Performances of 60 male and 60 female undergraduates on a conventional multiple-choice test were compared with their performances on a test that enabled them to assess the correctness of their own answers. The self-assessment approach reduced the bias seen in differences between male and female scores. (SLD)

The reliabilities of 2-, 3-, 4-, and 5-choice tests were compared through an incremental-option model on a test taken by 154 high school seniors. Creating the test forms incrementally more closely approximates actual test construction. The nonsignificant differences among the option choices support the three-option item. (SLD)

It is proposed that the multiple-choice test can be useful in helping language students understand the meaning of text better without overwhelming them. Techniques for drafting effective questions are discussed, and a sample list of questions is presented. (MSE)

To investigate the ability of multiple-choice tests to assess higher order thinking skills, examinations were constructed as half multiple choice and half constructed response. Results with 90 undergraduate and graduate students indicate that the 2 formats measure similar constructs at different levels of complexity. (SLD)

Analysis of educational psychology textbooks identified textbook authors' guidelines for teachers to follow when writing multiple-choice test items. Selected guidelines were used to evaluate multiple-choice items (N=1,080) from 18 college instructor guides to educational psychology texts. Results indicated that approximately 60 percent of the…

Instead of relying on standardized test scores and interdistrict comparisons, school systems must develop a more powerful, timely, and local approach to accountability that is truly client-centered and focused on results. Accountability requires giving successful teachers the freedom and opportunity to take effective ideas beyond their own…

Effects of experimenter-controlled incentives and feedback on the calibration of performance were examined for 85 undergraduates answering reading comprehension and mathematics multiple-choice questions. Results support the hypotheses that incentives affect the calibration of performance and that external incentives can enhance the degree to which…

Places performance assessment in the context of high-stakes uses, describes underlying technologies, and outlines the history of performance testing from 210 B.C.E. to the present. Historical issues of fairness, efficiency, cost, and infrastructure influence contemporary efforts to use performance assessments in large-scale, high-stakes testing…

The recall of information about Hepatitis B demonstrated by 180 seventh graders was tested with three test types: (1) short-answer; (2) true/false; and (3) multiple-choice. Short answer testing was the most reliable. Suggestions are made for the use of short-answer tests in evaluating student knowledge. (SLD)

Gender differences in performance on multiple-choice and constructed-response items in mathematics were studied with 3592 11th graders taking a high school graduation examination. Results suggest that gender differences in mathematics may be linked to content and item format, thus supporting the usefulness of a many-faceted Rasch model for…

In order to assess the impact of changes made by Citizenship and Immigration Canada, adult citizenship-education programs were examined nationwide. Far fewer programs were available to immigrants to Canada than in 1987. Content scope in citizenship-education programs has remained essentially unchanged or been reduced, and is heavily influenced by…

The effect of different item formats (multiple choice and open) on gender differences in test performance was studied for the Swedish Diagrams, Tables, and Maps (DTM) test with 90 secondary school students. The change to open format resulted in no reduction in gender differences on the DTM. (SLD)

A study evaluated use of Conceptests in a large pharmacy lecture course. Multiple-choice questions test students' ability to apply a concept presented in lecture to a situation not yet encountered. Students are presented with two or three Conceptests during each lecture, and must reevaluate responses after 15 minutes of additional lecture.…

Explores the relationship between the strategies second language learners use to interpret unfamiliar words in a target language. English-speaking learners of Japanese completed a beliefs questionnaire and multiple choice Kanji compounds test. Examined how learner beliefs are related to the ability to combine information from word parts and…

Considers how to use the World Wide Web for education. Topics include computer-based education initiatives; assessment, including multiple choice questions; a prototype Web site design based on a client-server model; and future work, including distance education and providing better graphic user interfaces. (LRW)