A critical review of the Expressive One-Word Picture Vocabulary Test (Gardner) is offered. The reviewer feels that the instrument cannot be recommended in its present form. Further research concerning the manual, and theoretical issues, (particularly test-retest stability) is strongly recommended. (Author/PN)

Handicapped and nonhandicapped students were administered the Bruininks-Oseretsky Test of Motor Proficiency to determine regular or specially designed physical education placement. Two of the three functions on the test were significant, indicating usefulness in placement. Fewer than half the test items for each function contributed discriminatory…

Subjects were immediately tested on a reading passage, and received feedback after 30 minutes or one day, or no feedback. After a week, subjects identified their original responses to three types of test items. One-day delay of feedback gave better results than the 30-minute delay. (GDC)

This article discusses an experiment carried out with university English as foreign language students to test the validity of multiple choice examinations for reading comprehension. (CFM)

The results of a series of simulation studies designed to investigate the influence of guessing and item pool characteristics on the bias, accuracy, and information properties of the trait estimates derived from Owen's Bayesian sequential testing strategy are reported. (RC)

Using item response theory to investigate differential item functioning (DIF), students' expected course grades were examined and found to function similarly across sex and race. These grades were incorporated into the matching criterion, enhancing the validity of subgroup comparisons for the third-grade mathematics test taken by 1,050 students.…

The agreement of diagnostic classifications from two parallel subtests assessing a mathematics skill with three levels of scoring was studied with 431 Arab Israeli 10th graders. Results indicate that, even when parallel form reliability is high, less agreement is apparent when performance is evaluated at the micro level. (SLD)

Different methodologies for linking tests across languages are reviewed and evaluated, focusing on monolingual item response theory, bilingual group designs, and matched monolingual group designs. These methods, although not without weaknesses, are superior for promoting score comparability than methods that rely on translation or expert judgment…

Examined the longitudinal invariance of method effects associated with negatively worded items on a self-report measure of global self-esteem. Data from the National Educational Longitudinal Study for 3,950 junior high school and high school students show that the method effects associated with negatively worded items exhibit invariance across…

Assumptions of a servocontrol model of test item feedback were tested in a study of 94 junior and senior high school students receiving feedback or no feedback with 2 retention intervals. Results support the assumption that response certitude is related to the learner's ability to comprehend a given item. (SLD)

Examines the ways in which computers can assist teachers in developing good tests. Describes the program TESTWORKS in detail and provides charts comparing this program with 11 others in the areas of price, type of questions generated, computer functions, and the usefulness of each. Discusses the use of word processors and databases. (KO)

How computers are used to analyze item data is reviewed, and the information that existing item-analysis programs provide is described. Summaries of studies comparing the performance of some of these packages reveal some of their current limitations. Emphasis is on the usefulness to educational practice of these packages. (SLD)

A study of gender bias in the Dental Admission Test's mathematics test and its validity in predicting dental school success found no significant difference between male and female performance and no significant difference in the predictive validity of items favoring males or females. (Author/MSE)

Compares the effect of content ordering and scrambled ordering on examinations in courses, such as economics, that require quantitative skills. Empirical results suggest that students do no better if they are given a content-ordered rather than a scrambled examination as student performance is not adversely affected by scrambled ordered…

The effects of using item response theory (IRT) ability estimates based on customized tests formed by selecting areas from a nationally standardized achievement test were examined. For some populations, in some conditions, IRT ability estimates can be equivalent to scores based on full-length tests. (SLD)