Examined the effects of test length and sample size on the alternate forms reliability and equating of simulated mathematics tests composed of constructed response items scaled using the two-parameter partial credit model. Results suggest that, to obtain acceptable reliabilities and accurate equated scores, tests should have at least 8 6-point…

This overview of the role of error and tolerance for error in measurement asserts that the generic precision associated with a measurement procedure is defined as the root mean square error, or standard error, in some relevant population. This view of precision is explored in several applications of measurement. (SLD)

Calculated year-1 to year-2 stability of assessment data from 21 states and 2 Canadian provinces. The median stability coefficient was 0.78 in mathematics and reading, and lower in writing. A stability coefficient of 0.80 is recommended as the standard for large-scale assessments of student performance. (SLD)

Empirical reliabilities of scores based on item-pattern scoring, using 3-parameter item-response theory and number-correct scoring, were compared within each of 5 score metrics for at least 900 elementary school students for 5 content areas. Average increases in reliability were produced by item-pattern scoring. (SLD)

An extension of the Rasch model is used to obtain objective measurements for examinations graded by judges. The model calibrates elements of each facet of the examination on a common log-linear scale. Real examination data illustrate the way correcting for judge severity improves fairness of examinee measures. (SLD)

Studied four forms of score resolution used by testing agencies and investigated the effect that each has on the interrater reliability associated with the resulting operational scores. Results, based on 120 essays from the Georgia High School Writing Test, show some forms of resolution to be associated with higher reliability and some associated…

A taxonomy of 43 rules for writing multiple-choice test items is presented, based on a consensus of 46 textbooks. These guidelines are presented as complete and authoritative, with solid consensus apparent for 33 of the rules. Four rules lack consensus, and 5 rules were cited fewer than 10 times. (SLD)

The consistency between raters over three years of a high-stakes performance assessment was examined in two studies involving a total of approximately 3,000 students in grades three, five, and eight. Results show that raters in different years differ in severity, with raters in mathematics most consistent, and those in language arts least…

When low-stakes assessments are administered, the degree to which examinees give their best effort is often unclear, complicating the validity and interpretation of the resulting test scores. This study introduces a new method, based on item response time, for measuring examinee test-taking effort on computer-based test items. This measure, termed…

In low-stakes testing, the motivation levels of examinees are often a matter of concern to test givers because a lack of examinee effort represents a direct threat to the validity of the test data. This study investigated the use of response time to assess the amount of examinee effort received by individual test items. In 2 studies, it was found…

Scores from 324 Louisiana schools on the Louisiana Graduation Exit Examination and a within-school split sample of 255 schools indicate that a single subject or grade provides a less consistent and more narrow perspective on school effectiveness than a subcomposite made up of 2 subject areas. (SLD)

Examined the validity and reliability of the 1995-96 reading/language arts portfolio assessment developed in the Dallas (Texas) public schools for prekindergarten through second grade. Ratings by 42 teachers show that portfolio contents do not provide a valid sample of student work and the assessment reliability is low. (SLD)

Several extensions to the Angoff method of standard setting are described that can accommodate characteristics of performance-based assessment. A study involving 12 panelists supported the effectiveness of the new approach but suggested that panelists preferred an approach that was at least partially conjunctive. (SLD)

Two studies involving a total of 99 experts examined the reproducibility of standards for 2 medical certifying examinations set under different conditions. Together, results of both studies provide evidence that a modified version of the Angoff method is quite reliable and produces stable results under varying conditions. (SLD)

Seventeen educational measurement textbooks were reviewed to analyze current perceptions regarding true-false achievement testing. A synthesis of the rules for item writing is presented, and the purported advantages and disadvantages of the true-false format derived from those texts are reviewed. (TJH)