Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Response Sets and Test Validity

Response Sets and Test Validity LEE J. CRONBACH University of Chicago A PSYCHOLOGICAL or educational test is constructed, by choosing items of the desired content, and refining them by empirical techniques. The assumption is generally made, and validated as well as possible, ,that what the test measures is determined by the content oflthe items. Yet the final score of the person on any test is a composite of effects resulting from the content of the item and effects resulting from the form of item used. A test supposedly measuring one variable may also be measuring another trait which would not influence the score if another type of item were used. This paper attempts to demonstrate these influences in a variety of tests, to examine the effect of these extraneous factors, and to suggest means of controlling them. Numerous studies show that scores may be influenced by variables other than the one sppposedly tested. In the Min- nerotn Mziltiphasic Personality Inveatory, for example, it is expIicitly recognized that a spbject may evade questions by the excessive use of the response “Cannot Say,” even though his actual behavior might be properly described by the response “True” or “False.” This tendency invalidates the test profile for persons http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Educational and Psychological Measurement SAGE

Response Sets and Test Validity

Cronbach, Lee J.
Educational and Psychological Measurement , Volume 6 (4): 20 – Dec 1, 1946
Download PDF
20 pages

Loading next page...
 
/lp/sage/response-sets-and-test-validity-FvlBBLaykB

References (30)

Publisher
SAGE
Copyright
© 1946 SAGE Publications
ISSN
0013-1644
eISSN
1552-3888
DOI
10.1177/001316444600600405
Publisher site
See Article on Publisher Site

Abstract

LEE J. CRONBACH University of Chicago A PSYCHOLOGICAL or educational test is constructed, by choosing items of the desired content, and refining them by empirical techniques. The assumption is generally made, and validated as well as possible, ,that what the test measures is determined by the content oflthe items. Yet the final score of the person on any test is a composite of effects resulting from the content of the item and effects resulting from the form of item used. A test supposedly measuring one variable may also be measuring another trait which would not influence the score if another type of item were used. This paper attempts to demonstrate these influences in a variety of tests, to examine the effect of these extraneous factors, and to suggest means of controlling them. Numerous studies show that scores may be influenced by variables other than the one sppposedly tested. In the Min- nerotn Mziltiphasic Personality Inveatory, for example, it is expIicitly recognized that a spbject may evade questions by the excessive use of the response “Cannot Say,” even though his actual behavior might be properly described by the response “True” or “False.” This tendency invalidates the test profile for persons

Journal

Educational and Psychological Measurement – SAGE

Published: Dec 1, 1946

There are no references for this article.