A meta-analytic review of published and unpublished intervention studies to change teachers' attitudes and behaviors toward students with disabilities was conducted in order to clarify the concepts of exceptionalities, attitudes, and change. The second objective of this study was to examine the moderating variables that may account for disparate results. The moderators that were examined were the level of intentionality of teachers at the beginning of the interventions, intervention characteristics, measurement characteristics, and teachers' specialization (regular vs. special education). The findings showed that lectures that used an indirect approach to attitude change and that were accompanied by field experience resulted in more change in attitudes. In addition, those studies that used self-report questionnaires showed more attitude change than the studies that used observational measures. Teachers' expectations and beliefs were easier to change than their behaviors and emotions. Conclusions and implications of the findings are discussed.