A Secondary Analysis of Claims Regarding the Reflective Judgment Interview [microform] : Internal Consistency, Sequentiality and Intra-Individual Differences in Ill-Structured Problem Solving / Phillip Wood

The Reflective Judgment Model and associated interview (RJI) (Kitchener and King, 1981) measure the ability of individuals to reason about ill-structured problems. It has gained popularity as a measure of college outcomes associated with postsecondary education. This study examines general claims for the model made from existing data from 15 of 25 studies of the RJI, representing 1,334 subjects and 4 of 5 available longitudinal studies. About 0.3% of scores were found to have coding or clerical errors or to have been incorrectly read into statistical programs. Discrepant definitions of rater agreement were found for some studies, making it difficult to interpret reported agreement rates. An improved scoring system is proposed that results in improved internal consistency estimates for the RJI. Examination of data reveals a systematic pattern of increasing performance and variability of the RJI as a function of educational level, replicating past findings. Confirmatory factor analysis suggests that the dilemma topics of the RJI represent tau-equivalent measures with significant intra-individual variation for the individual topics of the RJI. Seven figures and 11 tables present analysis results. (Contains 67 references.) (SLD)