Presents the procedures and results of an item analysis on the Watson- Glaser Critical Thinking Appraisal Form Ym, based on the responses of 200 graduating high school students in South Australia. (LC)

Discusses the design of effective questionnaires based on the concept of a hierarchical classification of Levels of Information: global (Level I), general concept (Level II), and specific (Level III). By tying question types to the information desired, the entire questionnaire can be improved. (JJD)

The mathematics required to calculate the asymptotic standard errors of the parameters of three commonly used logistic item response models is described and used to generate values for common situations. Difficulties in using maximum likelihood estimation with the three parameter model are discussed. (Author/JKS)

The internal reliability of the Speech Sounds Perception Test was examined in two independent samples. Both indicated a significantly greater number of errors on the first half of the test; the largest number of errors occurred on Subtests B and A. Reliability coefficients and a performance classification formula are discussed. (Author/CM)

Sixteen approaches for detecting item bias were compared on samples of Black, White, and Chicano elementary school pupils using the Lorge-Thorndike and Raven's Coloured Progressive Matrices tests. Recommendations for practical use are made. (JKS)

An extension of the Rasch model for partial credit scoring of test items is presented. An unconditional maximum likelihood procedure for estimating the model parameters is developed. The relationship of this model to Andrich's Rating Scale model and Samejima's Graded Response model are discussed. (Author/JKS)

In many personality and interest inventories, a score profile, rather than a single score, is attributed to each subject. Six applicable criteria are suggested for use in examining the adequacy of items in such inventories. These criteria relate to the items' response distributions, internal consistency, and discriminative value. (Author/BW)

Presents a table of discriminatory values for selecting test items based on mean differences in responses between upper and lower scorers in an normally distributed population. Provides a means for classifying items as poor, fair, or good. Describes a simple method for rank ordering accepted items. (Author/MJL)

Describes a practical approach that teachers can use to check for test-item validity in test construction. The Kuder-Richardson Reliability Formula is used. Detailed instructions describe the procedure for evaluating items for difficulty and using statistical methods to determine test validity. (AM)

Concludes that inferential reading test questions are as likely to be passage dependent as factual test questions. (MKM)

Lists five objectives in creating an item bank for language proficiency test items and describes the design of an item bank card. Briefly discusses item analysis and population distribution in this system and points out the numerous advantages of the system and conditions for its success. (PMJ)

Technical problems in achievement testing associated with using latent structure models to estimate the probability of guessing correct responses by examinees is studied; also the lack of problems associated with using Wilcox's formula score. Maximum likelihood estimates are derived which may be applied when items are hierarchically related.…

The computer program WEIGAN makes the weighted G analysis available for computer users. The input and output of the program are described. (Author/JKS)

Wilcox has described three probability models which characterize a single test item in terms of a population of examinees (ED 156 718). This note indicates indicates that similar models can be derived which characterize a single examinee in terms of an item domain. A numerical illustration is given. (Author/JKS)