Since the incarcerated population began to rise in the 1980's, overcrowding has been an issue for Federal, State, and local correctional facilities across the United States. Density and the perception of crowding have long been assumed to be a factor in violent behavior. The rapid pace at which inmates enter and leave a correctional institution makes proper classification and separation of troublesome inmates difficult. In addition, it can hinder the rapport between inmates and staff. Using the measures of both social and spatial density, this study examined the relationship between density and jail violence. Data were collected from 646 jails across the United States. Responding facilities were operating at an average of 94 percent capacity. The jail population was made up of 88 percent men and 52 percent of the inmates were awaiting trial. The analysis was divided into two parts: (1) a correlational analysis with independent and dependent variables and (2) multiple regression models to determine the change in coefficients of multiple determination as social density, spatial density, and transiency, then jail-level inmate characteristics, and final supervision variables were included. This measure of spatial density was found to be significantly associated with inmate-inmate assaults only after controlling for the social density and transiency variables. An unexpected finding was that the relationship between spatial density and assaults against inmates was in the opposite direction that was predicted. As the number of inmates residing in a jail rose above the facility’s rated capacity, the number of inmate-inmate assaults was found to decrease. Additional findings included: (1) bivariate analysis results indicated that the use of dormitories was negatively correlated with reported inmate-staff assaults; (2) transiency had a positive significant relationship with inmate-staff assaults; and (3) inmate/officer ratio was negatively correlated with both assaults against inmates and staff. The results of the study raise the possibility that other factors within the jails are accounting for variation in assault rates in jails, such as facility location; inmate participation in work, education, vocational, or rehabilitation programs; and classification techniques. References