tion (see Hagan and Bumiller, 1983). It may also be the case that immigrant drug offenders are sentenced with greater severity than others.

The most detailed recent study of criminal justice processing decisions involving immigrants was undertaken in El Paso and San Diego by Pennell et al. (1989). This study found in both cities that illegal aliens made up the largest proportion of immigrants prosecuted, and that these illegal aliens were much less likely than others to be released from jail prior to trial. For example, in El Paso only 14 percent of illegal immigrants compared with over 50 percent of all others were able to "bailout" prior to trial. This difference may be associated with the fact that the Immigration and Naturalization Service can place "holds" on illegal immigrants, that illegal aliens are financially less able to post bail, and that illegal immigrants are less likely to have the community ties that often are required for early release. When accused persons are unable to obtain release they may have greater difficulty generating resources to defend themselves in court, leaving them more vulnerable to conviction, and ultimately to imprisonment.

To assess the above possibilities we reanalyzed the Pennell et al. data that were collected in El Paso and San Diego. The results are presented in Table 9-2. These results confirm that immigrants in general in El Paso and San Diego are more likely to be detained prior to trial, and that, in turn, detention before trial increases the risks of conviction and imprisonment. That is, immigrants are at greater risk of conviction and imprisonment because they are more vulnerable to pretrial detention. In addition, immigrants in El Paso and San Diego who are charged with drug offenses are more likely than others to be sentenced to prison. These differences cannot be the result of immigrants having more extensive criminal histories, because Scalia (1996:Table 4) demonstrates that noncitizens are much less likely to have a prior known criminal history. A likely implication of these findings is that processing differences result in immigrants being overrepresented in prison populations. It will be important for further research to examine counterbalancing decisions, for example involving deportation, but the current findings suggest the likelihood that immigrants are overrepresented in prison as a result of justice system decision making.

An additional source of misperceptions about immigration and crime that may result from the uncritical reliance on prison statistics involves ways in which these figures are often used in characterizing immigrant criminality. The 1991 Survey of State Prison Inmates reported that nearly half of all alien inmates were incarcerated for drug offenses, that about 40 percent of these alien inmates used drugs during the month prior to their arrest, and that about 20 percent were under the influence of drugs at the time of their current offense. The 1991 survey reports that very high proportions of alien inmates from Colombia (87 percent) and the Dominican Republic (67 percent) were incarcerated for drug offenses.

Although it is the case that just over 20 percent of all inmates were incarcerated for drug offenses in 1991, about half of all inmates said they had been using