Risk Factors For Gang Membership Research Paper

This sample Risk Factors For Gang Membership Research Paper is published for educational and informational purposes only. If you need help writing your assignment, please use our research paper writing service and buy a paper on any topic at affordable price. Also check our tips on how to write a research paper, see the lists of criminal justice research paper topics, and browse research paper examples.

The question “Why do people join gangs?” has been of particular interest to those hoping to prevent gang membership. There are a number of potential methods to gain insight into the people who join gangs and the conditions or factors that lead to membership. These methods include the descriptive approach (a descriptive account of the characteristics of gang members), the direct approach (asking gang members why they joined), and the risk factor approach (identifying predictors of gang joining). Each approach is accompanied by limitations and benefits. The risk factor approach offers a number of methodological benefits and has significantly moved the gang literature forward in the most recent decades. A summary of gang risk factor studies and their theoretical contributions is offered here. While great strides have been made in the gang risk factor area, much work remains. A discussion of needed areas of research and potential hurdles is included.


Gangs have historically been a complicated phenomenon to understand. They form, they change, some disappear, some reappear, and some show amazing stability. People join, and some stay, but many leave quickly. In short, gangs and gang members come in so many shapes and sizes it has been difficult to define with confidence, let alone identify them all and study the extent and nature of the problem. But, there is a great desire to do so. One major reason for a preoccupation with gangs and their members is the strong link between gang membership and delinquency.

Gang members commit more delinquency than nongang members. This robust finding has been replicated over time, samples, and place (both across the United States and around the world). The rate of delinquency and violence has been so noteworthy that much public attention (media, research, policy, policing) on gangs has been in an attempt to address this issue. However, the history of concentrated gang suppression efforts has shown that gangs can be quite resilient. There is, in fact, evidence to suggest that attempting to force a gang to disband may actually increase their cohesion and group bond, thus actually making them stronger entities and less likely to dissipate. Therefore, prevention efforts (i.e., keeping gangs from forming or keeping people from joining) have become particularly important options for this type of social problem. For prevention to be feasible, however, the impetus for gang membership must be well understood. In other words, we must ask “Why do people join gangs?” If there is a greater understanding about the reasons for gang joining, there may be better venues to combat the problem before it starts.

Background Description

Gang studies have been challenged with definition and identification issues. There has not been a universal consensus (among practitioners, policy makers, or researchers) on exactly what constitutes a gang or a gang member. Therefore, efforts to identify, classify, or study gangs have always been fraught with questions of validity, comparability, and generalizability (for a review, see Klein and Maxson 2006). While this limitation has not hindered the ability to produce quality research studies, it must always be considered when interpreting gang findings.

The history of gang research has spanned both macro-and micro-level inquiry and has included both qualitative and quantitative methodologies. The earliest gang studies were generally qualitative studies that focused on a specific area or particular gang. These primarily descriptive studies were central in providing a rich understanding of gangs and their behaviors. A number of these studies reveal the conditions under which gangs form, and some implicated reasons why gang members joined their gangs. Quantitative studies soon followed. Many of these studies cast a wider sampling net by including both gang and nongang youth. These types of studies have been critical in assessing differences between gang and nongang members, the factors led some to join and others to resist, as well as a plethora of other subjects.

The quality of the quantitative, individual-level studies has improved substantially as gang research began to adopt both the use of cross-sectional methods and the more methodologically rigorous longitudinal data collection efforts. The expansion of knowledge in the area of risk factors for gang membership has benefited tremendously from data produced by prospective, quantitative, and longitudinal studies like the Denver Youth Survey (Esbensen et al. 1993), the Rochester Youth Development Study (Thornberry et al. 2003), the Pittsburgh Youth Study (Lahey et al. 1999), the Montreal Longitudinal Experimental Study (Craig et al. 2002), and the Seattle Social Development Project (Hill et al. 1999). These studies have been used to test and validate conclusions of cross-sectional studies and to provide firm, temporal ordering for predictors of gang membership. Longitudinal research efforts are integral to the study of risk factors (see Thornberry et al. 2004 for overview). They are, however, incredibly costly, time intensive, difficult to conduct, and therefore rare. As such, we are limited to only a handful of these types of studies. It is difficult to make sweeping generalizations regarding the importance of specific factors with such few studies. Cross-sectional studies help to bolster confidence in results, but they suffer from temporal ordering problems and cannot be given as much weight. Therefore, while much progress has been made in area of gang risk factor research, conclusions are still limited and many of our conclusions are still inconclusive.


There are a few ways in which to answer the question “Why do people join gangs?” The first method would be to describe the types of people that are gang members and infer that these characteristics were important determinants in why they joined a gang. This approach is often used in media or law enforcement depiction of gangs. These methods examine a group, neighborhood, city, state, or country and identify the gang members and describe them. Using this approach, results might suggest that gang members are often minority, males from highly impoverished areas. The implication for the question “Why do people join gangs?” then becomes that something about a person’s status as a minority, male, or resident of an impoverished area leads to gang membership. Generally, the descriptors available are superficial (sex, race/ethnicity, socioeconomic status, location, age) because the data being collected are from an outside observer’s determination of gang membership. In other words, this method does not include surveying every resident of a community about their gang status and then presenting descriptive information. This approach is limited to the observer’s ability to determine gang membership and the quality of the data that can be collected (usually) from afar. This approach is also limited in that it does not offer a good context for comparison. Even if a majority of gang members are minority, male, and residents of an impoverished area (which is debatable), it is not true that a majority of all minority, male, residents of impoverished areas are gang members.

The second way to find out why people join gangs is to ask them directly. Both qualitative and quantitative researchers have used this approach to gain insight into gang joining. This is a very direct and valuable approach. It is also limited in certain ways. First, usually, the sample captured in the direct method is more limited than the observational method due to the work intensity. Second, this approach requires finding gang members that are willing to disclose sensitive information. Self-report will always raise questions about recall and honesty. Finally, gang members can either be asked to offer their reasons for gang joining with or without prompts (i.e., answer choices). If gang members are allowed to answer without prompting the gamut of responses can be infinite and difficult to classify and generalize. If they are probed with specific responses, then an argument can be made that their responses are not as genuine. Regardless, the direct approach provides unique insight into the cognition of gang members.

Finally, there is a risk factor approach to determining who joins gangs and why. Risk factors are conditions (either individual or environmental) that increase a person’s likelihood of developing some negative condition. Risk factors precede the outcome of interest and, therefore, should be measured at an earlier period than the outcome. This requires the use of at least two observational points (i.e., a longitudinal study). In addition, the risk factor approach requires a comparison group (in this case, nongang members) to identify characteristics that distinguish gang members as a special group. In order to obtain this information from gang and comparison groups, a larger sampling net must be cast. This method often relies on respondent’s assessment or admittance that they are gang involved, which can assist in identifying gang members that may not be as “obvious” in the observational approach first described (but also suffers from the same limitations as the direct method described above). Usually in the study, a broad range of questions and statuses can be asked. One benefit to this method is that, unlike directly asking a respondent why they joined a gang, the predictors of gang joining need not be known to the respondent.

The Descriptive Approach

The descriptive approach uses the characteristics of gang members to understand “who joins a gang.” If a status is overrepresented in a gang sample than what would be expected in the general population, one could infer that status is related to gang membership. Official data (i.e., arrest statistics, incarceration rates) is one example of data that can offer a descriptive account of gang members. These data are collected when gang members come to the attention of the justice system. These data are often limited to whatever characteristics are collected by law enforcement (i.e., mostly superficial like sex, race, age). Using official data and the descriptive approach, gang members are often described as older, male, racially/ethnically minority, and from impoverished areas.

It is important to note that the descriptive picture of gang members differs depending on the use of self-report versus official data. Conducting a similar descriptive analysis of gang members using self-report data can yield a different depiction of what a “gang member” is. Self-report data collection efforts find that the peak age of gang joining is much younger than official data would offer. Self-report also shows that though males are more likely to be gang members, females make up a more significant proportion of gang members than official data would indicate. Self-report studies have identified gang members in both high-risk communities and low-risk communities. Though official data may suggest that gang members are usually racial/ethnic minorities, self-report studies identify substantial proportions of white gang members.

From Gang Members Themselves

A number of cross-sectional and longitudinal studies have asked gang members why they joined gangs. The most common response prooffered by gang members varies by study. Many of these studies collect these data by offering choices that gang members can select as the reasons they joined; therefore, if an answer is not supported by a study, it may be because it was not offered as a choice or because it was never selected by a gang member.

Many studies report that protection is a common reason for gang joining as suggested by members. This answer is both commonly asked by many studies and is a commonly reported reason offered by gang members. Another common reason for gang membership is that family or friends are involved in the gang. Studies that looked specifically at differences between girls’ and boys’ responses to gang joining found that girls were more likely to say that they joined because friends and family were in the gang than boys (though it was a common response for boys as well). Gang members also cite a desire for respect or a reputation, to feel like they are a part of their neighborhood, and to fit in better or to feel like they belong. It is also quite clear that the sheer fun and excitement of gang membership is an attractive feature for gang members.

This method has been very useful in debunking certain myths associated with gang joining. Gangs are often depicted as being organized, disciplined groups that are brought together for profit making in illicit business – in particular, drug selling. Studies show that it is actually rare for youth to report that they joined the gang for instrumental purposes like drug selling. Gang members identified by Esbensen and colleagues (1999) from a national sample of American youth, however, did often state that the general notion of “for money” was an important impetus for joining. Another commonly hypothesized cause of gang membership is that youth are forced to join gangs. A national survey of youth found that youth (boys and girls) rarely reported that they joined a gang because they were forced.

Risk Factor Approach

The risk factor approach requires both temporal ordering and a comparison group in order to truly identify the characteristics that lead to gang membership. As such, risk factor studies are generally far more labor intensive and costly than a cross-sectional descriptive study or even an interview study with gang members. Thus, the number of “true” risk factor studies is quite low. More cross-sectional studies (no temporal ordering) have been conducted that identify differences between gang and nongang members. These studies are useful and informative but will always suffer the lack of temporal ordering. However, longitudinal studies have generally supported the results of cross-sectional studies which lend greater confidence to the results. For the purposes of this review, the results of both longitudinal and cross-sectional studies will be included, though any result based solely on cross-sectional studies will be viewed with more caution.

Drawing from the more general literature on juvenile delinquency, gang risk factors are often presented in five domains: individual, family, school, peers, and community. Studies have shown that gang risk factors span all five domains. Therefore, no domain should be ignored. Previous work on risk factors for juvenile delinquency, more generally, shows that the importance of any of the five domains varies by social development. In other words, during some periods of the life course, certain domains may be more important than at others. It has been strongly argued that the causes of gang membership are unlikely to be set at the earliest stages of the life course. There is no evidence of an innate predisposition for gang membership. This makes intuitive sense because gang joining necessitates a gang to join – which is strongly influenced by life and environmental circumstance. This suggests that infant gang risk factor studies may not be a top priority, but starting early is still of serious consideration. Longitudinal research shows that important risk factors are present at relatively early ages. For example, using the Seattle Development Study, (Hill et al. 1999) measured risk factors at ages 10–12 that predicted gang joining at ages 13–18.

It has been repeatedly shown that risk factors for gang joining have a cumulative effect. In other words, the more risk factors present, the greater the likelihood of gang involvement. In the Seattle Study, youth with seven or more risk factors were 13 times more likely to join a gang than those with no risk factors or only one. The cumulative effect of risk factors has been shown both cross-sectionally and longitudinally. And this pattern has also been frequently demonstrated for juvenile delinquency more generally. It also appears that individuals with risk factors across multiple domains are at an increased likelihood for gang membership.

In reviewing the risk factor literature, it is important to keep measurement issues in mind. For ease of generalization and an overview, the intended spirit of measures needs to be grouped together, but the specific wording of each measure varies from study to study. And, as with any interpretation of quantitative data analysis, the determination of what constitutes “significant” is important. Bivariate relationships are the easiest types of analyses to detect any relationship between an independent and dependent variable. These types of relationships may not “hold up” in multivariate models. Does that make them irrelevant? In some cases, these same factors may be included in other multivariate models specified with other measures and be significant. Does that make them relevant? Clearly, model specification matters to the ability to detect significance. For better or for worse, this review mostly relies on the results of the most sophisticated statistical model presented by the researchers.

Individual Domain

Of all risk factor domains, the testing of the “individual factors” domain is the most expansive. Individual factors include demographic features, individual attitudes and beliefs, as well as behavioral characteristics. The ability of demographic features to predict the likelihood of gang joining has certainly depended on the sample under consideration and the type of analysis being conducted. As stated prior, males are more often thought to be gang members especially by media or law enforcement estimates, but self-report shows that a significant proportion of gang members are female. Similarly, the overrepresentation of minorities as gang members is less dramatic in self-report studies. Still, sex and race/ethnicity as significant risk factors of gang membership vary tremendously by study and the samples under investigation. Numerous school-based, self-report studies fail to find differences by sex and racial/ethnic propensity for gang membership, and other youth studies find differences. A handful of studies have examined the role of ethnic identity on an individual’s propensity to join a gang. These studies consider the role of ethnic identity separately for different racial/ ethnic groups under the theoretical assumption that gang membership may be related to the level of integration that youth feel within their racial or ethnic group. Studies testing ethnic identity have failed to find a significant relationship between identification within their ethnic group and gang joining.

It has been hypothesized that one factor that leads to an increased likelihood to participate in gangs are internalizing or externalizing disorders or affective states. Factors that have been examined in this category include disruptive behavior, conduct disorder, impulsivity or hyperactivity, risk seeking, low self-esteem/self-derogation, or the lack of guilt. Self-esteem is one of the most commonly tested factors and not commonly found to be associated with gang joining in either cross-sectional or longitudinal studies. Nondelinquency problem behaviors like conduct disorder, disruptive behaviors, or oppositional defiant disorder are more often available for consideration in longitudinal studies and have been consistently supported as a significant risk factor for gang membership. Impulsivity, risk seeking, and guilt have been less frequently tested (and mostly by cross-sectional studies) and have, thus far, garnered preliminary support.

Scholars have also examined whether pro-crime or pro-gang attitudes increase the likelihood of gang joining. There is some cross-sectional support that pro-gang attitudes are associated with increased gang joining, but there has not been support for the notion that youth who perceive more benefits to gang joining are more likely to join (or that youth who perceive many cons to gangs will not join). There is cross-sectional and, more limited, longitudinal support for the notion that youth with higher tolerance toward deviance, antisocial beliefs, more extensive use of techniques of neutralization for crime, or more supportive attitudes for alcohol and drugs are more likely to join gangs. Related to this, youth who hold more conventional beliefs or a stronger belief in rules have been more often shown to refrain from gang activity.

Numerous theories posit that gang membership is a response to blocked opportunities, status frustration, or other negative life events. Generally, there is consistent support that exposure to negative life events is associated with gang joining, but perceived (future) opportunities have not shown the same pattern. Experienced barriers to success or strain have been a strong basis for theories of gang membership in qualitative studies. Support for this notion in quantitative studies, both cross-sectionally and longitudinally, suggests that it may matter for male gang joining, but not for females.

Finally, gang joining may also be predicted by an individual’s behaviors. As stated prior, the link between gang membership and increased rates of offending (e.g., general delinquency, violent offending and aggressive behavior, graffiti, official contact with justice system agencies, loitering/ hanging out/cruising, drug sales) has been well established. Much of this research is based on correlational studies where causal ordering could not be definitively established (i.e., which came first, the gang joining or the offending). However, cross-sectional studies that incorporate nondelinquent, problem behaviors and longitudinal studies that establish temporal ordering can be used to assess the degree to which behavior predicts gang joining. Sexual promiscuity or early sexual experiences, for example, have been found to predict gang joining. Alcohol use and drug use (both legal and illicit) and additionally, early alcohol or drug use have predicted gang joining.

Longitudinal studies have supported the notion that youth with delinquent histories may be more likely to join gangs, but the amplification effect of gang membership on the rate of delinquency (post-gang joining) has a very strong grounding. In other words, some studies show that gang members are more likely to have committed certain types of crimes before they joined gangs (e.g., violent, total delinquency), but it is also clear that the rate of offending significantly increases after youth are a part of the gang and then falls off after leaving. There has been very little support for the notion that involvement in prosocial activities decreases the likelihood of gang joining in either cross-sectional or longitudinal studies. Researchers have examined the role of religion (e.g., religious service attendance, “religiosity,” or bonds to religion), employment (during school year and during the summer), and community-based activities like sports or events without noteworthy results.

Family Domain

There is some cross-sectional and longitudinal evidence that youth in families of lower socioeconomic status are more likely to become gang members. There are, however, more studies that fail to find the connection. The same can be said of research on family structure. There is inconsistent support for the role of nuclear family structure versus single parent households (or other family structures) on the likelihood of gang joining. Scholars have suggested that what is more important than the structure of the family is the quality of the parenting within the family. Risk factor studies often test the degree or quality of parental monitoring or control on youth. There is strong cross-sectional and longitudinal support for the idea that more parental monitoring decreases the likelihood of gang joining. The same has not been shown of the level of familial attachment. There is limited cross-sectional and longitudinal support that youth that are less attached to his/her parent or family are more likely to join a gang.

Many alternative family domain factors have been identified and proposed, but strong statements regarding generalization cannot yet be made. In a review of major risk factor studies, Klein and Maxson (2006) declare some results as “inconclusive” because an equal number of studies find support, fail to find support, or are inconclusive. There are also a number of factors that have not been extensively tested or exclusively tested in cross-sectional studies, and therefore, definitive conclusions would be premature. The role of familial deviance is still unclear. There is some support that the lack of parental role models, the presence of parents with prodeviant norms, or antisocial family members like siblings may increase the likelihood of gang joining. However, studies have also failed to show a link between actual familial deviance/violence and gang joining. There is preliminary evidence that familial or parental conflict, family bonding, and familial involvement and support are not as strongly associated, but these factors have had limited scrutiny. Other conceptually appealing factors like the effect of relatives in a gang have been presented as important in numerous qualitative studies (e.g., Vigil 1988), but have had limited quantitative testing.

School Domain

Most school domain risk factors are still considered “inconclusive.” That is not to say that various factors have not been examined, but inconsistent results across studies and only a handful of longitudinal studies make it hard to generalize conclusions. While no variable can be argued as having “overwhelming” support, the factors with the most support at this point are educational aspirations, school achievement (as measured by grades at various times or testing), and attachment to school. These factors have been tested more often than others. The implication of these results is that the more strongly a youth is tied to school, the less likely he/she would become gang involved. There is preliminary cross-sectional and longitudinal support for the effect of negative labeling by teachers and schools on the likelihood of gang involvement. Studies have found that negative labels like “bad,” “disturbed,” and “disabled” are associated with gang involvement. The opposite effect was not shown. Being labeled as a “good student” did not buffer gang joining. Safety in schools has been tested less often and generally without much support.

Other factors that have been investigated in just a few studies and require more evidence before trends can be discerned are features like attachment to teachers, self-esteem about school, educational marginality, school environment, educational frustration, winning awards in school, truancy/absenteeism, participation in school sanctioned activities, and having gang members in the classroom.

Peer Domain

The peer domain seems particularly important given that gangs are a peer group. This is why temporal ordering is of particular concern to this domain of risk factors. Cross-sectional studies that report peer group characteristics at the same time as an individual reports gang membership could very well be reported on the qualities of the gang – which is interesting and informative but not predictive. Only longitudinal studies can decisively establish whether the characteristics of or commitment to certain peer group types increases the likelihood of future dealings with gangs.

While the wording of questions or scales can differ from study to study, peer domain variables have generally clustered around two types of measures: the quality of the peers present and the level of attachment to peers. In measuring the quality of the peers that are present, scholars generally attempt to capture whether youth surround themselves with delinquent and/or prosocial peers. Delinquent peers would be a risk factor, whereas the presence of many prosocial peers would indicate a potential buffer for gang joining. Cross-sectional studies often find that gang members are more likely to be surrounded by delinquent peers, but due to the lack of temporal ordering, it is not clear whether this is because their delinquent peers are, in fact, their fellow gang members. There is, however, also longitudinal support for the effect of delinquent peers to lead to gang membership. Fewer studies have looked at the buffering effect of having prosocial or good peers on the risk of gang joining. There is preliminary support that the presence of more prosocial peers is important though more work is certainly needed.

The second cluster of peer domain factors has been the extent of an individual’s commitment to their peers. These measures have been tested less often and generally with weaker methodology, but the pattern generally follows that of the quality of peers present. Youth that are more strongly committed to their negative/delinquent peers or who have low commitment to prosocial peers are more likely to be in gangs. Youth that are more strongly committed to conventional or prosocial peers are less likely to join gangs.

Other factors that have had limited examination, but show some promise as risk factors, include the presence of or ability to resist peer pressure, the experience of being approached by a gang, and the time spent unsupervised or “hanging around.” Future studies should begin to examine some of these factors more directly related to socialization with peers.

Community Domain

Despite being one of the first domains to emerge in the gang literature (mostly from qualitative studies), the community domain is one of the least often tested in the quantitative work on gang risk factors. In qualitative studies about gangs, the structural features of the community are quite prominent. Decades of research show that gangs often form in highly disadvantaged and disorganized communities that lack effective social controls. Therefore, it is known that community level factors should not be neglected as important predictors for individual-level gang membership. However, risk factor data are most commonly collected at the individual level (i.e., people are asked questions about themselves and their lives). These measures often capture an individual’s perceptions of their community characteristics. These perceptions may or may not be consistent with the actual state of their neighborhood condition. Therefore, the most common method for testing community factors may not be sufficient measures of the actual phenomenon and may fail to find support even if it does exist.

Two common community measures are the level of (1) crime and (2) disorganization or disorder in a neighborhood. Studies have shown mixed results in regard to the effects of these two characteristics on the likelihood of gang joining for youth. There are longitudinal studies that show that community arrest rate and neighborhood disorganization predict later gang joining, though other studies (longitudinal and cross-sectional) do not. Other community related factors have been investigated, but not extensively. There is at least some support to suggest that perceived access to drugs in the neighborhood, attachment to a neighborhood, availability of firearms in the community, drug use in neighborhood, degree of poverty of the community, and the presence of gangs in the community should be considered.

Risk Factor Theories Of Gang Membership

There are a number of theories of gang formation and gang membership. Many of the most popular theories are based on early qualitative studies on gang groups and their dynamics. Most of these theories fit under the broader umbrella of other major criminological theories. For example, many theories of gang formation have their roots in macro-crime theories like social disorganization, strain, and subculture of violence. Learning theories have also been widely cited as the mechanism to explain shared norms and their transmission among gang members.

More recently, however, the rise of longitudinal risk factor research has spurred developmental theories of gang membership. The risk factor approach assumes that there are multiple paths to the same outcome. The timing, order of risk factors, and cumulative nature play a dynamic role in the likelihood of gang joining in youth. These theories would not be possible (nor testable) without these multiple-wave, longitudinal studies that can capture stimuli at various points during development. Arguably, the most well-known theory and documented risk factor theory is Thornberry and colleagues’ (2003) Interactional Theory of Gang Membership.

As the name suggests, this theory stresses that interactions between variables and bidirectional causality are important. According to this theory, it is not the case that exposure to every type of risk factor directly influences gang members. Factors like the neighborhood and family are predictive of gang membership indirectly because they loosen prosocial bonds. The decline of the bonds increases the chances for antisocial influences or values. A culmination of risk factors and disadvantage increases stress, delinquency, and general problem behaviors. Engaging in these behaviors then increases the perceived benefits of gangs (e.g., excitement, protection) and, hence, increases the odds of gang joining.

Thornberry and colleagues found support for their model using the Rochester Survey of Youth. Their theory focuses mostly on late childhood and early adolescent predictors of mid-adolescent gang membership. Since their work, other proponents of a developmental framework to explain gang membership have argued that the theory should incorporate earlier risk factors like at preschool or upon school entry (see Howell and Egley 2005) because a number of longitudinal studies (using gang and delinquency in general) have shown risk factors measured at earlier ages to be important. Regardless, risk factor research has made it possible for the understanding of gang joining and membership to be placed within a developmental context.

Future Directions

Risk Factors For Gangs Vs. Other Serious Offenders

One important aspect of the gang risk factor literature that is still up for debate is the difference between gang members and nongang serious offenders. The studies reviewed here compared gang members with nongang youth. In this direct comparison, a number of significant differences can be found. The most noteworthy and consistent difference between gang and nongang members is involvement in delinquency and other problem behaviors. Gang members commit significantly more offenses than their nongang counterparts, and, as shown here, there are a number of risk factors that distinguish these two groups.

When gang members are compared to nongang, but deviant peer group members, their behavior is also shown to be significantly more delinquent. A number of gang scholars contend that gangs are substantively different than even nongang deviant peer groups. If this is the case, one would expect that there are unique risk factors that significantly distinguish gang members from these other serious offenders. There are a limited number of studies that have attempted to address this issue. In one study using the Denver Youth Survey (a longitudinal study of over 1,500 youth), the researchers found that while many factors distinguished gang members from the non-offending youth, few differences existed between the gang members and nongang street offenders. A separate study using a national sample of youth (n ¼ 5,935) used the same 18 predictors on violent offenders and gang members. Of the 18 factors, 6 of them predicted both violent offending and gang membership in multivariate models (i.e., low guilt, neutralizations, delinquent peers, delinquent peer commitment, time spent with drugs and alcohol, and negative school environment). Four additional variables predicted violent offending but not gang membership (i.e., impulsivity, risk seeking, few prosocial peers, and time spent without adults). No factors independently predicted gang membership.

Scholars have struggled to predict gang membership above and beyond what we already know about delinquency more generally. Risk factors for gang members are very similar to those for violent and/or serious offenders (despite differences in level of behavior). This leads to the question of whether there is reason to study gangs as separate entities from other highly delinquent youth groups. The study using a national sample of youth also found that gang members, on average, had more cumulative risk factors than even violent offenders. This may be an important point of inquiry that requires further investigation. Perhaps, it is not specific risk factors that predict gang versus nongang serious offenders, but the sheer magnitude of cumulative risk factors that makes the difference. Gang risk factor research needs more investigation into the mechanisms or factors that lead gang members to gang groups as compared to other deviant peer groups.

Longitudinal Studies Are Needed To Move The Field Forward

True risk factor studies need to be longitudinal. Cross-sectional studies are informative, often more feasible to execute, and can be used to add support to longitudinal findings. However, they will never be able to definitively establish causal ordering and, thus, can never identify a true “risk factor” or predictor of future behavior. With newer theories like Thornberry and colleagues (2003) interactional framework for gang membership, more attention has been focused on pathway models as explanations for behavior. This makes intuitive sense. Human behavior is a very complicated thing to predict. It is highly unlikely that all factors simply work in a unidirectional fashion. The progress that the gang risk factor literature and theory building has made in recent decades has come from a handful of well-executed, longitudinal studies.

Unfortunately, there have been a limited number of these types of data collection efforts. They are costly, time-consuming, and labor intensive to conduct. The data available from these established studies have also been used quite extensively to study gang risk factors. No data collection effort can provide an infinite number of gang studies. The major longitudinal study efforts currently available also represent a limited geographic section of America, and especially with regard to gang research, diversity in location and population is an important consideration. Therefore, without more longitudinal data collection efforts, new information would not be available and gang risk factor work will not move forward. To complicate matters, tests of interactional theories (as opposed to simply direct effects of risk factors) require analysis with sufficient statistical power (i.e., lots of people to study). This requires studies with even more resources and time commitment. These types of studies are a monumental undertaking, but without more efforts, this area of work cannot make forward progress.


  1. Craig WM, Vitaro F, Gagnon C, Tremblay RE (2002) The road to gang membership: characteristics of male gang and nongang members from ages 10 to 14. Soc Dev 11:53–67
  2. Esbensen FA, Deschenes EP, Winfree LT Jr (1999) Differences between gang girls and gang boys: results from a multisite survey. Youth Soc 31:799–828
  3. Esbensen FA, Huizinga D, Weiher A (1993) Gang and non-gang youth: differences in explanatory factors. J Contemp Crim Justice 9:94–116
  4. Hill KG, Howell JC, Hawkins JD, Battin-Pearson SR (1999) Childhood risk factors for adolescent gang membership: results from the Seattle Social Development Study. J Res Crime Delinq 36:300–322
  5. Howell JC, Egley A Jr (2005) Moving risk factors into developmental theories of gang membership. Youth Violence Juv Justice 3:334–354
  6. Klein MW, Maxson CL (2006) Street gang patterns and policies. Oxford University Press, New York
  7. Lahey BB, Gordon RA, Loeber R, Stouthamer-Loeber M, Farrington DP (1999) Boys who join gangs: a prospective study of predictors of first gang entry. J Abnorm Child Psychol 27:261–276
  8. Thornberry T (1998) Membership in youth gangs and involvement in serious and violent offending. In: Loeber R, Farrington DP (eds) Serious and violent juvenile offenders: risk factors and successful interventions. Sage, Newbury Park
  9. Thornberry TP, Huizinga D, Loeber R (2004) The causes and correlates studies: finding and policy implications. Juv Justice 9:3–19
  10. Thornberry TP, Krohn MD, Lizotte AJ, Smith CA, Tobin K (2003) Gangs and delinquency in developmental perspective. Cambridge University Press, Cambridge
  11. Vigil JD (1988) Barrio gangs: street life and identity in Southern California. University of Texas Press

See also:

Free research papers are not written to satisfy your specific instructions. You can use our professional writing services to buy a custom research paper on any topic and get your high quality paper at affordable price.


Always on-time


100% Confidentiality
Special offer! Get discount 10% for the first order. Promo code: cd1a428655