killings in the late 1990s. This raises the possibility that there may have been some kind of epidemic of rampage shootings in the late 1990s that cut across all ages, including youth. Consistent with this hypothesis is the evidence from the cases that copycat mechanisms, which clearly were at work in at least one of the shootings, also may have influenced two of the other three suburban and rural school shootings examined.
Trend data on school shootings indicated that the school rampage cases listed in the legislation were not in fact new or unique: there had been similar incidents of school violence as far back as 1974. Remarkably, we could not find similar rampage shootings in the nation’s inner cities, a fact that surprised the committee. Only three events in inner-city schools across the country met our formal criteria for inclusion in the study. And when we looked closely at these cases, we found that they looked quite different from the cases in the suburban and rural schools.
Looking across the cases, we found that the eight shooters exhibited a number of similar traits. While these are consistent with risk factors for serious youthful violence identified in the literature, this study can do no more than claim them as tendencies or propensities. All were boys. Five had recently begun hanging out with delinquent or more troubled friends. Five had a relatively recent drop in their grades at school. Five had engaged in previous serious delinquent acts and the other three in minor delinquent behavior. Serious mental health problems, including schizophrenia, clinical depression, and personality disorders, surfaced after the shootings for six of the eight boys in these cases. All had easy access to guns. The rural and suburban boys had experience with guns, and one of the urban teens appears to have practiced with the gun he used.
However, there were also some characteristics that are usually thought of as protective. Half of the shooters came from intact and stable two-parent families, and five of the eight were good students, at least until 8th grade. Only three of the shooters struggled with grades or experienced the early school failure that frequently precedes the development of serious delinquent behavior. Only one of the eight shooters was a loner, and only two were gang members. Most had friends, although the quality of the friendships differed. Most of these shooters were not considered to be at high risk for this kind of behavior by the adults around them.
The central differences in these cases can be found in community structure. The two urban neighborhoods were characterized by commu