Population Size and Area Type
Table 1 compares the onset of gang problems by population size and shows that later onset is more common in less populated jurisdictions. Nearly three-fourths (73 percent) of cities with populations of 250,000 or more reported onset of gang problems before 1991. A majority (54 percent) of jurisdictions with populations between 50,000 and 99,999 reported onset during 1986–90 or earlier. In contrast, a large majority (61 percent) of jurisdictions with populations between 25,000 and 49,999 reported onset during 1991–92 or later. Jurisdictions with populations of less than 25,000 were especially likely to report onset during 1993–96; nearly two-thirds (64 percent) of the smallest jurisdictions (less than 10,000 population) reported onset of gang problems during this period.
Table 2 compares the onset of gang problems by area type: larger cities, smaller cities, suburban counties, and rural counties. A majority of larger cities (55 percent) reported onset of gang problems before 1991, whereas majorities of smaller cities (73 percent), suburban counties (61 percent), and rural counties (82 percent) reported onset during 1991 or later. Rural counties tended to have the latest onset of gang problems, with a majority (65 percent) reporting onset during 1993–96. A smaller majority (51 percent) of smaller cities reported onset during 1993–96. Compared with rural counties and smaller cities, suburban counties tended to have slightly earlier onset of gang problems, with the majority (52 percent) reporting onset during 1991–94.
Age, gender, and race/ethnicity. Tables 3, 4, and 5 show the age, gender, and racial/ethnic composition of gangs by year of gang problem onset.5 Compared with gangs in earlier onset jurisdictions, gangs in later onset jurisdictions tended to have younger members, a slightly larger proportion of female members, and a much larger proportion of Caucasian and African American members.
As shown in table 3, gangs in later onset jurisdictions had about the same proportion of younger members (younger than age 15) as gangs in earlier onset jurisdictions, a much larger proportion of members ages 15–17, and a much smaller proportion of older members (18 or older). Thus, gangs in later onset jurisdictions included a greater proportion of juveniles (i.e., individuals younger than age 18). This finding is especially evident when comparing the earliest (before 1981) and latest (1995–96) onset jurisdictions.
As shown in table 4, females represented a much smaller proportion of gang members than males, regardless of when a jurisdiction’s gang problem began. However, jurisdictions with the latest onset of gang problems (1995-96) had the largest proportion of female gang members (14 percent)2 to 4 percent greater than the female proportion for other onset groups.
As shown in table 5, racial/ethnic differences between gangs in later versus earlier onset jurisdictions were even more extreme than age differences. In jurisdictions with onset before 1981, a majority of gang members were Hispanic (58 percent). In contrast, in the later onset jurisdictions (1991 and later), Caucasians were the predominant group, followed by African Americans.
Multiethnic/multiracial gangs. The 1996 survey asked: “What percentage of the gangs in your jurisdiction are multiethnic or multiracial?” As shown in table 6, earlier onset jurisdictions reported a much smaller proportion of racially mixed gangs than later onset jurisdictions. Such gangs represented about one-third (32 percent) of all gangs in jurisdictions with onset before 1981, compared with more than half (56 percent) in jurisdictions with onset during 1991-92, 50 percent in those with 1993-94 onset, and 40 percent in those with 1995-96 onset.
In a more specific question, the 1998 survey asked respondents to estimate the percentage of gangs in their jurisdictions with a “significant mixture of two or more racial/ethnic groups.” Table 6 shows that such gangs were far more prevalent in later onset jurisdictions than in earlier onset jurisdictions. Only 18 percent of the gangs in jurisdictions with onset before 1981 had a significant racial/ethnic mixture, in contrast with 55 percent in jurisdictions with 1991–92 onset, 48 percent in those with 1993–94 onset, and 47 percent in those with 1995–96 onset. As in the 1996 survey, gangs with a significant multiethnic/multiracial mixture were most commonly reported in jurisdictions with onset of gang problems in 1991–92.
Firearms. The 1998 survey asked agencies to estimate how frequently gang members in their jurisdictions used firearms in assault crimes: “often,” “sometimes,” “rarely,” or “not at all.” More than half of all respondents (53 percent) said gang members used firearms often or sometimes. As shown in figure 1, firearm use by gang members in assault crimes was much less common in later onset jurisdictions than in earlier onset jurisdictions. A large majority (84 percent) of agencies in the earliest onset group (before 1981) reported that gangs often or sometimes used firearms in assault crimes, compared with only 32 percent of agencies in the latest onset group (1995–96)—a difference of 52 percent.
Drug trafficking. The 1996 survey asked respondents to estimate the percentage of drug sales in their jurisdictions that involved gang members and the proportion of drug distribution that was controlled or managed by gangs. As shown in table 7, the average share of drug sales involving gang members was 45 percent for the earliest onset jurisdictions (before 1981) and 35 percent for the latest onset jurisdictions (1995–96)—a difference of 10 percent. The same comparison for gang control of drug distribution shows a 30-percent difference between the earliest and latest onset groups (41 percent and 11 percent, respectively). Thus, both gang member involvement in drug sales and gang control of drug distribution were lower in the late-onset localities, but the difference was much greater for the latter measure. In other words, gangs in late-onset localities were, relatively speaking, less involved in drug distribution than in drug sales.
The 1998 survey asked respondents to estimate the percentage of gang members in their jurisdictions who were involved in drug sales. As shown in table 7, the earliest onset jurisdictions reported an average of 83 percent of gang members involved in drug sales, compared with an average of 65 percent for the latest onset jurisdictions—a difference of 18 percent.
Homicides. The 1998 survey also asked respondents to report the number of gang-related homicides in their jurisdictions. Figure 2 shows the percentage of jurisdictions in each gang problem onset period reporting no gang-related homicides, one or two such homicides, and three or more. The patterns for the number of gang homicides relative to gang problem onset period were most consistent in jurisdictions reporting either no homicides or three or more homicides. One-third (35 percent) of jurisdictions with gang problem onset before 1981 had no gang homicides. With one exception, this proportion consistently increased over the onset periods to 85 percent in the latest period (1995– 96)a difference of 50 percent between the earliest and latest periods. Conversely, the proportion of jurisdictions with three or more gang homicides decreased overall from 40 percent in the earliest onset period to only 4 percent in the latest onset period—a difference of 36 percent. The pattern was somewhat less consistent for jurisdictions reporting one or two gang-related homicides.
Although the proportion of jurisdictions reporting one or two homicides decreased between the earliest and latest onset periods, the proportion is slightly larger for jurisdictions with onset during 1986–90 than for those with onset before 1981 or during 1981–85 and is also slightly larger for those with onset during 1993–94 than for those with onset during 1991–92. In general, however, gang-related homicides were far less prevalent in jurisdictions with later onset of gang problems than in jurisdictions with earlier onset.
Other crimes. The 1998 survey asked what proportions of gang members were involved in aggravated assault, robbery, larceny/theft, burglary/breaking and entering, and motor vehicle theft: “most/all” (75–100 percent), “some” (26–74 percent), “few” (1–25 percent), or “none” (0 percent). Figure 3 shows that the percentage of agencies reporting involvement of most/all or some gang members in the two violent crimes (aggravated assault and robbery) in 1998 was consistently lower in the latest onset jurisdictions than in the earliest onset jurisdictions—a difference of 41 percent for aggravated assault and 37 percent for robbery. However, as shown in figure 4, a different pattern emerges for property crimes (larceny/theft, burglary/ breaking and entering, and motor vehicle theft). Compared with the earliest onset jurisdictions, the percentage of latest onset jurisdictions reporting involvement of most/all or some gang members was 38 percent lower for motor vehicle theft but 5 percent higher for burglary/breaking and entering and 4 percent higher for larceny/theft.
Thus, the 1998 crime measures indicate that gang members in the latest onset jurisdictions were most likely to be involved in burglary/breaking and entering and larceny/theft. Involvement of most/all or some gang members in these two property offenses was reported by 63 and 73 percent of the latest onset jurisdictions, respectively. Fewer than half of the latest onset jurisdictions reported similar levels of involvement for the other three criminal activities measured.