Work on the impact of religion on suicide has not yielded consistent results. Classic Durkheimian models stressing religious affiliation (Catholic vs Protestant) have been questioned in recent years and some authors have argued for a model based on the alternative conception of religious commitment (e.g. religiosity per se) as a prophylactic against suicide. Almost all of this work has been based on highly aggregated data where it is not known if nonreligious people account for the suicides. The present study tests both models with national data on 1,687 respondents. No support is found for the Durkheimian model at the individual level, but some is found for the religious commitment model: the greater the church attendance the lower the approval of suicide. The effect of religiosity on suicide ideation is independent of education, gender, marital status, and age.