Abstract: The Bureau of Justice Statistics and the National Center for Educational Statistics, in conjunction with Demographic Surveys Division (DSD) of the U.S. Census Bureau requested that staff in the Census Bureau’s Statistical Research Division (SRD) pretest proposed changes to the 2009 School Crime Supplement (SCS) of the National Crime Victimization Survey (NCVS). The sponsors requested a series of changes to some of questions, and SRD staff tested those suggested changes across two different rounds of cognitive interviews. This report contains the results of both rounds of cognitive interviews. Our main findings include: Overall, the changes to the questions were not problematic. The sponsors made changes to the questions about participation in school activities, the presence of friends and teachers with whom students can talk, the availability of drugs, avoiding certain school locations, and weapons at school. The sponsors also added a question about anonymously reporting of school threats and expanded the question on internet bullying. Overall, the changes and the additional question were not problematic. Respondents had little difficulty with the altered questions. Respondents had some trouble with questions that did not undergo any revisions. Respondents had some difficulty with the questions about getting to and from school, classroom disruptions, involvement in physicals fighting, and bullying. We recommended changes to these questions and the respondents accepted most of the suggested changes. Some questions continue to be problematic across multiple waves of the survey. In particular, the question about the availability of certain drugs yielded pretesting results that were inconsistent across both of the current rounds of interviews. These pretesting results were also inconsistent with previous pretesting research and the actual survey data. Respondents had some trouble with the different drug categories and were unfamiliar with some of the examples in the question. We encouraged the sponsors to consider this question in subsequent pretesting research.