Scalia, 1996:Table 2). About 45 percent of the legal immigrants to the United States are Hispanic in origin (Heer, 1996:Table 5.4). Of course this grouping is itself quite heterogeneous, including immigrants from Mexico, Cuba, the Dominican Republic, Colombia, El Salvador, Guatemala, and other countries in Central and South America and the Caribbean islands.

The heterogeneity of immigrant groups is an important source of complication in the relationship between immigration and crime. Not only is there social and cultural variation, for example, among Hispanic immigrants, there may be further significant variation in self-selection to participate in particular kinds of crime, as well as in legal documentation, with regard to vulnerability to statutory mandatory sentences and further group-linked variations in age and gender. Of course, such sources of variation also plague studies of other general correlates of crime, such as poverty and family breakdown, and crime itself is a heterogeneous entity. These problems do not stop research on these topics, but they do constitute reasons for caution in reaching policy-related conclusions. We attempt to address some of these concerns below.

Calculated on the basis of the Survey of State Prisons, the imprisonment rate for U.S. citizens is about 3.5 per 1,000 population. This rate is used as the base for calculating ratios presented in the first column of Table 9-1 for the most frequently imprisoned immigrant groups in U.S. state prisons. These figures indicate that immigrants from Cuba and the Dominican Republic are incarcerated at rates between four and five times those of citizens, that immigrants from Mexico, Jamaica and Colombia are incarcerated at rates from two to two and one-a half times those of citizens, and that immigrants from Guatemala and El Salva-