Three methods of item selection were used to select sets of 38 items from a 50-item verbal analogies test and the resulting item sets were compared for internal consistency, standard errors of measurement, item difficulty, biserial item-test correlations, and relative efficiency. Three groups of 1,500 cases each were used for item selection. First…

Six approaches to assessing test item bias were examined: transformed item difficulty, point biserial correlations, chi-square, factor analysis, one parameter item characteristic curve, and three parameter item characteristic curve. Data sets for analysis were generated by a Monte Carlo technique based on the three parameter model; thus, four…

Transformed item difficulties, chi-square, item characteristic curve (icc) theory and factor score techniques were evaluated as approaches for the identification of biased test items. The study was implemented to determine whether the approaches would provide identical classifications of items as to degree of aberrance for culturally different…

After tests are administered, they are scored and the scores are given back to the students. If the real purpose of the test is to improve student learning, simply returning the scores is not sufficient. The first step in evaluating test results is to be sure that the test has tested the intended concepts and content. Calculating the mean and the…

A study was conducted to measure faculty perceptions of academic standards and the level of academic intensity in transfer courses. Questionnaires were sent to chief academic officers (CAO's) at 30 community colleges, asking them to distribute five instruments to members of the academic senate and five to instructors who were not members of the…

The initial study was based on responses of 103 examinees on an objective classroom test. Results indicated that response changes increased item difficulties as well as biserial and point biserial correlations. Test reliability was not decreased. This implies that not all students benefit from changing initial responses. The personal biserial…

Test results from the Florida Educational Assessment of third and fifth grade communications and mathematics skills were used to compare linear and Rasch equating results. The samples consisted of over 5,000 cases for each grade and content area. The tests contained some items common to both the 1976 and 1977 test forms, but no fewer than 20…

An equating study was conducted on the Certifying Examination for Primary Care Physician's Assistants to compare the ability of current examinees with the standardization group and to determine if current test items are more difficult than previous items. Using 46 common items from the multiple choice section, the 1978 exam was equated to the 1976…

This report examines how selection fairness is influenced by the item characteristics of a selection instrument in terms of its distribution of item difficulties, level of item discrimination, and degree of item bias. Computer simulation was used in the administration of conventional ability tests to a hypothetical target population consisting of…

In a study of the detection of test item bias, Chinese speakers and Spanish speakers were administered a measure of pronunciation accuracy for the (d) and (z) morphemes. Six indices were calculated for each item: item difficulty, point biserial correlation coefficient, item variance, chi-square, Rasch logit of difficulty, and the average…

This paper focuses on the questions which arise as test practitioners monitor score scales derived from latent trait theory. Large scale assessment programs are dynamic and constantly challenge the assumptions and limits of latent trait models. Even though testing programs evolve, test scores must remain reliable indicators of progress.…

To explore the potential relationships among general attitude toward school, evaluative ratings of courses and instructors, and selected respondent characteristics, an attitude inventory, and, for each class, a course and instructor rating scale, with items which emphasized the nature of student-teacher interaction and level of course difficulty,…

We describe the item modeling development and evaluation process as applied to a quantitative assessment with high-stakes outcomes. In addition to expediting the item-creation process, a model-based approach may reduce pretesting costs, if the difficulty and discrimination of model-generated items may be predicted to a predefined level of…

TAILOR is a computer program that uses the implied orders concept as the basis for computerized adaptive testing. The basic characteristics of TAILOR, which does not involve pretesting, are reviewed here and two studies of it are reported. One is a Monte Carlo simulation based on the four-parameter Birnbaum model and the other uses a matrix of…

Technical improvements, which may be made in the reliability and validity of tests through confidence scores, are discussed. However, studies indicate that subjects do not handle their confidence uniformly. (MS)