlations. Suggestions for the implementation of differential option weighting…

A two-step weighted least squares estimator for multiple factor analysis of dichotomized variables is discussed. The estimator is based on the first and second order joint probabilities. Asymptotic standard errors and a model test are obtained by applying the Jackknife procedure. (Author)

The present study replicated the procedure of Kausler et al. (1975) as a means of testing further the hypothesis that the processing of wrong items differs qualitatively, as well as quantitatively, from the processing of right items. (Author)