MULTIPLE CHOICE VERSUS TRUE‐FALSE: A COMPARISON OF RELIABILITIES AND CONCURRENT VALIDITIES

Publisher Website
Google Scholar
Abstract
The purpose of this study was to compare the reliabilities of true‐false (TF) and multiple choice (MC) tests and to determine the concurrent validities of both. Two methods, judgmental and discrimination, were devised for objectively converting MC items to TF form. The TF items generated by the two methods from 70‐item MC natural science and social studies tests were incorporated in eight final forms which were differentiated by subject matter, conversion method, and item form order. A sample of 1018 nonurban high school students each responded to one of the eight forms. Examinees tried three TF items for every pair of MC items attempted. The TF tests were significantly less reliable than the MC tests but did tend to measure the same thing as the corresponding MC tests.

This publication has 1 reference indexed in Scilit: