The use of students as mock jurors in the majority of legal psychology studies on jury behavior has been criticized (e.g., Bray & Kerr, 1979; Diamond, 1997). This study examined the degree to which student mock jurors' decisions were generalizable to those of real jurors. The participants of the study included 297 jury-eligible university students and 297 volunteers from the venire in the same community as that in which the students resided. All participants viewed one of six versions of a videotaped criminal trial. The defendant testified in English or in Spanish. In addition, the race of defendant was varied. Three bilingual individuals served as defendants with one appearing to be of northern European origin, one of Latino background, and one of African origin. Dependent variables included verdict and, for those who found the defendant guilty, the number of years to which he should be sentenced, and whether he should be fined. Student mock jurors differed reliably from their community counterparts on several demographic characteristics. However, the two groups had mixed results in relation to decision-making tasks. There was no difference in individual verdict preferences, but the students' sentence recommendations were more punitive. These results are interpreted in the context of the generalizability of mock juror studies.