Assessing the Race‐violence Relationship at the Macro Level: The Assumption of Racial Invariance and the Problem of Restricted Distributions*

Macrolevel research on the race‐violence relationship has focused on the assumption of racial invariance in the effects of structural disadvantage measures on violence. Yet in most urban areas black and white disadvantage distributions only partially overlap, which precludes a critical empirical test of the assumption. I refer to this as the problem of “restricted distributions.” Using block group data for Atlanta, results show that the effect of a disadvantage index on violence is similar in black and white neighborhoods within the low range of the disadvantage distribution, but diminishes significantly at the higher levels prevalent in black areas. I discuss the implications of the findings and suggest avenues for future research.