from Mexico, is potentially misleading. Our data suggest that Mexican immigrants are more like their age and gender peers than is commonly assumed. This finding helps to resolve a paradox in the picture of Mexican immigration to the United States, because by other measures of health and well-being—including smoking, drug use, and the birthweight of babies—Mexican immigrants are generally found to do as well or better than U.S. citizens. One argument is that this is because of the strength of extended and nuclear families and religion in Mexican families. Insofar as this is the case, we may wish to place the priority on finding ways to preserve, protect, and promote the social and cultural capital that Mexican immigrants bring to their experience in the United States, rather than overemphasize issues of crime and punishment.

One especially important way in which Mexican and other immigrants to this country might benefit from better protection involves the risks to which they are exposed as victims of crime. In this chapter we have not considered the criminal victimization of immigrants largely because we could find no available source of data on this topic (but see Sorenson and Shen, 1996). The U.S. government invests large sums in public surveys of crime victimization, but these surveys do not include significant numbers of immigrants, and no special surveys of immigrants have been undertaken for this purpose. This task deserves special priority.

Also, and notwithstanding the potentially misleading picture we have found with regard to issues of immigration and crime, especially in the Mexican context, further work should be attentive to at least two poorly understood issues. The first involves the important task of making projections into the future that take into account interrelations among immigration, differential fertility, and social behavior. As we discuss in the Appendix to this chapter, higher rates of fertility among immigrant groups, either alone or in combination with other factors, such as criminal justice system bias, could result in immigrants forming larger proportions of prison populations in the future, even if their group propensities to crime remain constant. These processes are in need of further study.

Second, it is also likely the case that specific groups of immigrants, much like specific groups of citizens, do have a heightened propensity that leads them to be disproportionately involved in crime. This typically involves countries with relatively few immigrants to the United States. In these cases in which immigration is rather limited, there may be unique social networks and selection processes that explain the higher rates of crime involvement. If legal immigration from these countries was greater, it is plausible that rates of crime associated with these immigrant groups in the United States would be less striking, if only because the effects of these selection processes and social networks would be diluted. We know too little about these special cases to say much more, but enough to recognize that it is likely misleading to extrapolate from such cases to the experience of immigrants more generally.