Variables that represent the major theoretical perspectives were used to examine differences in a much larger sample of nations than has been used in previous research. Regression analysis indicated that homicide rates and, to a lesser degree, total crime rates had positive associations with prison rates; whereas, street-level law enforcement had a negative association. Dummy variables that represented nations grouped by geographic region indicated a historical-cultural component in the use of imprisonment. Little to no support was found for civilization theory, inequality, or unemployment variables. Discriminant analysis indicated that degree of human development and freedom, as well as inequality, helped classify nations according to prison conditions; whereas, the freedom variable helped classify nations that did and did not use capital punishment. Dummy geographic region variables were important in classifying nations by prison conditions and capital punishment, again indicating the probable importance of historical-cultural factors. 6 tables and 72 references