The paper argues that as reported, international homicide statistics are incomplete and unreliable for global descriptions of and comparisons of this crime. In addition, international quantitative measures of homicide rates are superficial and one-dimensional; and the commonly used definition of homicide as interpersonal violence is too narrow. The authors suggest that certain types of comparative questions about homicide might be best answered by using the homicide event; and the research agenda should be broadened to include both interpersonal homicide and collective homicide. The paper argues for a mixed method approach that takes advantage of all qualitative and quantitative data sources available. The QCA method developed by Ragin (1987: 2000) is used as an illustration. The QCA is not a statistical technique in the traditional sense. Instead, it helps delineate characteristics among relatively small groups of cases. A key quality of QCA is its application with small datasets, such as those often confronted in cross-national research. The focus of QCA is on summary measures. Individual cases have a background role that is overshadowed by the interpretation of relationships among variables. In QCA, cases are brought to the foreground to be examined in their full historical, sociological, and political contexts. Key issues of judgment in QCA analysis include the variables that are to be included as independent and dependent variables, as well as the cutoff points needed to create the binary data if the original data are at other than the binary/nominal level. The QCA methodology is demonstrated in the cross-national homicide data from 21 countries. 5 tables, 3 figures, 20 notes, 37 references, and appended illustration of the use of the International Homicide Index