Information about Tests from Sato's S-P Chart

Abstract

Educators often find themselves in opposing camps on the value of multiple-choice examinations com pared to essays because, as some es sayists would have it, You can't sum marize a student's work with a single number. Our purpose is not to enter this debate directly but to illustrate that the instructor who is willing to take the time can currently derive more informa tion about students from a multiple choice test than was possible in the past. For years many college teachers, either by hand or with computer pro grams, have examined their students' multiple-choice tests to discover which items their classes have found easy and which difficult. They often do this to help in evaluating the examination or their own teaching. Often, also, instruc tors might discuss the exams with indi vidual students in order to help these students diagnose their test-taking be haviors. In both these situations the in structor is getting more information about the students than can be gotten from the total score. Anyone who has done this type of prying into the make-up of multiple choice exams must have felt uneasy, at