Data reliability and validity are major methodological concerns in cross-national analyses of crime. Despite the large literature on cross-national homicide rates, there is little agreement on which source of data provides the most reliable estimates. In addition, few studies have examined the potential threat to validity posed by unclassified deaths. Through a description of trends over time as well as multivariate analyses, the current study aims to shed some light on these questions by (1) assessing the reliability of cross-national homicide data from the United Nations and the World Health Organization, and (2) investigating the impact of unclassified deaths on the validity of WHO data. Findings indicate that UN and WHO homicide rates (n=56) differ in magnitude but produce similar outcomes. Drawing on well-known correlates of cross-national homicide rates, the UN data provide more robust results and produce statistical models with less error. We find that WHO data are more stable and reliable over time, and better suited for longitudinal analyses. Findings also suggest that analyses drawing on WHO homicide data should not disregard unclassified deaths because their inclusion produces better fitted statistical models and provides a closer estimate of the true number of homicides.