It is important to count the number of jurisdictions reporting youth gangs because youth gang problems are an significant but largely ignored public policy issue for two main reasons. First, gang policy in the United States does not make a distinction between youth gangs and adult street gangs and ganglike criminal organizations. Understandably, a distinction would not need to be made if public policy were concerned only with apprehending and prosecuting persons who commit street crimes. The police mantra is "Investigate the crime, not the culture." However, public crime policy has a broader aim—preventing and reducing gang problems. To inform such a policy, detailed information on youth gang problems is needed (Miller, 1990). The lack of such detailed information has driven Miller's 28-year compilation of gang problem jurisdictions.

Second, the contribution of youth gangs and their members to juvenile delinquency, especially violence, has not been incorporated into juvenile delinquency policy and program development. Studies of large urban adolescent samples reported since 1995 show that gang members are responsible for a large proportion of violent offenses committed by the entire sample. Rochester, NY, gang members (30 percent of the sample) self-reported committing 68 percent of all violent offenses (Thornberry, 1998). Two-thirds of the chronic violent offenders in this urban sample were gang members for a time (Thornberry, Huizinga, and Loeber, 1995). In a Seattle, WA, sample of adolescents, gang members (15 percent of the sample) self-reported committing 85 percent of robberies perpetrated by the entire sample (Battin et al., 1998). Youth gang members in Denver, CO (14 percent of the sample) self-reported committing 79 percent of all serious violent offenses (Huizinga, 1997). The studies in Denver and Rochester were parts of OJJDP's Program of Research on the Causes and Correlates of Delinquency. Until 1998, delinquency studies reported that the worst influences on nondelinquents were delinquent friends; a study in Seattle revealed that gang members were far worse influences (Battin et al., 1998). This finding was replicated in the Rochester adolescent sample (Thornberry, 1998). The policy implication of these discoveries has been stated succinctly by youth gang researchers: "Because gangs have such a major effect on delinquent behavior, prevention efforts aimed at reducing delinquency and substance use should seek to prevent and reduce gang involvement" (Battin-Pearson et al., 1998, p. 10).

This Report supports this policy recommendation. Miller's compilation of localities (States, cities, and counties) affected by youth gangs documents the urgency of incorporating this recommendation into juvenile delinquency and crime policy in the United States.

Miller's method of compiling youth gang localities adds to the great policy value of his Report. There are three methods of tallying the number of youth gang problem jurisdictions nationwide, all of which have strengths and limitations. The first method, an ethnographic (direct observation) census of every jurisdiction in the United States, guided by one explicit definition, is cost prohibitive. Such a study might not even be feasible because of lack of agreement on what constitutes a youth gang and because youth gangs are amorphous and difficult to count.

The second method uses written questions to survey a nationally representative sample of jurisdictions. Unfortunately, this method was not used before the National Youth Gang Center (NYGC) surveys conducted in 1996 (Moore and Terrett, 1998; NYGC, 1999), 1997 (Moore and Terrett, 1999; NYGC, 1999), 1998, and 1999. These are the first national surveys of all large cities and suburban counties and representative samples of small cities and towns and rural counties. The survey method produces invaluable information on the existence, scope, and nature of youth gang problems that can be generalized to the Nation as a whole.

Because fully representative national surveys were not conducted in earlier periods, Miller uses the third method—examining historical trends in jurisdictions where youth gang problems have been reported in national surveys and other sources. His baseline is the first national survey of youth gangs, which he conducted in the 1970's (Miller, 1982). Most of the jurisdictions in his compilation were respondents to the 1995 (Moore, 1997; NYGC, 1997), 1996, and 1997 NYGC surveys. He also draws on other national surveys; Federal, State, and city agency reports; and his own tabulation of youth gang jurisdictions reported by local media and other sources. Using this large database, Miller traces developments in States, cities, and counties over a 28-year period. Thus, his Report provides a historical perspective on youth gang problems that informs delinquency and youth crime policies. Only a few of his findings (based on 1998 data) are highlighted here to illustrate the policy implications of youth gang problems; his Report contains many others.

• In the 1970's, only 19 States reported youth gang problems. By the late 1990's, all 50 States and the District of Columbia had reported gang problems.

• The number of cities reporting youth gang problems rose from 270 in the 1970's to 2,547 in 1998—an increase of 843 percent.

• The number of counties reporting gang problems rose from 101 in the 1970's to 1,152 in 1998—an increase of more than 1,000 percent.

• The regional location of gang cities changed substantially from the 1970's to the 1990's. In the 1970's, the West led the Nation, while the South ranked lowest. By 1998, the South had risen to second place, with a 33-fold increase, while the number of gang cities in the West had increased only by a factor of 4.

• In the 1970's, only about 27 percent of the population of all cities was affected by gang problems. This proportion rose to about 60 percent of the population of all cities by the late 1990's.

Between the 1970's and the 1990's, the number of smaller cities with gang problems increased much more rapidly than that for larger cities. The number of gang cities with populations larger than 10,000 increased about 7 times, while the number of gang cities with populations smaller than 10,000 increased almost 30 times. Later in this Report, Miller summarizes his findings in the following manner:

Youth gang problems in the United States grew dramatically between the 1970's and the 1990's, with the prevalence of gangs reaching unprecedented levels. This growth was manifested by steep increases in the number of cities, counties, and States reporting gang problems. Increases in the number of gang localities were paralleled by increases in the proportions and populations of localities reporting gang problems. There was a shift in the location of regions containing larger numbers of gang cities, with the Old South showing the most dramatic increases. The size of gang-problem localities also changed, with gang problems spreading to cities, towns, villages, and counties smaller in size than at any time in the past.

When combined with youth gang research showing the contribution of gang members to juvenile violence, Miller's study makes a compelling case for assigning higher priority to youth gangs in crime and delinquency policy. Jurisdictions experiencing youth gang problems cannot make significant progress in preventing and reducing juvenile and young adult violence without addressing youth gang problems at the same time.

James C. Howell, Ph.D.
Pinehurst, NC