Standardizing Perceived Task Difficulty for Studies of Cheating: A Methodological Refinement and Sample Tasks

Abstract

As a contribution to research on cheating (in experiments that investigate ethical action) three “tests” of intellectual competence which permit cheating were standardized for perceived level of difficulty. A vocabulary, remote-associates, and trigram word-making task were rated for perceived difficulty and ranked for relative perceived difficulty by 71 college students. Analyses of variance of the mean rankings and ratings showed that the three tasks are of widely (and significantly) different degrees of perceived difficulty. Moreover, the three tasks occupy nearly equal distances from one another along the dimension of perceived difficulty, permitting classification by experimenters as “easy,” “medium,” and “very difficult” in false norms provided to subjects. Suggestions for use in cheating studies included manipulations that require credible false-norm feedback as well as those which attempt to assess the clash between perceived difficulty and discrepant instructions concerning level of difficulty.