Schizophrenia patients and their relatives have been found to exhibit increased reflexive errors on the antisaccade task, suggesting the deficit reflects genetic susceptibility for schizophrenia. To evaluate the degree to which antisaccade error is elevated in schizophrenia relatives, we carried out a meta-analysis of the existing literature and a primary study examining whether the magnitude of reported differences between relative and nonpsychiatric comparison groups could be due to differences in participant inclusion criteria. Meta-analysis yielded a moderate to large effect size across studies comparing relatives and controls (Cohen's d=0.61; Glass' d(g)=0.87). Antisaccade performance in medically and psychiatrically healthy relatives (n=45), who were selected from a larger sample of relatives based on criteria applied to healthy controls, was significantly more impaired than in healthy control participants (d=0.81, d(g)=0.93). Moreover, excluded (n=71) and included relatives did not differ (d=0.14, d(g)=0.13). The results indicate that the antisaccade deficit is a robust phenomenon in unaffected schizophrenia relatives that is not due to differences in inclusion criteria between relatives and controls, and thus are consistent with a growing literature indicating that the antisaccade deficit will be a valuable endophenotype of schizophrenia.