Intimacy between two disputants provides insulation against the occurrence of violence. However, once violence becomes a viable option for conflict resolution, intimacy may exacerbate incident outcomes. Furthermore, because of the emotional intensity, relationally close assailants may be less attuned to the situational characteristics, a known predictor of incident outcomes such as injury. This study explores the role of relational distance as a determinant of the completion (versus the attempt or threat) of non-sexual assault, as well as a potential moderator of the relationship between situational characteristics and assault completion. The analysis uses 35,616 incidents of non-sexual assault from the National Crime Victimization Survey (1992–2008). A logistic regression model of assault completion is estimated, with attention devoted to the influence of relational distance, demographic, and situational characteristics. A heteroscedastic logistic regression model is then estimated, including product terms between the regressors and an indicator for whether the assailant was relationally distant (versus relationally close). Relational distance is strongly and inversely related to the likelihood of assault completion. Furthermore, 10 of 12 situational characteristics (e.g. bystander presence use of a weapon) are significantly more strongly correlated with assault completion in incidents where the victim and assailant(s) are relationally distant.
Read full abstract