Aggravated Assault and the Urban System: Dallas, 1980-81
Keith D. Harries
Stephen J. Stadler
DOI: 10.2190/EY9P-0EW0-DK43-XTDT
Abstract
Prior research on the relationship between violence, heat stress, and the urban socioeconomic environment has suggested links between these three variables. Specifically, low-status neighborhoods appeared to exhibit a relatively exaggerated response to heat stress compared to higher status areas. This article uses prior work on the severe summer of 1980 as a point of departure for comparisons between 1980 and 1981 (a "normal" year). In the study period consisting of 609 days, some 9,994 assaults were recorded in Dallas. The underlying expectations were that 1) 1981 would, in general, show a diminished level of assault in the summer, owing, in part, to diminished heat stress, and 2) certain environmental relationships revealed in the analysis of 1980 data would be replicated with a substantially expanded data set. Initially, weather data for 1980 and 1981 were compared in order to establish that the summer of 1981 was indeed meteorologically different from that of 1980. Then a general model was developed, incorporating a Discomfort Index, day of the week, month, and selected interaction terms. Residuals were analyzed. Pairwise month-by-month comparisons of mean assault frequencies were made, by neighborhoods classified according to their socioeconomic status. In general, 1980-81 differences were less pronounced than expected, possibly due in part to population growth in the Dallas area. Other explanations lie in the so-called curvilinear effect, and the calendar effect. This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 United States License.