CORRELATES OF NET GAIN FROM CHANGING MULTIPLE-CHOICE ANSWERS: REPLICATION AND EXTENSION

Abstract

Summary.-This study replicates the 1975 research of MueUer and Shwedel on changing multiple-choice answers and extends the research to students of business administration. The results support earlier findings of a strong positive net point gain when students changed answers. Sex of the student was not a factor, while high-scoring students benefitted more from changing answers than low-scoring students. Students are confronted with a multitude of objective examinations. With this onset comes the students' dilemma of whether to change original selected answers when later scrutiny indicates a different answer may be correct. Several researchers have examined the answer-changing behavior of college students in psychology, education, and educational psychology courses. This research has shown that most students do change answers to objective questions, but that their perception is that they lose more points than they gain from answer-changing (Foote & Belinsky, 1972; Jacobs, 1972; Mathews, 1929). Contrary to these expectations, all studies report that students generally increase their total scores as a consequence of changing their answers. Several researchers have also examined individual characteristics and potential causes of answer-switching behavior (Jackson, 1978; Videler & Hansen, 1980). However, prior research has been performed primarily with college students of psychology, education, or educational psychology. Mueller and Shwedel (1975) examined the correlates of net gain for graduate students enrolled in an educational measurement course. In identifying in the literature the limited subject matter taken by the student-subjects, the authors called for work on individuals studying a different subject matter to validate the results obtained from the closely related disciplines. The purpose of this study is to replicate the prior study and extend research on answerchanging to behavior of students of business administration. Specifically, this study examined the relationship of sex, answer-changing, and total test score to net changes in total score resulting from changed answers.