Analysis of “Second-Guessed” Training Tests: Improvement on Objective Choice Tests Which are Reviewed and Re-Marked After Initial Correction

Abstract

IN SOME training programs, a corrected objective examination is re turned to the student for review. Particularly in correspondence courses, below-standard papers may be returned for re-doing. Errors are indicated, but neither errors nor correct answers are explained by the corrector. The trainee reviews the questions that he has missed, and chooses one of the remaining answers as correct. The paper, which the trainee has re-marked, is then re-corrected by the school. The school then returns the paper to the trainee, with correct answers indi cated and explained. The fallacies of wrong choices may also be ex plained. The school makes these comments on the questions whether or not the trainee has passed the examination. While the procedure is some what laborious, its purpose is to make the trainee study intensively in his areas of weakness. The present article presents formulae which seem to show for ex aminations made up of objective questions, that the improvement on re examination depends on the type of objective question used. For some types of test item, the examinee will find it easy to pick the right an swers to questions which have been marked as errors on the basis of the first testing and on some not. Too great an apparent improvem e n t through inflated test scores after the review, may weaken discrimina tion by bringing all scores close to the ceiling of perfection. Then too, different types of questions have different effects on the extent to which the trainee can review effectively the subject-matter covered by the ex amination. Lowering the area of effective review obviously decreas es the peculiar training value of this system of correcting papers. The second remains a guess and fails as a review. The situation can be summarized in terms of certain formulas that may be helpful to workers in this area.