The testlet, a bundle of test items, alleviates some problems associated with computerized adaptive testing: context effects, lack of robustness, and item difficulty ordering. While testlets may be linear or hierarchical, the most useful ones are four-level hierarchical units, containing 15 items and partitioning examinees into 16 classes. (GDC)

This article suggests that mechanical application of Golden Rule-like procedures is inappropriate. The fundamental idea embodied in them, namely, that of taking issues of equity into account in test construction, may reasonably be done without doing violence to test validity. (JAZ)

This is a reprint of a 1986 letter by the former president of the National Council on Measurement in Education (NCME) to New York and California legislators. The author outlines why NCME is opposed to legislative initiatives to extend Golden Rule procedures to tests in those states. (JAZ)

Examines three errors and learning difficulties found to be typical in general chemistry, gives the source of each problem, and presents a method of correcting each problem. Suggests using analysis of examination errors for the purpose of altering instruction for the improvement of student learning. (CW)

Presented is an electricity problem on simple series circuit. Provided are the problem diagram and answers with explanation. (YP)

A method for ordering persons and items when all responses are ordinal was developed and applied to several sets of questionnaire data (from undergraduates) and one set of archeological data. The method provides a possible nonparametric treatment of data usually treated by more traditional psychometric methods. (SLD)

An experiment was conducted that contrasted a variant of computerized adaptive testing, self-adapted testing, with two traditional tests. Participants completed a self-report of text anxiety and were randomly assigned to take one of the three tests of verbal ability. Subjects generally chose more difficult items as the test progressed. (Author/LMO)

A taxonomic matrix classification was used to assess the curricular validity of the Stanford Achievement Tests for the mathematics textbooks used in a school district's fifth and sixth grades. Rasch item difficulty was also examined. Results indicated only small differences between textbooks. (GDC)

Using item responses from an in-training examination in diagnostic radiology, the application of a strength of association statistic to the general problem of item analysis is illustrated. Criteria for item selection, general issues of reliability, and error of measurement are discussed. (Author/LMO)

This paper discusses various issues involved in using the Rasch Model with multiple-choice tests and questions the suitability of this model for multiple-choice items. Results of some past studies supporting the model are shown to be irrelevant. The effects of the model's misfit on test equating are demonstrated. (Author JAZ)

A content-oriented analysis of written solutions to test items in Israeli high school graduation examinations in mathematics yielded six error categories: misused data; misinterpreted language; logically invalid inference; distorted theorem or definition; unverified solution; and technical error. (Authors/MNS)

Under the assumptions of classical measurement theory and the condition of normality, a formula is derived for the reliability of composite scores. The formula represents an extension of the Spearman-Brown formula to the case of truncated data. (Author/JAZ)

Argues that though college examiners seem to desire candidates' personal responses to literature, the examination questions require students to reply to the opinions of others rather than explain their own opinions. (SRT)

The test equating accuracy of content-representative anchors (subsets of items in common) versus nonrepresentative, but substantially longer, anchors was compared for a professional certification examination. Through a chain of equatings, it was found that content representation in anchors was critical. (Author/GDC)

Response changes of two classes of 30 dental hygiene students each were tabulated on multiple-choice questions on quizzes, midterms, and final examinations. Response changes were classified as wrong to right, right to wrong, or wrong to wrong. Significantly more responses were changed from wrong to right than from right to wrong. (Author/CT)