A growing body of literature has emerged in the last few years that has begun to clarify the risk factors that are both unique to stalking as well as common to other forms of criminal behavior. Numerous definitions of stalking exist, but most of these definitions share several key elements, including the occurrence of repetitive, unwanted contact that is perceived by the victim as intrusive and/or threatening. The small samples and idiosyncratic methods that characterize many of the studies here clearly limit the conclusiveness of any interpretations based on this new but rapidly growing literature. A growing number of somewhat larger studies, each using different study methodologies, have generated greater consistency. The most reassuring conclusion that can be drawn from this review is that although some form of violence appears to occur quite often in stalking cases (between 30 and 50 percent of the cases), severe violence has been much less frequent. Cases of serious violence appear to be relatively infrequent (6 percent). In an analysis of victim reports, a higher rate of injuries was observed resulting from stalking-related violence (37 percent), but the definition of injury included cuts and bruising. No cases of life-threatening violence were reported. Several risk factors that are unique to stalking-related violence have emerged from the growing literature. Among the most consistent correlates of violence are threats and a previous intimate relationship between the victim and offender. A number of consistent findings have also emerged with regard to clinical diagnosis and stalking violence. A history of substance abuse has typically corresponded to an increased rate of violence among stalking offenders. The presence of a personality disorder yielded a statistically significant effect in meta-analysis. Prior criminal history and history of violence unrelated to stalking revealed substantial inconsistencies across the studies analyzed, and neither yielded a significant overall effect size in the meta-analysis. Too little research has studied the personal characteristics of stalking offenders, such as age, race, gender, and level of education, to justify any firm conclusions. 3 tables, 35 references