THE EFFECTS OF QUESTION LEVEL IN PROSE MATERIAL ON LEARNING, ATTITUDE, AND SUBSEQUENT BEHAVIOR (SOURCE)

This study compared effects of question level in prose material on learning, attitude, and subsequent behavior in an applied setting. Subjects were randomly assigned to Fact, Application or No-question groups. In a Randomized Subjects Design, 186 clinical training students read a 20 page packet of prose material which explained the concept of empathy and how to express empathy in relationships with patients. The Fact group received 20 pages of prose content with two fact post-placed questions. The Application group received the same prose content with two application post-placed questions. The No-question group received only the prose material.;An 80 item multiple choice posttest and/or the 64 item Hogan Scale of Empathy or no test administrations occurred immediately after the prose material had been read.;The posttest was composed of 20 fact questions based directly on in-text fact questions in the fact question prose material, 20 application questions based directly on in-text application questions in the application question prose material, and 20 fact questions and 20 application questions based on the prose material but not the questions in the prose material, not in-text. All subjects took the posttest two weeks later.;Within two weeks of the original prose study, 34 of the subjects were observed and rated by two faculty raters during a doctor/patient appointment. A 10 item clinical rating scale adapted from the IAQ (Interviewing Attributes Questionnaire) was used for rating.;Data analyses found attitude and clinical behavior not affected by the question condition. Analysis of posttest administration data produced significant findings for question level, question source, and testing, but not for treatment group. Analyses of significant interactions indicated in-text or not in-text question effect significant with fact questions but not application questions, and question level effect and in-text or not in-text question effect significant for both tests. Analyses of only delayed posttest data was consistent with the earlier analyses.;Results of the study were attributed to familiarity of subjects with recall processing and the probability that processing of information at higher levels is of a multidimensional character deserving of further research.