Do choice-type tests (multiple-choice, true-false, etc.) measure the same abilities or traits as free response (essay, recall, completion, etc.) tests? A large number of studies conducted with several different methodologies and spanning a long period of time have addressed this question. In this review, attention will be focused almost…

In the area of large scale assessment, there is increasing interest in the measurement of students' written performance. At issue is whether the task demands in writing assessment can be simplified to involve the production of paragraph-length writing samples and/or multiple choice testing, rather than full-length essays. This study considers data…

Use of latent class models to examine strategies that examinees (92 college students) use for a specific task is illustrated, via a multiple-choice test of spatial ability. Under an answer-until-correct scoring procedure, models representing an improvement over simplistic random guessing are proposed. (SLD)

Compares multiple choice testing with free response formats in terms of the cognitive operations each type demands. (HOD)

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)

Investigated measurement characteristics of both forms of Finding Embedded Figures Test (FEFT). College students (N=302) completed both forms of FEFT or one form of FEFT and Group Embedded Figures Test. Results suggest that FEFT forms provide reasonable reliable and valid data. (Author/NB)

A meta-analysis of the difficulty and discrimination of the "none-of-the-above" (NOTA) test option was conducted with 12 articles (20 effect sizes) for difficulty and 7 studies (11 effect sizes) for discrimination. Findings indicate that using the NOTA option does not result in items of lesser quality. (SLD)

We investigated the use of quizzes administered prior to lecture (i.e., prelecture quizzes) and compared them to no-quiz control groups. In previous research, the success of administering quizzes after covering a topic (i.e., postlecture quizzes) was contingent on the quizzes and the subsequent exams being of similar level and content. However,…

Despite the prevalence of multiple choice items in educational testing, there is a dearth of empirical evidence for multiple choice item writing rules. The purpose of this study was to expand the base of empirical evidence by examining the use of the "all-of-the-above" option in a multiple choice examination in order to assess how…

This study of survey research experts was conducted to determine desirable characteristics of mail questionnaires. The 82-item Likert-scale instrument used in the study covered general appearance, instructions, choice of items, choice of response options, wording, order of items, and item format. The instrument was administered to: 8 subjects who…

An alternate-choice test item is a simple declarative sentence, one portion of which is given with two different wordings. For example, "Foundations like Ford and Carnegie tend to be (1) eager (2) hesitant to support innovative solutions to educational problems." The examinee's task is to choose the alternative that makes the sentence…

Investigated the effect on student performance of changes in question structure and sequence on a GCE 0-level multiple-choice chemistry test. One finding noted is that there was virtually no change in test reliability on reducing the number of options (from five to per test item). (JN)

This study, one of a series designed to answer practical questions about performance based assessment, examined the comparability of school scores on short, nonparallel test forms. The data were obtained from mathematics tests with both multiple choice (MC) and performance assessment (PA) items. The tests were administered in a statewide testing…

When determining criterion-referenced test length, problems of guessing are shown to be more serious than expected. A new method of scoring is presented that corrects for guessing without assuming that guessing is random. Empirical investigations of the procedure are examined. Test length can be substantially reduced. (Author/CM)

A comparison of double as opposed to single multiple-choice questions yielded significant differences in regard to item difficulty, item discrimination, and internal reliability, but not concurrent validity. (Author/PN)