TY - JOUR
AU1 - Cronbach, Lee J.
AB - 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 
TI - Response Sets and Test Validity
JF - Educational and Psychological Measurement
DO - 10.1177/001316444600600405
DA - 1946-12-01
UR - https://www.deepdyve.com/lp/sage/response-sets-and-test-validity-FvlBBLaykB
SP - 475
EP - 494
VL - 6
IS - 4
DP - DeepDyve
ER -