The Dynamics of Domestic Violence
利用明尼阿波利斯实验数据，分析警察处理家庭暴力案件的方式（逮捕、劝告或短期分离）对夫妻后续暴力的影响，发现逮捕更有效但效果短暂，并考察了年龄、就业等社会经济因素的作用。
The 0. J. Simpson affair is only the latest in a series of events that have focused attention on domestic violence. Using data collected by Murray Strauss and his colleagues, Lawrence Sherman (1992) estimates that each year there are 18 million incidents of domestic violence fitting the criminal-justice classification of an assault and that police officers handle at least two million cases of violence involving a spouse or lover. Domestic-violence assault is more common than all other forms of violence combined. Recognizing that domestic violence is a social as well as a private issue, a number of police departments have participated in experiments designed to learn how they can best handle domestic-violence calls. In this paper, we use data from the experiment in Minneapolis to determine how police treatments in cases of domestic violence affect the couple's subsequent violence. We also examine how socioeconomic factors such as age and employment are related to domestic violence. These factors are considered in part because they may affect the police actions. Our model is a stochastic dynamic model that allows the effects of the police actions to vary over time. In brief, we find that arrest is more effective than advising or short-term separation but that the differential effect is transitory.