Defining youth gangs
As noted in the Methodology section, definitions continue to pose problems in evaluating youth gang activity on a national level. To address these problems, the 1998 National Youth Gang Survey asked respondents about the characteristics they consider important in defining a youth gang. The survey also provided respondents with an alternative definition of gangs and asked them what percentage of gangs in their jurisdictions would fall within this definition. The response to these questions is discussed below.
Identifying the characteristics that define a youth gang is difficult because individual law enforcement agencies think differently about gangs, in terms of each agencys own laws, language, observations, and experiences. This problem can be magnified when agencies lack a formal written definition and even more so when defining characteristics vary among units within law enforcement agencies. To better understand how law enforcement agencies characterize gangs, the 1998 National Youth Gang Survey asked respondents to rank six gang characteristics according to their importance as criteria in defining a youth gang, with 1 indicating the most important criterion and 6 indicating the least important.
Table 45 lists the six characteristics and shows the number and percentage of first-place selections each received as a criterion for defining a gang. Although none received an overwhelming majority of votes as most important, commits crimes together clearly was the most important criterion, receiving a first-place selection by 50 percent of the respondents. The next most popular characteristic, has a name, was ranked most important by only 19 percent of the respondents.
Tables 46 and 47 permit further analysis of how agencies use the six characteristics to define youth gangs. Table 46 presents a frequency analysis for the six characteristics, showing the number of respondents who selected each characteristic as most important (i.e., the frequency of choice 1), second most important (frequency of choice 2), and so forth. The table also shows the spread for each characteristic, that is, the difference between the frequencies of choice 1 and choice 6. Table 47 then lists the frequency sequences for each characteristic. For example, the characteristic commits crimes together was choice 1 for 613 respondents, choice 2 for 166, choice 3 for 161, choice 4 for 113, choice 5 for 122, and choice 6 for 46. The spread for that characteristic is 567. The frequency sequence is 123546.
Commits crimes together and has a name both received substantially more rankings of 1 (most important) than the other characteristics. As noted earlier, commits crimes together had primary importance, with one-half of all respondents assigning it a 1. The relevance of this characteristic is clear: its highest frequency (choice 1) is very high; its lowest frequency (choice 6) is very low; its spread is, therefore, very large (567 votes); and its frequency sequence, noted in the example above, is almost in straight numeric order. The relevance of having a name is less clear: the frequency spread for this characteristic is very small (only 53 votes), which means that the six ranking choices received almost equal numbers of votes.
For other characteristics, table 47 shows that the ranking choices receiving the most votes were as follows: has a leader or several leaders (choice 3), hangs out together (choice 4), displays or wears common colors or other insignia (choice 5), and claims a turf or territory (choice 6). It is puzzling that choice 6 was most popular for claims a turf or territory. This indicates that a majority of respondents thought of this as the least important criterion in defining a gang. However, a closer look in table 46 reveals that the next most common ranking for this characteristic was choice 2 and that the difference in votes between choices 2 and 6 (54) was negligible. Although many respondents considered this characteristic least important, almost as many considered it quite important. This contradiction points to a lack of agreement on how to identify a gang.
The data indicate that these characteristics all play a part in defining a gang. That committing crimes together emerged as a very important criterion could be attributed to the fact that the respondents are law enforcement officials. However, this characteristic was selected as most important by only one-half of the respondents. The lack of agreement on gang characteristics is further illustrated in respondents answers to the second part of the question on gang characteristics, in which they were asked to list any other characteristics their agencies considered important in defining a youth gang.
Several answers addressed the minimum number of gang members, which varied from only two to at least five. Some agencies looked for structure and leadership to identify gangs; others looked for a lack of leadership. Another theme several respondents mentioned was the physical appearance of gang members (e.g., tattoos, similar haircuts, and similar style of dress) or their automobiles (e.g., driving the same kind of automobile and having the same markings on their automobiles). Another common group of characteristics involved factors such as attitude, antisocial or intimidating behavior, development of a language, or a common set of beliefs. Some agencies identified a youth as a gang member if the youth corresponded with, was friends with, or had a picture taken with a known gang member.
The survey findings reported in this section make it clear that law enforcement agencies use a wide variety of characteristicsnot just committing crimes togetherto define youth gangs. The findings also suggest an opportunity for further research in this area. The next section analyzes responses to the survey question regarding the applicability of an alternative definition of youth gangs. That analysis sheds some light on the findings presented here.
In an attempt to determine the effect of various youth gang definitions on responses, an alternative definition of youth gang was tested in the 1998 survey. The alternative definition was devised by an eminent gang researcher, Malcolm Klein. The objective was to see whether the alternative definition helped specify the criteria law enforcement agencies use to identify youth gangs. Respondents were asked what percentage of the youth gangs they had reported would fit Kleins alternative definition:
This alternative definition is very similar to the one that respondents were instructed to use in the 199698 National Youth Gang Surveys: a group of youths or young adults in [the respondents] jurisdiction that [the respondent] or other responsible persons in [the respondents] agency or community are willing to identify or classify as a gang. As in the alternative definition, the original definition instructed survey respondents to exclude motorcycle gangs, hate or ideology groups, prison gangs, and exclusively adult gangs. There is, however, one important difference between the two definitions. The alternative definition specifies gang involvement in illegal activities, whereas the original definition does not.
Because of the similarity of the two definitions, it is not surprising that respondents who reported gang problems said that nearly 8 out of 10 (79 percent) of the gangs they had reported under the original definition would fit the alternative definition. Many agencies (546, or 54 percent of all agencies responding to the question) reported that 100 percent of their gangs met the alternative definition. Most of the 21-percent difference likely is accounted for by the fact that, as reported in the previous section, law enforcement agencies use a broad set of criteria in defining gangs. One hundred and five agencies (10 percent) reported that none of their gangs met the alternative definition. Although involvement in criminal activities is the predominant criterion, it is not the only one.
Indeed, although law enforcement agencies might not perceive some youth gangs as involved in illegal activity, such gangs are nevertheless an important matter of concern because of their potential criminal involvement. Surveys of students have found that many gangs identified by students are not particularly active in serious and violent crimes.
In a recent national survey of students ages 1219, more than one-third (37 percent) reported gangs at their schools (Howell and Lynch, 2000). Although about two-thirds of the surveyed students reported that gangs at school were involved in one or more of three types of illegal activity (violence, drug sales, or gun carrying), 40 percent of the students said gangs were involved in only one of these types of activity. About one in five students (21 percent) said gangs were involved in two of the measured activities, and just 8 percent said they were involved in all three types of activity. Thus, only a small fraction of the students surveyed said gangs were highly active in all three types of serious criminal activity.5
Currys (2000) 5-year followup study of self-reported gang members among young students (grades 68) found continuity between gang membership at young ages and later police records. About one-half of the young gang members had subsequent police records, and about 20 percent of them were eventually identified by police as gang-related offenders. Thus, Curry argues, gangs and gang members identified in student surveys are not a separate gang problem from that indicated in law enforcement data.
Because only a small fraction of student-affiliated youth gangs may be highly active in serious and violent crimes, law enforcement agencies may be reporting only a portion of gangs and gang members that are actually present in schools. Analyses of the 1998 National Youth Gang Survey data show that within youth gangs, adult members (i.e., those age 18 or older) were more heavily involved in criminal activity than were juveniles (age 17 or younger), with the exceptions of burglary/breaking and entering and larceny/theft. Much of the 21-percent difference between the number of gangs reported under the original National Youth Gang Survey definition and the number reported under the alternative definition might well be accounted for by younger adolescent gangs that are not yet extensively involved in criminal activity. Curry concluded that although student survey data and official law enforcement data do not coincide perfectly, together they can produce a picture of gang activity that contributes to a better understanding of youth involvement in gangs over time.
Tables 48 and 49 show responses to the question on the alternative definition, by area type and region. There was very little variation by area type. This is surprising given that other 1998 survey results based on the original National Youth Gang Survey definition showed less gang and gang member involvement in serious and violent crime in small cities and rural areas (with the exception of gang member involvement in drug sales). There were, however, regional differences, which may be a function of State definitions of gangs that emphasize involvement in illegal activity. The strongest correspondence between gangs reported under the original and alternative definitions (84 percent) was in the West. This may be explained by the proliferation in this region of the California STEP Act gang definition, which emphasizes that criminal street gangs have as one of their primary activities the commission of criminal activities (Klein, 1995:28).
The section on youth gang defining characteristics noted that survey findings suggest an opportunity for further research in this area. The findings on an alternative definition also point to the need for more research on the criteria law enforcement agencies and others use to define youth gangs.