Researchers agree that official data, especially those from developing countries, must be viewed with caution, because they suffer from the two major defects of all official crime data: their precision and reliability is dependent on the vagaries of citizens reporting their experiences to the police and to the data-collection and processing decisions made by government agents. A major problem for research on crime in developing countries is that the most common resource and method for estimating the degree of unreliability and inaccuracy of official data, i.e., victimization surveys, is not consistently available. One way to address this situation is for researchers to become familiar with data-collection methodologies of specific countries as well as international research and findings. Further, in order to allow for comparisons of nonreporting levels and reasons across countries, researchers must conduct victimization surveys and continue them over some time to develop trend lines. Questions about reasons for not reporting crime to police should be included in such surveys. Specific types of victimizations believed to be underreported should be studied separately as problems in methodology and interpretation. 15 notes and 87 references