Assessing Changes in the Effect of Divorce Rates on Homicide Rates Across Large U.S. Cities, 1960-2000: Revisiting the Chicago School.

Researchers commonly include a measure of the level of divorce among the standard covariates in macro-level studies of homicide, justifying this practice with reference to social disorganization theory. We review the underlying logic for a divorce/homicide relationship, distinguishing between a "cultural/normative conflict" variant advanced by the classical Chicago School theorists and a "structural/control" variant associated with the neosocial disorganization perspective. We suggest that the cultural/normative conflict variant implies that the effects of divorce will become attenuated over time, whereas the structural/control variant implies stability in effects. We then assess the degree to which the effects of levels of divorce on homicide rates have changed with panel data for a sample of large U.S. cities during the period 1960-2000. The results of seemingly unrelated regression (SUR) analyses reveal considerable stability in the effects of a measure of divorce on homicide rates, especially if the divorce measure is combined with a "sibling" measure of family disorganization—the percentage of children not living with two parents. Our analyses suggest that the commonly observed positive effect of measures of divorce on homicide rates over recent decades is most plausibly interpreted with reference to the "structural/control" arguments associated with the neosocial disorganization perspective. [ABSTRACT FROM AUTHOR]