The study was undertaken to develop a practical method of adjusting changes (between pretest and posttest) in self-report scores of self-concept for concurrent changes in defensiveness. Data were drawn from a population of teachers and paraprofessionals in training, or independent subscales of the Tennessee Self-Concept Scale. Changes prior to adjustment were not statistically significant on either the self-concept or the defensiveness scale. However, when changes for each subject on the defensiveness scale were converted to a standard score and applied as a correction factor to the self-concept score for the individual, differences became statistically significant. (Author)