• The number of jurisdictions reporting youth gangs increased by 4.1 percent for the Nation as a whole, as shown by comparison of the estimated number of jurisdictions reporting gangs prior to 1996 with those reporting gangs in 1996. Contrary to this overall increase, gang activity decreased by 3.2 percent in large cities. Although the causes of these changes in gang activity are not immediately apparent, followup interviews were scheduled with agencies that experienced changes, and the results of this analysis will be reported in the future.

• Most gang problems began quite recently, especially in small cities and rural counties. Survey respondents most frequently cited 1994 as the year gangs began to pose a problem in their jurisdictions. The average year of onset was significantly earlier in the West (1986).

• When results were weighted for the number of gang members reported in each jurisdiction, half of the gang members nationwide were juveniles; the other half were adults. However, the majority of gang members (71 percent) were reported to be between the ages of 15 and 24.

• The average proportion of juvenile gang members was less in jurisdictions that reported their gang problem began prior to 1990 as compared with jurisdictions that reported their gang problem began between 1990 and 1996. The average proportion of adult gang members increased as population size increased.

• The average proportion of adult gang members increased as the level of gang member involvement in drug sales and the degree of gang control of drug distribution increased.

• Females were reported to be substantially less involved in gangs than males in 1996, despite other recent findings indicating relatively more female involvement. The average proportion of female gang members was affected slightly by population size, decreasing from 14 percent in populations of 1-9,999 to 9 percent in populations of 250,000 or more.

• The racial and ethnic composition of gangs in 1996 appears to be different from what earlier national surveys and research had indicated. When the number of gang members reported in each jurisdiction was controlled for, Caucasians accounted for 14 percent of all gang members. In addition, the proportion of Caucasian gang members was especially high in rural counties (32 percent) and small cities (31 percent).

• As population size increased, the average proportion of African-American, Hispanic, and Asian gang members increased and the average proportion of Caucasian gang members decreased.

• The average proportion of minority gang members, especially Hispanics, in newer gang jurisdictions was lower than the average proportion in older gang jurisdictions. Conversely, the average proportion of Caucasian gang members was substantially higher in newer gang jurisdictions compared with older gang jurisdictions, suggesting that the increase in Caucasian gang membership was a recent trend.

• The average proportion of African-American gang members increased as the level of gang member involvement in drug sales and the degree of gang control of drug distribution increased. Under the same circumstances, the average proportion of Caucasian, Hispanic, and Asian gang members decreased.

• Forty-six percent of gangs throughout the country were shown to be multiethnic/multiracial when the number of gang members reported in each jurisdiction was controlled for. Multiethnic/multiracial gangs were especially prevalent in suburban counties (55 percent) and small cities (52 percent).

• When unweighted percentages were used, the average proportion of multiethnic/multiracial gangs was highest in the Midwest (55 percent) and lowest in the Northeast (39 percent).

• Approximately 84 percent of respondents indicated that their jurisdictions experienced some gang migration. In addition, when the number of gang members reported in each jurisdiction was taken into account, 21 percent of the gang members in jurisdictions that experienced some migration were estimated to be migrants.

• Respondents in small cities reported the highest average proportion of migrants (36 percent). The larger the jurisdiction, the smaller the average proportion of gang migrants. Regionally, the average proportion of migrants was highest in the Midwest (34 percent) and the Northeast (33 percent).

• An estimated 2,364 homicides that occurred in large cities and 561 homicides that occurred in suburban counties involved gang members. The larger the population of a jurisdiction, the higher the number of homicides involving gang members.

• Respondents indicated that youth gang members were, relatively, more involved in larceny/theft, followed fairly closely, in the order of degree of involvement, by aggravated assault, burglary, and motor vehicle theft. The number of jurisdictions that reported a high degree of involvement in aggravated assault, robbery, and motor vehicle theft increased as the population of jurisdictions increased.

• On average, respondents estimated that 43 percent of the drug sales in their jurisdictions involved gang members, although most respondents reported gang member involvement at the high and low ends of the spectrum. Almost half of the jurisdictions in large cities and small cities reported that gang members were not very involved in drug sales. Additionally, the majority of jurisdictions in the West and Northeast indicated that gang members were not very involved in drug sales.

• Nearly half (47 percent) of the respondents indicated that gang members controlled or managed less than one-quarter of all drug distribution in their jurisdictions. Jurisdictions in the Midwest and Northeast reported a high degree of gang control of drug distribution. Approximately 36 percent of respondents in the Midwest and 33 percent in the Northeast indicated that more than half of the drug distribution in their jurisdictions was controlled by gangs.

Analysis of these data by the National Youth Gang Center will continue, and subsequent surveys will gather additional information in areas that require further examination. The NYGC survey database also will be accessible to other gang researchers for analysis.