You are not currently logged in.
Access JSTOR through your library or other institution:
Incremental Reliability and Validity of Multiple-Choice Tests with an Answer-Until-Correct Procedure
Gerald S. Hanna
Journal of Educational Measurement
Vol. 12, No. 3 (Autumn, 1975), pp. 175-178
Published by: National Council on Measurement in Education
Stable URL: http://www.jstor.org/stable/1433950
Page Count: 4
Preview not available
It was theorized that an experimental answer-until-correct procedure, whereby an examinee continues to respond to each multiple-choice item until feedback signifies that the keyed answer has been selected, would yield scores of greater reliability and validity than would conventional procedures. A paper, an interpretive exercise, and an essay examination provided criterion measures against which to compare the criterion-related validities of multiple-choice test scores derived both by the experimental method and by an inferred conventional procedure. Compared with the inferred conventional scores, the experimental scores were more reliable but less valid. Implications and relationships to previous research are discussed.
Journal of Educational Measurement © 1975 National Council on Measurement in Education