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)

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)

Provides recommendations to aid teachers (elementary and secondary) in planning tests, constructing item types (essay, short answer, true-false, or other alternate response, matching, multiple-choice), administering tests, scoring tests, analyzing tests, and using test results. (KH)

Data were collected from 117 college students to examine the validity of the Self-Directed Learning Readiness Scale. Two assumptions were generated from the results: (1) most items on the scale are correlated with the subject's total score and (2) most items are not correlated with the subject's age. (CH)

For measuring the face validity of a test, Nevo suggested that test takers and nonprofessional users rate items on a five point scale. This article questions the ability of those raters and the credibility of the aggregated judgment as evidence of the validity of the test. (JAZ)

The purpose of the present paper is to derive linear equating methods for the common item nonequivalent populations design from explicitly stated congeneric type test score models. The equating methods developed are compared with previously developed methods and applied to five professionally constructed examinations administered to approximately…

The purpose of this paper is to propose a general multicomponent latent trait model (LTM) for response processes. It combines the linear logistic LTM with the multicomponent LTM. Joint maximum likelihood estimators are presented for parameters of the general multicomponent LTM and an application to cognitive test items is described. (Author/BW)

The microcomputer version of the Ohio Vocational Interest Survey (OVIS II) differs from the machine-scored version in its ability to incorporate data from the OVIS II:Career Planner in its printed report. It differs from the hand-scored version in its ability to include data from the OVIS II:Work Characteristic Analysis in its printed report.…

Suggests that the English-as-a-foreign-language field would benefit by using the same types of tests as those of "Graded Objectives in Modern Languages," which are used in British schools to test foreign language learning. Presents test items which are typical of the British tests. (SED)

A supplementary treatment is proposed which helps identify sources of bias affecting groups of test items. This treatment is illustrated with the tranformed item-difficulty method as applied to an evaluation of a test used to help select applicants to be admitted to an aviation training program. (Author/BW)

Suggestions for software design in computer managed testing (CMT) cover instructions to testees, their physical format, provision of practice items, and time limit information; test item presentation, physical format, discussion of task demands, review capabilities, and rate of presentation; pedagogically helpful utilities; typefonts; vocabulary;…

Compared items from psychologically naive vs. psychologically sophisticated item-writers vs. a standardized test (N=552). Results showed that nonpsychologists with no formal definition of the construct they were to measure were able to write items that were as valid as those elicited from psychologists. (BH)

Two strategies for assessing item bias are discussed: methods comparing item difficulties unconditional on ability and methods comparing probabilities of response conditional on ability. Results suggest that the iterative logit method is an improvement on the noniterative one and is efficient in detecting biased and unbiased items. (Author/DWH)

A paradoxical phenomenon of decreases in reliability as the number of elements averaged over increases is shown to be possible in multifacet reliability procedures (intraclass correlations or generalizability coefficients). Conditions governing this phenomenon are presented along with implications and cautions. (Author)

This document contains the framework and a set of recommendations for the 2005 NAEP mathematics assessment. It includes descriptions of the mathematical content of the test, the types of test questions, and recommendations for administration of the test. In broad terms, this framework attempts to answer the question: What mathematics should be…