Allen Terry, Novak Shannon A, Bench Lawrence L
University of Utah, UT, USA.
Violence Against Women. 2007 Aug;13(8):802-16. doi: 10.1177/1077801207302040.
This study uses two types of independent variables, age and the location of the physical wound, to develop a model of injury patterning that identifies violent behavior without direct observation of the assault. In this research, domestic violence injuries are compared to accidental injuries. The results indicate that there are specific and predictable injury patterns that separate abuse from other kinds of wounds. A logistic regression model was developed to identify the regions of the body most susceptible to injury from domestic assault. Using the age of the victim and the injury regions, probabilities were calculated to determine which wounds were caused by abuse.
本研究使用两种类型的自变量，即年龄和身体伤口的位置，来建立一个损伤模式模型，该模型无需直接观察袭击行为就能识别暴力行为。在这项研究中，将家庭暴力造成的伤害与意外伤害进行了比较。结果表明，存在特定且可预测的损伤模式，可将虐待造成的伤害与其他类型的伤口区分开来。开发了一个逻辑回归模型，以确定身体上最容易受到家庭暴力袭击伤害的部位。利用受害者的年龄和受伤部位，计算概率以确定哪些伤口是由虐待造成的。
