Abstract:Fifty-two students sat two 60-question multiple choice examinations 8 months apart. The second examination consisted of 30 of the questions used at the first sitting, together with 30 questions with repeated stems but different responses. Immediately after the first sitting, 27 of the students went through the paper with the examiner. They were given the correct answers, and any problems were discussed. For the 30 questions with new options, the increase in marks was similar in the two groups. The group given … Show more
“…Similar increases in truth ratings have been found after reading false items on true-false tests (Toppino & Brochin, 1989). Likewise, a re-test with novel lures yields better performance than a re-test with old lures (Rees, 1986), because initial testing increases the perceived truth-value or familiarity of the lures (Jacoby, Shimizu, Daniels, & Rhodes, 2005). In a more direct demonstration, Roediger and Marsh (2005) showed that taking a multiple-choice test increased production of lures as answers on a final general knowledge test, even though subjects were warned against guessing.…”
Three experiments were conducted to investigate whether increasing the number of lures on a multiple-choice test helps, hinders or has no effect on later memory. All three patterns have been reported in the literature. In Experiment 1, the stimuli were unrelated word lists, and increasing the number of lures on an initial multiple-choice test led to better performance on later free recall and cued recall tasks. In contrast, in Experiments 2 and 3, stimuli were facts from prose materials, and increasing the number of multiple-choice lures led to robust costs in cued recall and smaller costs in free recall. The results indicate that performance on the initial multiple-choice test is a critical factor. When initial multiple-choice performance was near ceiling, testing with additional lures led to superior performance on subsequent tests. However, at lower levels of multiple-choice performance, testing with additional lures produced costs on later tests.
“…Similar increases in truth ratings have been found after reading false items on true-false tests (Toppino & Brochin, 1989). Likewise, a re-test with novel lures yields better performance than a re-test with old lures (Rees, 1986), because initial testing increases the perceived truth-value or familiarity of the lures (Jacoby, Shimizu, Daniels, & Rhodes, 2005). In a more direct demonstration, Roediger and Marsh (2005) showed that taking a multiple-choice test increased production of lures as answers on a final general knowledge test, even though subjects were warned against guessing.…”
Three experiments were conducted to investigate whether increasing the number of lures on a multiple-choice test helps, hinders or has no effect on later memory. All three patterns have been reported in the literature. In Experiment 1, the stimuli were unrelated word lists, and increasing the number of lures on an initial multiple-choice test led to better performance on later free recall and cued recall tasks. In contrast, in Experiments 2 and 3, stimuli were facts from prose materials, and increasing the number of multiple-choice lures led to robust costs in cued recall and smaller costs in free recall. The results indicate that performance on the initial multiple-choice test is a critical factor. When initial multiple-choice performance was near ceiling, testing with additional lures led to superior performance on subsequent tests. However, at lower levels of multiple-choice performance, testing with additional lures produced costs on later tests.
“…This phenomenon represents the memorial effects of exposure to misinformation and has been described (see Brown et aL, 1999) as the Negative Suggestion Effect (NSE). The robustness of the NSE has been demonstrated during multiple-choice tests (e.g., Rees, 1986), true-andfalse tests (e.g. , Toppino & Brochin , 1989), and spelling tests (e.g., Brown, 1986;Jacoby & Hollingshead, 1990).…”
Section: Repetition Of Inaccurate Perseverative Responding Responsesmentioning
Participants completed 5 classroom examinations during which the timing of knowledge of results (no feedback: Scantron form; delayed feedback: end-of-test, 24 hour delay; immediate feedback: educator, response form) and iterative responding (1 response, up to 4 responses) were manipulated. At the end of the semester, each participant completed a 100-item final examination which included 10 items randomly selected from each classroom examination, plus 50 entirely new items. Neither the source of feedback nor the number of responses permitted influenced performance on classroom examinations but both factors interacted significantly to enhance the final examination performance of participants provided with immediate feedback and iterative responding. The correction of initially inaccurate strategies by combining immediate feedback with iterative responding was not differentially effective as a function of information source: educator or the Immediate Feedback Assessment Technique (IF AT) form. For these participants, response identification accuracy, confidence ratings, and retention were higher and inaccurate perseverative responding was lower. Performance on the final examination permits the preliminary quantification of how immediate feedback coupled with iterative responding, when used during classroom examinations that contain items that will be repeated on a cumulative final examination, not only assesses student knowledge but also teaches in a manner that promotes the retention of course materials.Many of the earliest studies conducted in the psychological sciences were dedicated to examining changes in the performance of learners provided with information (Le., feedback) that either affirmed a correct response or corrected an error (e.g., Thorndike, 1913Thorndike, , 1927. This corrective information, viewed initially within an associationistic framework as a contingent event, strengthened correct responses
“…6 ' 7 MCQs may be useful in self-directed learning, 8 and there is evidence of useful student learning from feedback to students who have taken MCQ examinations. 9 The interactive nature of our postexamination review process facilitates discussion and debate and is an active, student-centered teaching activity that might be expected to enhance learning. The students in our study were enthusiastic about the learning experience, and our faculty reviewers were impressed that it was a useful learning experience for both the students and themselves.…”
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.