The interpretation of these results should be tempered by an awareness of the limitations of the USC study methodology. The surveys used to collect data relied heavily on law enforcement as a source of information. A logical next step would involve using an array of informants, including courts, schools, and social service providers in addition to community residents and gang members. It should also be noted that the USC data are cross-sectional in nature and cannot adequately describe second- or third-order waves of migration, wherein some individuals may travel from city to city.9 Another untapped dimension in the USC survey was termed "indirect migration," in which one gang is influenced by another gang that was influenced by a third gang. For example, Pocatello, ID, gangs were heavily influenced by gangs from Salt Lake City, which were started by gang members from Los Angeles (R. Olsen, Pocatello Police Department, personal communication, September 24, 1996). Other patterns of sequential mobility were reported on during the USC interviews, but did not occur with sufficient frequency to warrant further analysis.
The findings from the 1992 and 1993 USC surveys provide evidence that gang member migration, although widespread, should not be viewed as the major culprit in the nationwide proliferation of gangs. Local, indigenous gangs usually exist prior to gang migration, and migrants are not generally viewed by local law enforcement as the cause of gang problems. This pattern is less evident in cities in which gangs have emerged more recently, but these municipalities are no more likely to experience gang migration than chronic gang cities. Moreover, the motivations for gang member relocation (i.e., more often socially motivated than driven by crime opportunities) and patterns of gang participation (equally likely to join existing gangs as to retain original affiliation in order to initiate new gangs or branches) do not distinguish migrants in the two types of cities. Proponents of the "outside agitator" hypothesis of gang formation as described by Hagedorn (1988) will find little support in the data available from the USC national study.
On the whole, the USC findings agree with the research literature on gangs cited earlier. Many of the researchers -- Rosenbaum and Grant (1983), Hagedorn (1988), Huff (1989), Zevitz and Takata (1992), Decker and Van Winkle (1996), and Waldorf (1993) -- found that gang formation was only minimally affected by the diffusion of gang members from other cities. The findings reported by some researchers -- Skolnick et al. (1990) and NDIC (1994, 1996) -- are less consistent with those reported in the USC study. The Skolnick et al. and NDIC studies focused heavily on drug issues and may have disproportionately represented cities with drug-gang migration or with migrants that moved for drug expansion purposes.10 Such cities reflect a distinct pattern of gang migration -- older gang migrants, traveling longer distances, staying for briefer periods (see Maxson, Woods, and Klein, 1995, for full presentation of these analyses). Research that focuses on drug matters may fail to capture more prevalent trends. Although more often the subject of media coverage, migration for drug distribution purposes is less common than other types of migration. The differential patterns of gang migration, and their effects on local communities, require more research.
In addition, the USC findings are difficult to compare with those reported by Knox et al. (1996). Respondents in the Knox et al. study presented a widespread perception of outside gang influence. This may be the result of exposure to the media and products of the entertainment industry. Klein (1995) and others have suggested that the diffusion of gang culture in the media plays a key role in the proliferation of gang membership. Our Nation's youth are hardly dependent on direct contact with gang members for exposure to the more dramatic manifestations of gang culture, which is readily accessible in youth-oriented television programming, popular movies, and the recent spate of "tell-all" books from reputed urban gang leaders. The nature of this influence and its impact on gang participation and expansion have not been investigated systematically but are crucial in understanding fully the dynamics of gang proliferation.
Cities with emerging gang situations should examine the dynamics of their own communities before attributing their gang problems to outside influences. Socioeconomic factors, such as persistent unemployment, residential segregation, and the lack of recreational, educational, and vocational services for youth, are more likely sources of gang formation or expansion than is gang migration.