ERIC Documentation Service, TM 019 276, pp. 1-7 (1992).
Robert Ennis tested for deductive reasoning competence using a multiple-choice format in which the premises are given and it is asked whether the conclusion would then be true; invalid arguments should be answered "Maybe." It was decided to investigate whether a conversationally more appropriate cue "Not necessarily" would affect scores. Using these items in fairly stressful conditions it was found that the change of format made no difference to 18 valid items but led to significant differences in response pattern to 13 of 18 invalid items, in most cases with more correct responses.
Press Back to return to list of publications. Return to Home page.
HTML version prepared May 26th, 1997.