Assignment: Scope and Activities of Criminal Organizations

Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=rjqy20

Justice Quarterly

ISSN: 0741-8825 (Print) 1745-9109 (Online) Journal homepage: https://www.tandfonline.com/loi/rjqy20

Criminal and Routine Activities in Online Settings: Gangs, Offenders, and the Internet

David C. Pyrooz, Scott H. Decker & Richard K. Moule Jr.

To cite this article: David C. Pyrooz, Scott H. Decker & Richard K. Moule Jr. (2015) Criminal and Routine Activities in Online Settings: Gangs, Offenders, and the Internet, Justice Quarterly, 32:3, 471-499, DOI: 10.1080/07418825.2013.778326

To link to this article: https://doi.org/10.1080/07418825.2013.778326

Published online: 18 Mar 2013.

Submit your article to this journal

Article views: 4418

View related articles

View Crossmark data

Citing articles: 24 View citing articles

Criminal and Routine Activities in Online Settings: Gangs, Offenders, and the Internet

David C. Pyrooz, Scott H. Decker and Richard K. Moule Jr.

Crime and deviance reflect the dynamic nature of social life. The Internet has changed opportunities for crime and deviance, much as it has changed other aspects of social life. Accompanying the movement of offending and victimiza- tion to the Internet has been the expansion of deviant groups—including gangs—into online settings. Drawing from web-facilitated and web-enhanced classifications of online deviant behavior and identity, we extend the study of offending, gangs, and gang membership to online settings. Using data gathered in five cities from 585 respondents, including 418 current and former gang members, we study general online routine activities, online criminal and devi- ant behaviors, and gang-related online behaviors and processes. Based on our

David C. Pyrooz is assistant professor in the College of Criminal Justice at Sam Houston State University. He received his BS and MS in Criminology from California State University, Fresno and PhD in Criminology and Criminal Justice from Arizona State University. His research interests revolve around gangs and deviant networks, developmental and life course criminology, and vio- lent offending and victimization. He has authored articles that appear in Crime and Delinquency, the Journal of Quantitative Criminology, the Journal of Research in Crime and Delinquency, and Justice Quarterly. He is the author of Confronting Gangs: Crime and Community (Oxford, with G. David Curry and Scott H. Decker). Scott H. Decker is foundation professor of the School of Criminology and Criminal Justice at Arizona State University. He received his BA in social justice from DePauw University and MA and PhD in criminology from Florida State University. His main research interests are in the areas of gangs, criminal justice policy, and the offender’s perspec- tive. His most recent books include European Street Gangs and Troublesome Youth Groups (Winner of the American Society of Criminology, Division of International Criminology Outstanding Distinguished Book Award, 2006), Drug Smugglers on Drug Smuggling: Lessons from the Inside (2008, CHOICE Academic Press Book of the Year), and Criminology and Public Policy: Putting Theory to Work (2010, with Hugh Barlow). Richard K. Moule Jr is a doctoral student in the School of Criminology and Criminal Justice at Arizona State University. He received his BS in Criminology from The College of New Jersey and his MS in Criminology and Criminal Justice from Arizona State University. His research interests include gangs and deviant networks, life course criminology, and the intersection of technology and crime. He is currently the archivist for the Walter B. Miller Library, housed in the School of Criminology and Criminal Justice. Correspondence to: D.C. Pyrooz, Sam Houston State University, College of Criminal Justice, PO Box 2296, Huntsville 77341, USA. E-mail: [email protected]

JUSTICE QUARTERLY, 2015 Vol. 32, No. 3, 471–499, http://dx.doi.org/10.1080/07418825.2013.778326

� 2013 Academy of Criminal Justice Sciences

results, we arrive at three main conclusions: (1) gang members use the Inter- net and social networking sites as much, if not more, than their nongang coun- terparts, (2) gang members have a greater overall propensity for online crime and deviance than former and nongang respondents, based on our multivariate multi-level item response theory models, (3) the Internet is rarely used to further the instrumental goals of gangs, instead appealing to the symbolic needs of gangs and gang members. We conclude by discussing the conceptual and policy implications of these findings in relation to online activities of offenders and deviant groups.

Keywords gangs; the Internet; crime and deviance

Internet access and use is widespread. The number of global Internet users exceeds 2.2 billion (Internet World Stats, 2012). In the USA, with three-quar- ters of the population online, the Internet is a normative component of every-

day life (NTIA, 2011). Search engines like Google grant quick access to information, the social media that has accompanied Web 2.01 promotes digital

connectedness, and job applications, banking, shopping, and education are available online (Mehra, Merkel, & Bishop, 2004; Pasek, More, & Romer, 2008;

Wellman, Haase, Witte, & Hampton, 2001). There is a dark side associated with the expansion of the Internet that must

be contrasted with its social merits. The pirating of digital goods, the creation of viruses and cyber-attacks, victimization in the form of fraud, harassment, and stalking, and online sexual deviance, particularly involving children, are all

facilitated by computer-mediated communication (Fox, Nobles, & Akers, 2010; Holt, 2012; Holt, Blevins, & Burkert, 2010; Holtfreter, Reisig, & Pratt, 2008;

Wolfe, Higgins, & Marcum, 2007). Accompanying the rise of digital crime and deviance, criminal networks have seized on opportunities that the Internet

provides to further collective goals. An emerging literature has provided descriptive accounts of the nature and extent of online behaviors and pro-

cesses across a range of antisocial groups, including hate, terrorist, cult, and other antisocial and extremist groups (Corb, 2011; Dawson, 2009; Weimann,

2006; Zhou, Reid, Qin, Chen, & Lai, 2005). Crime is a reflection of the dynamic nature of social life. While the Internet

has changed opportunities for crime, criminologists have been slow to embrace

the study of new forms of offending. For example, the application of theory and method to the study of terrorism was integrated into mainstream criminol-

ogy only after September 11, 2001 (Agnew, 2010; Lafree, Dugan, & Korte,

1. The evolution of the Internet has been characterized by the concepts of Web 1.0 and Web 2.0 (Cormode & Krishnamurthy, 2008; O’Reilly, 2007). Web 2.0 emerged in the early 2000s after the dot.com bubble burst, and refers to the combination of advancing digital technologies, the growth of websites involving “social” components (profiles, friends, etc.), and content creation (videos, images) by Internet users. Web 1.0, in contrast, refers to the small concentration of content producers targeting a much wider base of content consumers during the 1990s.

472 PYROOZ ET AL.

2009; Rosenfeld, 2004). Because the Internet is a medium for communication, its capacities are at the hands of programmers and users who construct con-

tent and facilitate interaction. As such, it is important to understand how offenders adapt and use this powerful technology, and whether it parallels or

diverges from street offending. The online behavior of gangs and gang mem- bers is particularly important to understand in this context because (1) gang members fit the age group most likely to use the Internet, especially social

media, and (2) as criminological studies demonstrate, gang membership facili- tates increased involvement in crime. The Internet may be a venue for gangs

and gang members to expand criminal activities. We study the criminal and routine behaviors of gangs, gang members,

and nongang youth and young adults in online settings. Data were derived from in-person interviews with 585 respondents involved in or at-risk for

criminal justice system involvement in five cities. Drawing from online iden- tity and behavioral perspectives (Maratea & Kavanaugh, 2012; Tyler, 2002;

van Dijk, 2006), our study is guided by web-facilitated and web-enhanced classifications of online behavior and identity. The latter perspective leads us to a parallelism hypothesis: online behaviors and processes should con-

verge with “offline” settings. We examine this hypothesis in three ways, the first of which explores the online routine activities of gang and nongang

members, including use of the Internet, level of computer skills, and web- sites visited (e.g. social networking sites). The second set of questions

addresses involvement in criminal and deviant online activities, including piracy, harassment, threats, and the facilitation of drug sales, assault,

theft, and robbery. Multi-level item response theory (IRT) modeling is used to assess differential patterns in the overall propensity for online crime and deviance. The final set of questions explores the online presence of gangs

and the collective features of their online existence. As more components of our daily lives move online, this research integrates criminological theory

into our understanding of criminal and deviant individual and collective activity on the Internet.

Deviant groups online

The Internet introduces a vast social world to deviant groups.2 The advent of the Internet has fundamentally changed people’s ability to communicate and access information. Blogs, tweets, Facebook updates, news media message

boards, and instant messaging create immediate connections that span cities and continents. The virtues of a digitally connected world are substantial,

2. Terms such as organization, network, group, and collective, and other indicators are often used to convey some degree of “group-ness.” For the present purposes, we use the term “deviant groups” to capture the spectrum of collectives who engage in unconventional and extreme behaviors and ideologies.

CRIMINAL AND ROUTINE ACTIVITIES IN ONLINE SETTINGS 473

though such a value-neutral medium is an appealing platform for deviant groups. Indeed, Weimann (2004) argued that limited regulation, anonymity,

the rapid flow of information, and large audiences provide a setting for groups such as terrorists to flourish, as these features meet the instrumental and sym-

bolic needs of deviant groups. Therefore, it should come as no surprise that by the early 2000s, terrorist (Weimann, 2004; Weimann, 2006), cult (Dawson & Hennebry, 1999), hate and racial supremacy (Gerstenfeld, Grant, & Chiang,

2003; Sela-Shayovitz, 2012a; Zhou et al., 2005), and other extremists groups maintained an online presence.

Research has advanced from documenting the presence of deviant groups online to elaborating on how these groups use the web. Such groups have

harnessed the communication capabilities of the Internet in multiple ways. Foremost, many deviant groups have online networks of websites and forums

(Burris, Smith, & Strahm, 2000; Zhou et al., 2005). These websites create a navigable path between and within groups, and between “believers” and the

curious or sympathetic. Among the central features of these websites are recruitment and the dissemination of propaganda (Corb, 2011; Reid & Chen, 2002). Prior to the advent of the Internet, these groups relied on rudimen-

tary and localized methods, including AM radio, public access television, print materials, music, and in-person networks (Dawson & Hennebry, 1999;

Schafer, 2002). Contemporary recruitment and propaganda dissemination strategies have expanded beyond local or regional audiences to whole coun-

tries and continents, and have advanced in technological complexity and appeal (Burris et al., 2000; Reid & Chen, 2002; Zhou et al., 2005). Indeed,

web access has facilitated new methods of recruitment and propaganda dis- semination, including the creation of online video games and the merchan- dising of music and literature (Dawson & Hennebry, 1999; Gerstenfeld

et al., 2003; Selepak, 2010). While these methods are largely passive, rely- ing on online social or ideological homophily, web-based recruitment and

propaganda is troubling because it can facilitate or reinforce radicalized beliefs without human contact and promote “lone wolf” radicalized behav-

iors (Spaaij, 2010). Gangs share many characteristics with deviant groups who have an online

presence, including weak organizational structure, high turnover rates in mem- bership, and marginalization (Curry, 2010; Decker & Pyrooz, 2011; Horgan &

Bjorgo, 2009; Klein & Maxson, 2006). The absence of political and religious ide- ology among gangs, in combination with their street and youth orientation, might challenge their ability to exploit the Internet in a manner similar to

other deviant groups. Papachristos (2005, p.53) held that “few gang members ever discuss or mention the Internet. Many don’t possess the hardware, soft-

ware, or technical skills (not to mention the necessary telephone lines) to manage the Web.” Times have changed since Papachristos’s statement, as the

structural barriers hampering Internet access to disadvantaged populations have declined (DiMaggio, Hargittai, Celeste, & Shafer, 2004; Mossberger,

Kaplan, & Gilbert, 2008). Additionally, there is considerable overlap in the age

474 PYROOZ ET AL.

demographics of Internet users and gang members (Pew, 2012). As a result, we would expect a noticeable presence of gangs and gang members online.

Conceptual framework

We draw upon perspectives about online identity and behavior to guide our

understanding of gang processes and activity on the Internet (Maratea & Kava- naugh, 2012; McKenna & Bargh, 1999; 2000; Smith & Kollock, 1999; Tyler,

2002; van Dijk, 2006). As technology is integrated into the normal functions of everyday life, it presents a new way of doing old things (Tyler, 2002). There

are, however, important distinctions in how identity and behavior manifests online.

Web-facilitated online identities and behaviors may reflect offline realities, but there is heterogeneity across context (Tyler, 2002; van Dijk, 2006; Waskul

& Martin, 2010). This may mean experimentation with secret or idealized iden- tities online (Redmon, 2003; Waskul, 2003; Waskul & Martin, 2010). Some real- world experiences can easily translate into deviant online subcultures, where

“secret selves” are expressed away from friends, family, co-workers, and law enforcement (see Holt, 2007) that may include deviant identities. Online sub-

cultures assist in the construction and reification of online identities, where deviance is facilitated and supported by online communities of like-minded

individuals (e.g. Durkin, Forsyth, & Quinn, 2006) but maintains some level of anonymity. This approach has been applied to a number of deviant, primarily

sexual, behaviors, but suggests a unidirectional theoretical approach; such identities and resulting behaviors are unlikely to exist without the web (Blevins & Holt, 2009; Quinn & Forsyth, 2005).

Web-enhanced online identities and behaviors, alternatively, are reflections of offline identities and behaviors, thus one is the analog of the other. There

is substantial overlap in offline and online identities and behaviors, and there is less of a need to necessarily hide the deviant self. For individuals, this

integration creates fluidity between the digital and offline worlds (Maratea & Kavanaugh, 2012). Weimann (2006) described how various terrorist organiza-

tions engaged in ideological and religious debates online. These debates were able to frame, and likely facilitate, acts of violence throughout the Middle East

(see also Daniels, 2009, on race supremacy groups). Such instances illustrate a blurring of the online and offline worlds, and thus a bidirectional relationship between online and offline identity and behavior.

We view gang-related activities and processes on the Internet as consistent with the web-enhanced perspective of online and offline identity and behavior.

“Gang” identity is salient in both the real world and online—it carries with it a degree of “street cred” acknowledged by gang and nongang members alike,

despite its deviance. In contrast to other deviant identities, there is little reason to hide gang membership. To do so is to lose any such credibility. But

CRIMINAL AND ROUTINE ACTIVITIES IN ONLINE SETTINGS 475

identity theory itself does little to suggest whether gang members would differ from others in every form of Internet behavior. Rather, decades of research on

gang processes and behavior suggest that this identity will matter most with respect to deviance. Thus, a parallelism hypothesis guides our hypotheses:

gangs and gang members should behave online much like they do offline. The stakes are high on the Internet. Gangs can organize drug distribution,

coordinate attacks, stake out new territory, and replenish their ranks through

recruitment with the click of a mouse. Yet, a consistent finding in the litera- ture on gangs is that their organizational structure tends to be informal and

diffuse, drug sales tend to be entrepreneurial rather than corporate, violence tends to be sporadic rather than well planned, and leaders and rules are hard

to come by (Decker & Pyrooz, 2013). Therefore, the parallelism hypothesis intimates gangs are organized and behave on the Internet in ways that are

similar to the street. Websites should be poorly maintained and updated infrequently because their technological competencies are unspectacular,

Internet-mediated communication should be limited because of concerns regarding law enforcement scrutiny, and recruitment rare because anonymity undermines trust. Much like the street (Decker & Van Winkle, 1996), online

gang behavior should be more symbolic than instrumental because of limited group structure and regulation of behavior.

While gangs are unlikely to exploit the Internet to meet their collective goals, it does not mean that their online presence is limited or marginalized.

The standard online routine activities—e.g. hours online and use of social media—of gang members should be indistinguishable from similarly situated

peers. The Internet is useful to people of all walks of life, and gang members are no exception. Researchers have long argued that the day-to-day routines of gang youth are unexciting and very similar to nongang youth (e.g. Klein,

1971). Because the value-neutral medium of the Internet does not inherently provoke the oppositional nature of gang members, patterns of use, content

accessed, and noncriminal behaviors on the Internet will likely resemble the online repertoire of their nongang peers.

When turning to crime and deviance in online settings, a different story should emerge. The Internet is an emergent social field that functions as a vir-

tual convergence space for offenders and targets (Felson, 2006; Newman & Clarke, 2003; Pratt, Holtfreter, & Reisig, 2010). Consistent with criminal pro-

pensity theories, the offending profiles of individuals with high-criminal pro- pensities that select into gangs are unlikely to be dampened by the limited, unregulated barriers of the Internet (e.g. Gottfredson & Hirschi, 1990). Crimi-

nal propensities play out in online forums, as opportunities to bully, harass, steal, and other online exploits are abundant on the Internet. Consistent with

theories of socialization, the normative orientation of the gang provides codes and scripts that dictate and reinforce online behavior (Thornberry, Krohn,

Lizotte, Smith, & Tobin, 2003). The Internet appeals to the instrumental and symbolic needs of the gang and gang members in the form of individual or

group status enhancements or in response to status threats (Decker, 1996;

476 PYROOZ ET AL.

Hughes & Short, 2005; King, Walpole, & Lamon, 2007). Virtual social worlds could be a deadly medium, facilitating gun acquisitions and drug distribution,

as well as establishing new conflicts and exacerbating existing ones because of overlapping convergence in spaces with gang rivals. Essentially, the line

between street and online worlds may be blurred, but the key is determining whether the online behaviors of gang members are age-normative and consis- tent with their peers.

Existing research

The implicit assumption about the presence of gangs and gang members in online settings is that the Internet is a new venue for criminal and deviant activ-

ity. Law enforcement and legal commentary on this topic recognize the pres- ence of gangs online, but are ambivalent about the extent and nature of the

problem. On the one hand, in the FBI’s report by the National Gang Intelligence Center (2011), there is concern that gangs were not only online but also elevat- ing their technological sophistication of criminal involvement to new levels (see

also Hanser, 2011; O’Deane, 2011). Based on their “visual inspection” of social networking sites, the NGIC (p. 41) noted that “social network, microblogging,

and video sharing websites—such as Facebook, YouTube, and Twitter—are now more accessible, versatile, and allow tens of thousands of gang members to eas-

ily communicate, recruit, and form new gang alliances nationwide and world- wide.” On the other hand, documents, videos, and communications online often

leave a “digital trail” that is useful for law enforcement. After identifying numerous signs and symbols unique to a host of gangs online, Knox (2011, p. 6) held that while “gang beliefs and values [are] corrupting the entire idea of an

“internet community,” the online presence of gangs presents opportunities for intelligence gathering among police agencies to build criminal cases.

Content analyses of the websites, social networking pages, and twitter feeds involving gangs shed light on the uses of the Internet by gangs and gang mem-

bers. Morselli and Decary-Hetu (2012; Decary-Hetu & Morselli, 2011) conducted a detailed keyword search of 56 gang names common in Canada on Facebook,

MySpace, and Twitter. Additional terms of gangs, drugs, violence, and crime were added to the queries for additional discriminatory power. The number of

tweets, fans, friends, and profile views that met the search terms were recorded, as well as the number of gangs observed on these pages. Based on their findings, Morselli and Decary-Hetu held that the Internet provides a “new

channel for publicity,” where gangs and other deviant groups no longer have to rely on word of mouth to promote their reputations and activities. In

Womer and Bunker’s (2010) analysis of the social networking pages of Sureno gangs and Mexican drug cartels, they also observed that gangs posted pictures

and videos of themselves, posing with firearms and drugs, and referencing criminal activities. These authors, as well as others (see Decker, van Gemert,

CRIMINAL AND ROUTINE ACTIVITIES IN ONLINE SETTINGS 477

& Pyrooz, 2009; Sela-Shayovitz, 2012b), commented that online images help glorify gang culture and possibly encourage participation offline.

Based on these works, we have established that gangs maintain an online presence. But there is a “dark figure” to this information because it is limited to

what gangs and gang members make publicly available. When we asked whether their gang used the Internet, gang members from Los Angeles and St. Louis told us: “[We] don’t talk about it [gang business] because the police is on there” and

“That’s a no no. Only idiots do that. Why would you do that?” Gang members are keenly aware that police are monitoring their online activities, potentially

skewing our understanding of Internet gang activity. To date, two studies have supplied evidence from surveys of gang members’ use of the Internet. Based on

a survey of an urban student population about their Internet use, King et al. (2007) reported that 137 of the 100,000 respondents were gang associates. Key

findings were that 25% of gang members used the Internet for 4 hours a week, community centers were the location where 45% of gang members accessed the

Internet, and 70% of gang members found friendship was easier to establish online than in person. Sela-Shayovitz (2012a) reported a high level of reliance on the Internet, sophistication in online offending, and collective activities

among 30 gang members she interviewed in Israel. Absent a shared denominator or a comparison group, it is difficult to determine how the patterns among gang

youth in these studies converge or diverge with their nongang peers.

The current study

Deviant groups maintain an active online presence and use the Internet to fur- ther their collective goals. It is less clear, however, how gangs are using the

Internet, what gang members are doing online, and how their behaviors differ from their nongang peers. If crime reflects the dynamic nature of social life

and the Internet presents a new set of opportunities for offenders and deviant groups, it is important to understand the extent and nature of online gang

activity. What is needed is an empirical study of how adolescents and young adults—including gang and nongang youth—go about their criminal and routine

activities online. Such a study has implications not only for the emerging social world of the Internet, but also for online and criminological theories. Does

online behavior parallel offline realities? Answering such a question yields important insight into an emerging form of criminal and deviant behavior.

This study has three goals. First, we explore the prevalence of Internet

usage among gang and nongang youth and young adults. To examine online routine activities, individuals are surveyed about the amount of time they

spend online, their technological capacities, and what they do online, includ- ing their use of social networking sites like Facebook. Answers to this line of

questioning tap immersion in online activity, foreshadow potential online offending opportunities, and help us sort out whether gang members are using

the Internet in an age-normative manner. Our second set of questions relate

478 PYROOZ ET AL.

to criminal and deviant online activities. Because of what we know about gang members offline (i.e. on the street) and the online conceptual foundation pre-

sented above, we explore whether gang members have a higher propensity to engage in criminal and deviant activities online than their nongang peers.

Finally, we examine gang-related online behaviors and processes at the group level. If gangs are an online threat, they should exhibit an online presence and exploit the Internet for gang purposes, such as organizational functions. In

combination with direct surveys of gang members, these three sets of ques- tions provide an important conceptual and empirical advancement in the liter-

ature on gangs, deviant groups, crime and deviance, and the Internet.

Methods

Data

The sample used in this study consists of 585 respondents interviewed about their use of the Internet and involvement in gangs. The data were obtained

from interviews conducted with youth and young adults in five US cities: Cleve- land, OH; Fresno, CA; Los Angeles, CA; Phoenix, AZ; and St. Louis, MO. This

specific study was part of a larger project investigating gangs’ use of the Inter- net initiated by a large technology corporation. Respondents were interviewed

in settings chosen to include a large number of individuals with involvement in gangs and criminal behavior. All sites included individuals who were actively

involved in gangs, as well as those who have disengaged from gangs or who have avoided gang involvement entirely. In Cleveland, Los Angeles, and Phoe- nix, populations of individuals at high risk for involvement in crime in the com-

munity were interviewed. The respondents in Cleveland were the clients of street outreach workers; individuals at risk for involvement in crime whose

behavior was being monitored by a noncriminal justice agency. The majority of the Cleveland respondents had not been involved in the criminal justice sys-

tem. The respondents in Los Angeles and Phoenix were participants in street outreach programs. The goal of these agencies is to work with gang members

and former offenders seeking to change their lives. Many of their clients had extensive involvement in the criminal justice system. In Fresno, interviews

were conducted with a jail population, representing a deep end sample of indi- viduals arrested and processed by the criminal justice system. In St. Louis, interviews were conducted with individuals on probation or parole. This popu-

lation represents a group in the community under active supervision by the criminal justice following adjudication.

Interviews were conducted in 2011 in these cities over the course of several months.3 Trained members of the research project staff administered surveys

3. IRB approval was obtained so long as confidentiality and anonymity were maintained, as they were. Informed consent was obtained from each participant before interviews began.

CRIMINAL AND ROUTINE ACTIVITIES IN ONLINE SETTINGS 479

face-to-face in private locations at the facilities of each research site (e.g. classrooms and attorney rooms). Most surveys were completed within 45 min to

one-hour. In the rare case that a respondent did not speak English, they were assigned to a Spanish-speaking interviewer. Very few individuals refused to

participate in the study. In some cases, respondents declined to answer spe- cific items in the questionnaire. Respondents in street settings were provided a small monetary incentive or store coupon for participating that did not exceed

$25, but this was not permitted in the jail. By purposively surveying youth and young adults in these settings, these

data have several strengths, the first of which is that they provide rich insight into self-reported Internet behaviors unfiltered by publicly accessible webpage

content. Second, unlike population samples containing low prevalence rates of gang membership and offenders, these data include a large number of gang-

involved respondents and offenders to allow for a comprehensive investigation into online behavior. Finally, and perhaps most importantly, such a sampling

strategy accounts for considerable unobserved individual heterogeneity, which is built into population surveys. In other words, the high-risk nature of the sample aids in strengthening the validity of the findings in the current study

because the respondents are drawn from equally high-risk environments.

Outcome variables

We examine three sets of outcomes related to the activities of individuals and

gangs in online settings. First, with regard to noncriminal online routine activities, Internet use prevalence was assessed by asking respondents if they “go online to access the Internet or send and receive email” (Pratt et al.,

2010), where “yes” and “no” responses were coded 1 and 0, respectively. Among those who answered yes (N = 464, or 79 percent), follow-up questions

inquired about their routine activities online, including time spent, shopping, viewing videos, social network usage, and technological capacity. Internet use

frequency measures weekly time spent on the Internet, ranging from 0 to 30 hours. Responses were mean-adjusted based on five categories

(e.g. 5–10 hours = 7.5 hours) to approximate a meaningful distribution. Online shopping prevalence is a binary item asking if respondents “purchased items

from online websites” in the last year (Yes = 1, No = 0). YouTube viewing preva- lence is a binary item asking if respondents “go on YouTube or other sites to watch videos” (Yes = 1, No = 0).4

Social network use prevalence is a binary item asking if respondents have accounts and use social networking websites, including Facebook, MySpace,

Twitter, and Other SNS (Yes = 1, No = 0). Similar to internet use frequency, social network use frequency is an interval measure of weekly time spent on

4. A small fraction of the sample was not presented with this question (16%), as it was inserted in the survey at a later date from the study initiation.

480 PYROOZ ET AL.

social networking sites and was mean-adjusted to approximate a meaningful distribution, ranging from 0 to 21 hours. Finally, technological capacity tapped

respondents’ “knowledge and skills in using computers,” ranging from 0 (“I do not use them unless I absolutely have to”) to 3 (“I can use Linux, most soft-

ware, and fix most computer problems”) (Holt & Bossler, 2009). Our second set of questions relate to criminal and deviant online activities.

Eight questions were included to inquire whether the respondents engaged in

these activities over the six months prior to the interview, including (1) ille- gally downloading media or computer software, (2) selling stolen property on

websites, (3) setting up drug sales or purchases online, (4) harassing or threat- ening someone on comment forums, social network sites, blogs, or chat rooms,

(5) coordinating assaults through email or social networking sites, (6) searching social networking sites to steal from or rob people, (7) posting or uploading

deviant videos of fights or threats on sites like YouTube, and (8) targeting or attacking someone in the street because of events occurring online. These

items were derived from various sources, including the online deviance litera- ture, commentary on gangs online, conversations with site managers, and assumptions about technology and offending complexity based on gang

research (Curry, Decker, & Pyrooz, 2013; Gogolak, 2012; Klein & Maxson, 2006; Knox, 2011; May, Bossler, & Holt, 2011; Skinner & Fream, 1997). All of the

items are binary and positively related to each other (mean inter-item r = .25). As we soon describe, the multi-level framework produces a latent variable of

individual propensity to engage in online crime and deviance, consistent with studies on gang membership (Krohn & Thornberry, 2008; see also Melde &

Esbensen, 2012; Pyrooz & Decker, 2012). Because this is an initial inquiry into online behaviors, our analytic strategy is appropriate and offending (dis)aggre- gation remains a topic for future studies to consider.

The last set of questions is exclusive to gang-related processes and behaviors. For those with a history of gang membership, we ask four group-level questions

about the online activities of their current or former gang and four individual- level questions about their personal use and perceptions of gangs and the Inter-

net. The group-level questions are all binary, asking if their gang has a website (or social network page), gang organizes online (e.g. meetings, drug sales, par-

ties, and assaults), gang recruits online, and gang posts videos online (e.g. music and fights). The individual-level items ask if the respondent searches for

gang information online and watches gang-related videos. In addition, respon- dents are asked about the importance of the Internet to their gang and to other gangs, where 0 = “not important,” 1 = “somewhat or very important.”

Explanatory variables

Our key explanatory variables in this study concern gang membership. Gang membership was measured in two stages. First, all respondents were asked if

they have “ever been a member of a gang?” Second, those who responded

CRIMINAL AND ROUTINE ACTIVITIES IN ONLINE SETTINGS 481

“yes” to a history of gang membership were next asked whether he or she was “currently in a gang?” Those who responded “yes” to both items were recorded

as current gang members, while those answering “yes” to item 1 and “no” to item 2 were recorded as former gang members. Both items are binary, with

respondents answering “no” to both items serving as the reference category. Self-nomination is a reliable and valid method for operationalizing gang mem- bership (Esbensen, Winfree, He, & Taylor, 2001; Thornberry et al., 2003), and

important differences in attitudinal and behavioral profiles further require that we distinguish between current and former gang members (Katz, Webb, &

Decker, 2005; Sweeten, Pyrooz, & Piquero, 2012).

Control variables

Several demographic control variables are also included in our analysis, includ-

ing age (in years), male ( = 1, “female” = 0), foreign born ( = 1, “US born” = 0), racially Black (=1, “non-black” = 0), ethnically Hispanic (=1, “non-His- panic” = 0), and racially/ethnically other (=1, “non-other” = 0), with whites

serving as the reference category. The typical respondent is in their 20s, male, black or Hispanic, and born in the USA. Given our purposive sampling strategy,

these are characteristics we would expect to find of respondents drawn from inner cities with involvement or high-risk for involvement in the criminal jus-

tice system.

Analytic strategy

The results of this study are presented in three stages, moving from individual- level online routine activities to criminal and deviant activities, and then

gang-related activities. In the first and third stages of the analyses, simple univariate and bivariate statistics are presented. These findings establish the

nature and patterns of online activities—for individuals and groups—and whether these activities differ across demographic controls and gang member-

ship status (current, former, and nongang respondents). In the second stage of the analysis, multi-level logistic IRT modeling is used

to relate gang membership status to crime and deviance online. A multi-level IRT framework offers several advantages to the study of online crime and devi- ance (Osgood, McMorris, & Potenza, 2002; Raudenbush, Johnson, & Sampson,

2003). First, by virtue of the IRT measurement model, our outcome is a latent variable representing individual propensity to engage in crime and deviance

online (i.e. theta). As a result, it captures the general tendency to offend online. Second, affirmative responses to each item in the measurement model

are not only a function of individual propensity but also item seriousness (or base offending rates), preventing less severe forms of offending from dominat-

ing the model. Third, the random-effects component in multi-level modeling

482 PYROOZ ET AL.

allows us to relate covariates to between-person variability in offending propensities. Studies of online crime and deviance tend to model specific

offenses (e.g. piracy and bullying) over general forms of offending, therefore a multi-level IRT approach is an important step forward in our understanding of

this increasingly important form of criminal behavior. As a mixed-effects approach, individual online crime and deviance items

(level 1) are nested within respondents (level 2). Our level 1 equation takes

the following form:

Log½oddsðYij ¼ 1Þ� ¼ b0j þ X8

i¼1 bijDij

A dummy coding scheme is used for the eight online crime and deviance items, therefore our level 1 outcome refers to the log odds of a “yes” response

for person j to committing offense i. Responses are determined by the latent propensity for online crime and deviance ðb0jÞ and item seriousness ðbijÞ. The latent determinant of offending is allowed to vary across respondents (u0j),

while item seriousness ðbijÞ coefficients remain fixed to adjust for their base rates in the sample. Our level 2 equations are:

b0j ¼ c00 þ c01 Current Gang1j þ c02 Former Gang2j þ . . . þ u0j

bij ¼ ci0

The goal in Equation 2 is to explain between-person variability in online offending propensities. To do so, we include covariates for gang membership

status, demographic controls, and online routine activities. All analyses were carried out in Stata 12.0 using xtmelogit (StataCorp, College Station, TX).

Results

Do online routine activities differ by gang membership status?

Table 1 presents the findings for the online routine activities by gang member-

ship status. About four-fifths of the sample reported using the Internet, a rate that exceeds the general US population (NTIA, 2011). Given the strong link between age and Internet usage, we would expect to observe such high rates.

No differences are observed across gang membership status, which means current, former, and nongang youth and young adults are using the Internet at

roughly equal rates. The Internet-using subsample is the focus of the remain- der of our analyses.

We observe mixed evidence to support the contention that gang members use the Internet for general activities in a manner that differs from their

CRIMINAL AND ROUTINE ACTIVITIES IN ONLINE SETTINGS 483

T a b le

1 D e m o g ra p h ic

c h a ra c te ri st ic s a n d n o n c ri m in a l o n li n e ro u ti n e a c ti v it ie s b y g a n g m e m b e rs h ip

st a tu s

C u rr e n t g a n g

F o rm

e r g a n g

N o n g a n g

S ig . d if f.

M e a n / (S D )

M e a n / (S D )

M e a n / (S D )

A B

C

F u ll sa m p le

(N = 5 8 5 )

N / %

1 7 4 / 3 0 %

2 4 4 / 4 2 %

1 6 7 / 2 8 %

A g e

2 4 .9

(7 .9 )

2 8 .3

(9 .7 )

2 2 .7

(8 .0 )

⁄ ⁄

⁄

M a le

8 2 .2

8 4 .4

7 9 .6

n s

n s

n s

B la c k

3 1 .6

3 6 .1

4 1 .9

# #

n s

H is p a n ic

6 0 .9

5 2 .0

3 6 .5

⁄ ⁄

⁄

R a c e / e th n ic it y : O th e r

4 .6

5 .7

7 .8

n s

n s

n s

F o re ig n b o rn

8 .6

6 .6

1 0 .8

n s

n s

#

In te rn e t u se

p re v a le n c e (% )

7 8 .2

7 8 .7

8 1 .4

n s

n s

n s

In te rn e t S u b sa m p le

(N = 4 6 4 )

N / %

1 3 6 / 2 9 %

1 9 2 / 4 1 %

1 3 6 / 2 9 %

In te rn e t u se

fr e q u e n c y (h o u rs )

7 .7

(7 .8 )

9 .1

(8 .7 )

6 .0

(5 .3 )

# ⁄

⁄

O n li n e sh o p p in g p re v a le n c e (% )

2 8 .7

3 5 .4

3 3 .1

n s

n s

n s

Y o u tu b e v ie w in g p re v a le n c e (% )

8 5 .6

8 6 .2

9 3 .3

n s

n s

n s

S o c ia l n e tw

o rk

u se

p re v a le n c e (% )

8 0 .1

7 9 .2

8 4 .6

n s

n s

n s

F a c e b o o k u se r (% )

6 9 .1

7 0 .8

7 2 .1

n s

n s

n s

M y S p a c e u se r (% )

4 1 .2

3 5 .4

4 7 .8

n s

n s

⁄

T w it te r u se r (% )

1 4 .7

1 0 .9

2 1 .3

n s

n s

⁄

O th e r S N S u se r (% )

2 7 .9

2 9 .2

1 6 .2

n s

⁄ ⁄

S o c ia l n e tw

o rk

u se

fr e q u e n c y (h o u rs )

5 .5

(6 .9 )

5 .0

(6 .7 )

3 .8

(5 .5 )

n s

⁄ #

T e c h n o lo g ic a l c a p a c it y (0 – 3 )

1 .2

(0 .8 )

1 .4

(0 .8 )

1 .3

(0 .8 )

n s

n s

n s

⁄ p < .0 5 , ] p

< .1 0 .

N o te . In d e p e n d e n t sa m p le

t- te st s a n d c h i- sq u a re

te st s w e re

u se d to

d e te rm

in e b e tw

e e n g ro u p m e a n d if fe re n c e s.

A = c u rr e n t to

fo rm

e r g a n g st a ti st ic a ll y si g n ifi c a n t d if fe re n c e

B = c u rr e n t to

n o n g a n g st a ti st ic a ll y si g n ifi c a n t d if fe re n c e

C = fo rm

e r to

n o n g a n g st a ti st ic a ll y si g n ifi c a n t d if fe re n c e

484 PYROOZ ET AL.

nongang and former gang peers. On the one hand, gang-involved respondents shop online, watch videos on YouTube, and access social networking sites at a

rate that is indistinguishable from those who are not in gangs. Further, all three groups indicate similar competencies in their technological skills, with

the modal respondent reporting they can “surf the web and use basic soft- ware.” On the other hand, we find that current and former gang members spend more time on the Internet and social networking sites on a weekly basis,

about 25–50% more hours.5 We also see that all three groups use the most popular social networking site—Facebook, with over one billion users—at

equal rates, but this changes when we turn to MySpace, Twitter, and other social networking sites. Notably, over one-quarter of current and former gang

members, compared to one-sixth of nongang respondents, report using alterna- tive sites, such as YouTube, Hoodup, and WorldstarHipHop.

From these findings we reach several conclusions. First, gang members are online. We would expect this, as technological diffusion occurs and these tech-

nologies are incorporated into everyday life (Rogers, 2005), even youth and young adults at the economic and social margins should utilize the Internet. Gang membership has no inhibitory influence on stifling these processes. Sec-

ond, gang members spend considerable time online. While the prevalence of Internet and social networking use are similar, the frequency of usage for cur-

rent and former gang members exceeds that of nongang respondents. These differences remain even in a multivariate context with demographic controls.

Thus, if Facebook and related sites function as convergence spaces for criminal and deviant activity, gang members are situated within them, although it is

less clear how they operate in those spaces. Third, with the above points in mind, this sample lacks the technological competencies to wreak the level of havoc that is concerning to security professionals. While we find little evidence

to support the concerns of the FBI that gangs use the Internet for “identity theft, computer hacking, and phishing schemes” (NGIC, 2011, p. 41), it does

not mean that gang members do not present problems to online communities.

Do gang members engage in more online crime and deviance than their peers?

Table 2 reports the descriptive and bivariate statistics for online criminal and deviant activities. The illegal downloading of music and other media was the modal response category among the respondents, whereas searching social net-

working sites to steal from or rob people was the least endorsed category. We observe a considerable amount of online crime and deviance, with 45% of the

sample engaging in at least one form of offending in the last 6 months. Small proportions of respondents report that they facilitate drug sales, sell stolen

5. These differences remain remarkably similar when holding constant demographic control variables.

CRIMINAL AND ROUTINE ACTIVITIES IN ONLINE SETTINGS 485

T a b le

2 C ri m in a l a n d d e v ia n t a c ti v it ie s b y g a n g m e m b e rs h ip

st a tu s (N

= 4 6 4 )

C u rr e n t g a n g

F o rm

e r g a n g

N o n g a n g

S ig . d if f.

M e a n o r % (S D )

M e a n o r % (S D )

M e a n o r % (S D )

A B

C

Il le g a l d o w n lo a d s (% )

3 3 .1

– 2 2 .4

– 2 3 .5

– ⁄

# n s

S e ll in g st o le n p ro p e rt y (% )

4 .4

– 3 .1

– 3 .7

– n s

n s

n s

D ru g sa le s (% )

6 .6

– 2 .6

– 3 .7

– #

n s

n s

H a ra ss m e n t (% )

1 1 .8

– 9 .4

– 8 .1

– n s

n s

n s

C o o rd in a te

a ss a u lt s (% )

8 .1

– 3 .6

– 5 .1

– #

n s

n s

S e a rc h to

st e a l/ ro b (% )

4 .4

– 1 .6

– 0 .7

– n s

# n s

U p lo a d d e v ia n t v id e o s (% )

1 4 .0

– 3 .6

– 5 .9

– ⁄

⁄ n s

O n li n e -s tr e e t a ss a u lt

(% )

1 3 .2

– 8 .3

– 2 .9

– n s

⁄ ⁄

O n li n e v a ri e ty

sc o re

(8 it e m s)

0 .9 6

(1 .6 2 )

0 .5 5

(0 .9 9 )

0 .5 4

(0 .9 3 )

⁄ ⁄

n s

N / %

1 3 6 / 2 9 %

1 9 2 / 4 1 %

1 3 6 / 2 9 %

⁄ p < .0 5 , # p < .1 0

N o te . In d e p e n d e n t sa m p le

t- te st s a n d ch

i- sq u a re

te st s w e re

u se d to

d e te rm

in e b e tw

e e n g ro u p m e a n d if fe re n c e s.

A = c u rr e n t to

fo rm

e r g a n g st a ti st ic a ll y si g n ifi c a n t d if fe re n c e

B = c u rr e n t to

n o n g a n g st a ti st ic a ll y si g n ifi c a n t d if fe re n c e

C = fo rm

e r to

n o n g a n g st a ti st ic a ll y si g n ifi c a n t d if fe re n c e

486 PYROOZ ET AL.

property, harass and threaten people, and upload deviant videos in online settings. We also observe that online conflict is not limited to the web. Eight

percent of Internet users reported that online settings are a source for prob- lems that spill over into offline settings.

When we disaggregate the sample by gang membership status, we find that current gang members have elevated online offending profiles. They illegally download media, sell drugs, coordinate assaults, and upload deviant videos at

a higher rate than former gang members. Compared to nongang respondents, they illegally download media, search social network sites to steal and rob,

upload deviant videos, and participate in online-to-street conflicts at higher rates. One might have expected more pronounced differences by gang mem-

bership status across the offense categories, though this could be due to the 6 month exposure frame. Turning to an online crime and deviance variety

score6 reveals clear differences—the offense rate among current gang mem- bers that is about 70% greater than their former and nongang peers. This sug-

gests that the black box of processes associated with gang facilitation effects on the street might also be at work online (Krohn & Thornberry, 2008; Melde & Esbensen, 2011; Sweeten et al., 2012). Of course, these differences are inde-

pendent from extraneous influences such as online routine activities, which are important factors to consider when modeling online behaviors (e.g. Holt &

Bossler, 2009; Marcum, Higgins, & Ricketts, 2010; Pratt et al., 2010). We report the results of the two-level logistic IRT models of online offend-

ing propensity in Table 3. The findings are from the structural regression model (Equation 2, above), where covariates are related to the latent scores for

online crime and deviance while holding constant item seriousness and research site. In Model 1, we assess whether the bivariate differences in online offending observed in Table 2 persist when accounting for demographic control

variables. The results indicate that this is the case, where gang membership was associated with a 1.46 (p < .001) increase in the log odds of online offend-

ing. Former gang members were also more likely to engage in online offending, with log odds increases of 0.89 (p < .05). Age was the only demographic covari-

ate statistically related to online offending, where a standard deviation (about 9 years) decreased the log odds of online crime and deviance by 0.98 (p < .05).

While criminological “facts” about gender invariance do not appear to receive support online, the negative age effect should prompt further scrutiny about

the generality of the age-crime curve in online settings (e.g. Farrington, 1986; Lauritsen, Heimer, & Lynch, 2009).

Model 2 presents a more stringent test of the relationship between gang

membership and online offending, as it accounts for the online routine activi- ties of respondents. Several important findings are observed. First, all four

online routine activities are at least marginally statistically related to online offending in a positive direction. More technologically competent individuals

6. An offending variety score is reported because it closely approximates IRT offending score (Sweeten, 2012).

CRIMINAL AND ROUTINE ACTIVITIES IN ONLINE SETTINGS 487

who spend more time on the Internet, use social networking sites, and use them at higher frequencies are more likely to engage in crime and deviance

online. Second, the online routine activities covariates wash away online offending differences between former gang members and those who avoid gangs. Any residual ties between former gang members and their gang or other

related life-course factors diminish in influence once accounting for what peo- ple are doing online (Pyrooz, Decker, & Webb, 2010; Sweeten et al., 2012).

Third, even when considering patterns and competency in Internet use, we find that current gang members are engaging in more online criminal and deviant

activity than nongang respondents (b = 1.05, p < .01) and former gang respon- dents (b = .67, p < .05). These findings are consistent with what we know about

gang membership and delinquent behavior in street settings, especially with regard to selection, facilitation, and enhancement theoretical framework of

Thornberry et al. (2003), which provides an additional level of support for this classification scheme in a new (online) setting.

Do gangs use the Internet to further their collective interests?

Our final set of results is presented in Table 4. The univariate statistics provide

a baseline of online gang-related activities for those with a history of gang membership, whether they are current or former gang members. We asked the

gang subset of the sample to report about their gang’s activities online, as well as their own online activities. We supplement the baseline information with

Table 3 Two-level logistic IRT models of propensity for online crime and deviance

Model 1 Model 2

c (SE) c (SE)

Current gang member 1.46 (.35)⁄ 1.05 (.31)⁄

Former gang member 0.89 (.34)⁄ 0.38 (.31) Age (z) �0.98 (.19)⁄ �0.79 (.18)⁄ Male �0.37 (.33) �0.08 (.30) Black �0.51 (.56) �0.39 (.52) Hispanic 0.24 (.53) 0.68 (.50)

Other 0.05 (.71) 0.65 (.65)

Foreign born �0.18 (.49) �0.28 (.45) Internet use frequency (hours) 0.04 (.02)⁄

Social network user 1.25 (.44)⁄

Social network use frequency (hours) 0.07 (.02)⁄

Technological capacity 0.26 (.16)#

Intercept �1.96 (.63)⁄ �4.29 (.77)⁄ Variance (% explained) 2.85 (24%) 1.93 (48%)

⁄p < .05, #p < .10. N = 464 Note. Unconditional variance (holding constant seriousness and site) = 3.73.

488 PYROOZ ET AL.

T a b le

4 G a n g -r e la te d a c ti v it ie s in

o n li n e se tt in g s (N

= 4 1 8 )

U n iv a ri a te

B iv a ri a te

d if fe re n c e s

E v e r g a n g

m e m b e rs

C u rr e n t v s.

fo rm

e r

In te rn e t u se r

v s.

n o n u se r

A g e (z )

Y e a rs

si n c e

le a v in g g a n g

% S ta t.

S ig .

S ta t.

S ig .

S ta t.

S ig .

S ta t.

S ig .

G ro u p -l e v e l o u tc o m e

G a n g h a s a w e b si te

1 9

n s

n s

n s

n s

G a n g o rg a n iz e s o n li n e

1 1

n s

n s

� n s

G a n g re c ru it s o n li n e

8 +

n s

� �

G a n g p o st s v id e o s o n li n e

4 6

n s

+ �

� In d iv id u a l- le v e l o u tc o m e

S e a rc h fo r g a n g in fo rm

a ti o n

2 5

n s

+ �

� W a tc h g a n g -r e la te d v id e o s

5 6

+ +

� �

Im p o rt a n c e o f In te rn e t to

y o u r

g a n g a

S o m e w h a t/ v e ry

im p o rt a n t

2 1

n s

n s

n s

n s

N o t im

p o rt a n t

7 9

Im p o rt a n c e o f In te rn e t to

o th e r

g a n g sa

S o m e w h a t/ v e ry

im p o rt a n t

4 0

� n s

� n s

N o t im

p o rt a n t

6 0

(+ ) = d if fe re n c e is

p o si ti v e a n d st a ti st ic a ll y si g n ifi c a n t a t p < .0 5 , (�

) = d if fe re n c e is

n e g a ti v e a n d st a ti st ic a ll y si g n ifi c a n t a t p < .0 5 , (n s) = n o d if fe re n c e a n d p > .0 5

a O n ly

a su b se t o f g a n g re sp o n d e n ts

(N = 3 0 9 ) w e re

a sk e d th is

q u e st io n .

CRIMINAL AND ROUTINE ACTIVITIES IN ONLINE SETTINGS 489

bivariate differences to determine if the findings are conditioned by (1) current gang involvement, (2) Internet usage, (3) age, and (4) time in years

since leaving the gang. Nearly 20% of this subsample reported that their gang had a website or

social networking page. Among those indicating that their gang had a website, about one-third said that the site was password protected. A female gang member in Los Angeles reported: “you have to be a member to access it.” This

demonstrates a level of sophistication among gangs, with the site functioning as online “set space” or a virtual street corner (Papachristos, 2005; Tita,

Cohen, & Engberg, 2005). Only 11% reported that their gang organized activi- ties online. When asked this question, a standard response was “hell no.”

There were, however, some indicators of online organization. A gang member is St. Louis mentioned writing in code—“we got a baseball game”—on Face-

book to urge the gang to get together. A gang member in Fresno told us that they avoid drug involvement online, but used the Internet to organize meet-

ings, parties, and even fundraisers for “bail or other emergencies.” Older members were less likely to endorse this question, which is typical of most online gang behaviors, including gang recruitment.

Only eight percent of the sample reported that their gang recruited new members online, with current gang members more likely to answer “yes,” and

older respondents and those further removed from the gang more likely to answer “No.” As Densley (2012) has noted, trust is a major component of

street gang recruitment, something the anonymity of the Internet is unlikely to project. Finally, nearly 50% of the gang subset said that their gang posts videos

online. Internet users were more likely to indicate this, while older respon- dents and those more removed from the gang were more likely to indicate otherwise. A gang member in St. Louis told us: “Someone’s always got a phone

recording. Anything you record goes on Facebook or YouTube.” Querying “gang” and “gang fight” on YouTube confirms such a perspective. The sample

had strong and varied opinions about posting videos. Many reported that they would post fights, but remove them after a couple of days to prevent law

enforcement from securing the incriminating videos. For the individual-level outcomes, we find that one-quarter of the respon-

dents search for gang-related information online. A common response, as a former gang member in Fresno noted, was to “Google rival gang names, see

what shows up.”7 The majority (56%) of respondents reported that they watched gang-related videos online, which consisted mostly of fights and music. Many respondents were simply interested in gang-related fights and

assaults in general, finding them as entertaining as a boxing or UFC match. Finally, we examined the importance that gang subset placed on the Internet.

For their own gang, about 20% of respondents perceived the Internet as some-

7. Several respondents in Fresno mentioned the controversial video “Fresno Uncensored,” a DVD made by a local rapper featuring several Fresno area gangs that is now freely available on video sharing sites like YouTube.

490 PYROOZ ET AL.

what or very important, with no differences across the discriminating factors. Their perception of other gangs was much different, however. Indeed, around

40% of the respondents perceived the Internet as somewhat or very important. The difference between “own” and “other” gang is illuminating in light of

what Howell (2007) refers to as street gang “myths.” How gangs perceive other gangs has important implications for emerging trends and isomorphic pro- cesses, and can shed light on what Felson (2006) calls “big gang theory” (see

also Ayling, 2011; Densley, 2012; Klein, 1971). We draw several conclusions from these data. First, there is limited

evidence to support the claims that gangs utilize the Internet to further their collective interests. The overwhelming majority of respondents do not report

that their gang has a website to promote their activities, organizes meetings or related activities online, or uses the Internet to replenish their ranks. As a

result of this finding, second, rather than facilitating instrumental goals, it appears that the Internet satisfies the symbolic needs of gangs. The Internet is

less important to achieving instrumental outcomes such as drug sales or recruit- ment, and more important to the accomplishment of status goals. In this way, gangs use the Internet much like an electronic graffiti wall. Third, even if gangs

had the organizational capacity to exploit the Internet for instrumental gains, much like on the street, gangs know that the police monitor their activities. We

revisit these themes in the concluding section of this paper.

Discussion

Changes in technology—cars, telephones, and computers—all changed the nature of social life as well as the motives and methods for accomplishing

crime. Similarly, new crime types emerge over time. Prior to September 11, 2001 mainstream criminology paid little attention to terrorism, but theory and

method has since caught up (Agnew, 2010; Lafree et al., 2009; Rosenfeld, 2004). The narrowing digital divide and the advent of Web 2.0 has placed the

Internet and its capacities at the hands of offenders and nonoffenders alike. It is important to understand how offenders’ use of this powerful technology par-

allels or diverges from their street offending activities. Drawing from online perspectives of identity and behavior—web-facilitated or web-enhanced—our

findings strongly suggest that there is a dual parallelism, first between the non- criminal use of the Internet by active and nongang respondents. That is, with regard to the use of social media, online shopping and the like, gang members

look much like their nongang counterparts in terms of access, prevalence of use, and primary activities. There is a second source of parallelism our findings

support, between criminal activities that occur on the street and those that occur online. The online behavior of gangs and gang members is particularly

important to understand in this context because the Internet may serve as a new venue for crime and deviance. Based on the results presented above,

three larger points merit further discussion.

CRIMINAL AND ROUTINE ACTIVITIES IN ONLINE SETTINGS 491

First, much of the online behavior that gang members engage in is age-appropriate. The online findings resonate with a substantial body of

research on the age-appropriateness of gang member behavior in street or offline settings, particularly with regard to noncriminal activities (Decker &

Van Winkle, 1996; Klein, 1971; Miller, 2011). Our results extend this line of research to include a number of routine online activities. Gang members are online, using the Internet and social networking sites as much, if not more,

than their nongang counterparts. We also observe that neither gang members nor their nongang peers have the technological competency to engage in com-

plex forms of cybercrime, like phishing schemes, identity theft, or hacking into commercial enterprises. In short, while the Internet has reached these inner-

city populations, access alone is not translating into sophisticated technologi- cal know-how. Rather, these findings reflect the offline identities typical of

many American youth and young adults (Tyler, 2002; van Dijk, 2006). Interests in online shopping, video watching, and participation on social networking sites

are commonplace, and indicate the integration of the Internet into everyday life. These same offline identities prove problematic once we consider delinquent or antisocial behaviors.

Second, the overall propensity to participate in online forms of crime and deviance is greater among gang members than nongang youth and young

adults. When gang members are surveyed about their behavior in offline set- tings, studies consistently demonstrate between (using nongang youth as con-

trols) and within (using the gang member as a control) individual influences of gang involvement on criminal offending (Decker, Melde, & Pyrooz, 2012; Krohn

& Thornberry, 2008). The fact that the patterns of criminal propensity in rela- tion to gang membership status hold when applied to online settings provides initial support for an extension of the selection, facilitation, and enhancement

models of gang membership to the virtual world. The assumption that offline identity and behavior—web-enhancement—migrates into online settings makes

available myriad criminological theories to the empirical study of online crime and deviance.

The weakly regulated confines of the Internet are unlikely to inhibit the group processes within gangs that promote online criminal and deviant activi-

ties. In fact, the Internet is likely to encourage various activities—e.g. drug sales, harassment, threats, and fights—because of virtual convergence spaces

in the overlapping social worlds of offenders and targets. Indeed, the Internet can initiate new gang conflicts or exacerbate existing ones. In response to a question about gangs and criminal activity on the Internet, a gang member in

St. Louis recounted such an online conflict: “Motherfucker got shot up, sliced up, killed, and thrown in a trash can. Facebook is some shit, man. You forget

about real life outside.” While the Internet can complement the street as an extension of gang life, it can also supplement gang identity. Gang members

have a greater ability to meet the normative expectations of membership, enhancing their status by posting intimidating videos on YouTube or threaten-

ing messages on Facebook. What Katz (1988) referred to as “dread” has now

492 PYROOZ ET AL.

gone digital. The Internet is a space where gang members may extend their criminal involvement; whether that activity revolves around personal interests

and status enhancement as opposed to furthering the interests of the gang is a different question.

Third, the online activities of gangs reflect symbolic rather than instrumen- tal objectives. We find little evidence to support views that the Internet satis- fies the instrumental needs of gangs (e.g. organization and recruiting). One of

the thorny issues in the study of gangs and other groups involved in crime is when a member is acting in concert with other members to further group goals

and when their behavior is independent of that influence (Short, 1985). The street-level behavior of gangs has been the subject of such explanations

(though hardly in sufficient detail or volume) as have other groups such as ter- rorists and other extremist groups. The Internet poses a challenge to doing so,

as it typically engages individuals not in the aggregate but rather as solitary users. Despite this, the Internet does allow a new form of collective action

among groups, and may help stratify group norms. Our findings demonstrate that collective and group identity exist on the web through postings on social media and Facebook that prompt violence in ways that mirror street behavior.

Thus, gangs exploit the Internet to further their collective identity rather than for instrumental means. Much like offline realities, the Internet is an extension

of gang behavior on the street, but one with a much wider audience. Perhaps once gangs acquire the technological knowhow to exploit the instrumental

opportunities available online, with the ability to diffuse responsibility to limit law enforcement surveillance and intervention, we will see a more advanced

manipulation of the Internet among gangs. Until then, online activities should reflect offline realities, and policy should develop accordingly.

These findings lend themselves to a number of policies and practices target-

ing various aspects of online deviance. Though our findings suggest difficulties in responding to crimes facilitated by the Internet, technology may just as

likely provide the solution. Emerging evidence suggests suppression and social support responses on the part of Internet Service Providers (ISPs), websites,

and law enforcement can be used to affect the behavior of offenders as well as potential victims and offenders. Through a subsidiary (Google Ideas), Google

has launched a social media program designed to address the use of the Inter- net to attract and recruit members of extremist groups.8 Police in cities such

as Philadelphia are using Twitter to notify residents of shootings and crimes in their neighborhoods; the use of the Internet as a means of reporting crimes, as well as community outreach efforts, will become increasing valuable as more

citizens move online. The National Institute of Justice has announced a funding solicitation for the study use of social media as a response to crime, suggesting

the belief in a certain level of sophistication in such practices. We urge law

8. Strategies Against Violence Extremism (againstviolentextremism.org) is the product of a 2011 conference that produced strategies to employ Twitter to use ex-members of these groups to provide content and steer potential recruits away from chat rooms and extremist messages.

CRIMINAL AND ROUTINE ACTIVITIES IN ONLINE SETTINGS 493

enforcement to work more closely with different websites and ISPs as neces- sary to address gangs and crime online. For example, YouTube has previously

assisted law enforcement in identifying gang members and other offenders who post videos online (Hanser, 2011), and our findings suggest a number of

other websites that should be monitored for gang activity. Lastly, our results suggest that gang members are moving to multiple forms of communication. While we have focused primarily on the Internet, many of these same recom-

mendations apply to other forms of telecommunication (i.e. cell phones and email), areas that should not be overlooked by law enforcement or criminolo-

gists (Decker et al., 2012). Of course, this study did not directly observe gang-related behaviors and

processes online; our goal was to examine similarities and differences in online behavior across gang and nongang youth and young adults. While the

cross-sectional nature of our data is a limitation of the study with regard to online criminal propensity, the robustness these findings is enhanced (1) when

we compare gang effects by 6 month and lifetime exposure times for online offending, where no gang to nongang differences are observed for the latter, and (2) by the purposive sampling strategy of high-risk youth and young adults

from similar ecological settings, which goes a long way to address unobserved population heterogeneity that could render the findings spurious. We see lon-

gitudinal research in online settings as a productive route for future study to understand the evolution of online identity and behavior in relation to conti-

nuity and change in gang membership. This will allow researchers to under- stand both the benefits and consequences of gang membership in a setting

normally viewed as conferring benefits (Moule et al., 2013). Further, we encourage multi-measure approaches of the explanatory and outcome vari- ables and sampling strategies that include a broader range of youth and young

adults. The use of the Internet for criminal and deviant purposes is only likely to

grow as Internet access and use continues to expand. Barriers to Internet access and use have been lowered by the diffusion of technology while nor-

mative compliance has propelled newer (and older) generations online. The study of crime and deviance online is not esoteric; criminologists should

embrace the challenge that extending theory and method to the Internet presents. A first step would be to test the extent to which criminological

“facts”—the age-crime curve, gender invariance, victim-offender overlap, and peer effects—extend into online settings; these facts are at the heart of important theoretical and methodological advances in the field. The current

research has taken such a step by extending the study of offending, gang membership, and gangs to the Internet, providing evidence that gang

membership is related to offending in online settings. If online behavior can- not be reduced to offline behavior, it appears that the Internet may present

criminological challenges that will be a source of theory and method for years to come.

494 PYROOZ ET AL.

Acknowledgements

Funding from Google Ideas supported this project. We are grateful for their support. The content of this paper, however, is solely the responsibility of the

authors and does not necessarily represent the official views of Google. We would like to thank Travis Pratt for his comments on this manuscript. Portions of this manuscript were presented at the annual meeting of the Academy of

Criminal Justice Sciences in New York. Correspondence concerning this article should be addressed to David C. Pyrooz, College of Criminal Justice, Sam

Houston State University.

References

Agnew, R. (2010). A general strain theory of terrorism. Theoretical Criminology, 14, 131-153.

Ayling, J. (2011). Gang change and evolutionary theory. Crime, Law, and Social Change, 56, 1-26.

Bjorgo, T., & Horgan, J. (2009). Introduction. In T. Bjorgo & J. Horgan (Eds.), Leaving terrorism behind: Individual and collective disengagement (pp. 1-13). New York, NY: Routledge.

Blevins, K. R., & Holt, T. J. (2009). Examining the virtual subculture of johns. Journal of Contemporary Ethnography, 38, 619-648.

Burris, V., Smith, E., & Strahm, A. (2000). White supremacist networks on the Internet. Sociological Focus, 33, 215-234.

Corb, A. (2011). Into the minds of mayhem: White supremacy, recruitment, and the Internet. Cambridge: Google Ideas.

Cormode, G., & Krishnamurthy, B. (2008). Key differences between Web 1.0 and Web 2.0. Firzst Monday, 13.

Curry, G. D. (2010). Gangs, crime and terrorism. In B. Forst, J. R. Greene, & J. P. Lynch (Eds.), Criminologists on Terrorism and Homeland Security (pp. 97-112). New York, NY: Cambridge.

Curry, G. D., Decker, S. H., & Pyrooz, D. C. (2013). Confronting gangs: Crime and com- munity (3rd ed.). New York, NY: Oxford.

Daniels, J. (2009). Cyber racism: White supremacy online and the new attack on civil rights. Lanham, MD: Rowman & Littlefield.

Dawson, L. L. (2009). The study of new religious movements and the process of radical- ization in terrorist groups. Ottawa: Defense Research and Development Canada, Centre for Security Science.

Dawson, L. L., & Hennebry, J. (1999). New religions and the Internet: Recruiting in a new public space. Journal of Contemporary Religion, 14, 17-39.

Decary-Hetu, D., & Morselli, C. (2011). Gang presence in social network sites. Interna- tional Journal of Cyber Criminology, 5, 876-890.

Decker, S. H. (1996). Collective and normative features of gang violence. Justice Quar- terly, 13, 243-264.

Decker, S. H., & Pyrooz, D. C. (2011). Gangs, terrorism, and radicalization. Journal of Strategic Security, 4, 151-166.

Decker, S. H., Melde, C., & Pyrooz, D. C. (2012). What do we know about gangs and gang members and where to we go from here? Justice Quarterly. Online first. DOI:10.1080/07418825.2012.732101.

CRIMINAL AND ROUTINE ACTIVITIES IN ONLINE SETTINGS 495

Decker, S. H., & Pyrooz, D. C. (2013). Gangs: Another form of organized crime?. In L. Paoli (Ed.), Oxford handbook of organized crime. New York, NY: Oxford University Press.

Decker, S. H., van Gemert, F., & Pyrooz, D. C. (2009). Gangs, migration, and crime: The changing landscape in Europe and the United States. Journal of International Migration and Integration, 10, 393-408.

Decker, S. H., & Van Winkle, B. (1996). Life in the gang: Family, friends, and violence. New York, NY: Cambridge University Press.

Densley, J. A. (2012). Street gang recruitment. Social Problems, 59, 301-321. DiMaggio, P., Hargittai, E., Celeste, C., & Shafer, S. (2004). Digital inequality: From

unequal access to differentiated use. In K. Neckerman (Ed.), Social inequality (pp. 355-400). New York, NY: Russell Sage Foundation.

Durkin, K. F., Forsyth, C. J., & Quinn, J. F. (2006). Pathological Internet communities: A new research direction for sexual deviance research in a post modern era. Socio- logical Spectrum, 26, 595-606.

Esbensen, F.-A., Winfree, L. T., Jr., He, N., & Taylor, T. J. (2001). Youth gangs and definitional issues: When is a gang a gang, and why does it matter? Crime & Delin- quency, 47, 105-130.

Farrington, D. P. (1986). Age and crime. Crime and Justice, 7, 189-250. Felson, M. (2006). Crime and nature. Thousand Oaks, CA: Sage. Fox, K. A., Nobles, M. R., & Akers, R. L. (2010). Is stalking a learned phenomenon? An

empirical test of social learning theory. Journal of Criminal Justice, 39, 39-47. Gerstenfeld, P., Grant, D. R., & Chiang, C.-P. (2003). Hate online: A content analysis of

extremist Internet sites. Analyses of Social Issues and Public Policy, 3, 29-44. Gogolak, E. (2012). Inside the weird world of tracking gangs on social media. The Atlan-

tic Cities. Retrieved from http://www.theatlanticcities.com/neighborhoods/2012/ 07/inside-weird-world-tracking-gangs-social-media/2734/#.UBMZTf8X1F8.email

Gottfredson, M. R., & Hirschi, T. (1990). A general theory of crime. Stanford, CA: Stan- ford University Press.

Hanser, R. D. (2011). Gang-related cyber and computer crimes: Legal aspects and prac- tical points of consideration in investigations. International Review of Law, Comput- ers, and Technology, 25, 47-55.

Holt, T. J. (2007). Subcultural evolution? Examining the influence of on- and off-line experiences on deviant subcultures. Deviant Behavior, 28, 171-198.

Holt, T. J. (2012). Exploring the intersections of technology, crime, and terror. Terror- ism and Political Violence, 24, 337-354.

Holt, T. J., Blevins, K. R., & Burkert, N. (2010). Considering the pedophile subculture online. Sexual Abuse, 22, 3-24.

Holt, T. J., & Bossler, A. M. (2009). Examining the applicability of lifestyle-routine activities theory for cybercrime victimization. Deviant Behavior, 30, 1-25.

Holtfreter, K., Reisig, M. D., & Pratt, T. C. (2008). Low self-control, routine activities, and fraud victimization. Criminology, 46, 189-220.

Howell, J. C. (2007). Menacing or mimicking? Realities of youth gangs. Juvenile and Family Court Journal, 58, 39-50.

Hughes, L. A., & Short, J. F., Jr. (2005). Disputes involving youth street gang members: Micro-social contexts. Criminology, 43, 43-76.

Internet World Stats. (2012). World Internet users and population stats. Retrieved from http://www.internetworldstats.com/stats.htm

Jackson, B. A. (2001). Technology acquisition by terrorist groups: Threat assessment informed by lessons from private sector technology adoption. Studies in Conflict and Terrorism, 24, 183-213.

Katz, J. (1988). The seductions of crime. New York, NY: Basic Books.

496 PYROOZ ET AL.

Katz, C. M., Webb, V. J., & Decker, S. H. (2005). Using the Arrestee Drug Abuse Moni- toring (ADAM) program to further understand the relationship between drug use and gang membership. Justice Quarterly, 22, 58-88.

King, J. E., Walpole, C. E., & Lamon, K. (2007). Surf and turf wars online—growing implications of internet gang violence. Journal of Adolescent Health, 41, 66-68.

Klein, M. W. (1971). Street gangs and street workers. Englewood Cliffs, NJ: Prentice Hall.

Klein, M. W., & Maxson, C. L. (2006). Street gang patterns and policies. New York, NY: Oxford University Press.

Knox, G. (2011). Gang members on Facebook: Should we look the other way? Retrieved from http://www.ngcrc.com/gangface.html

Krohn, M. D., & Thornberry, T. P. (2008). Longitudinal perspectives on adolescent street gangs. In A. M. Liberman (Ed.), The long view of crime: A synthesis of longitudinal research (pp. 128-160). Washington, DC: National Institute of Justice.

Lafree, G., Dugan, L., & Korte, R. (2009). The impact of British counterterrorist strate- gies on political violence in Northern Ireland: Comparing deterrence and backlash models. Criminology, 39, 17-45.

Lauritsen, J. L., Heimer, K., & Lynch, J. P. (2009). Trends in the gender gap in violent offending: New evidence from the National Crime Victimization Survey. Criminology, 47, 361-399.

Maratea, R. J., & Kavanaugh, P. K. (2012). Deviant identity in online contexts: New directives in the study of a classic concept. Sociological Compass, 6, 102-112.

Marcum, C. D., Higgins, G. E., & Ricketts, M. L. (2010). Potential factors of online vic- timization of youth: An examination of adolescent online behaviors utilizing routine activity theory. Deviant Behavior, 31, 381-410.

May, D. C., Bossler, A. M., & Holt, T. J. (2011). Low self-control, deviant peer associa- tions, and juvenile cyberdeviance. American Journal of Criminal Justice, 37, 378-395.

McKenna, K. Y. A., & Bargh, J. A. (1999). Causes and consequences of social interaction on the Internet: A conceptual framework. Media Psychology, 1, 249-269.

McKenna, K. Y. A., & Bargh, J. A. (2000). Plan 9 from cyberspace: The implications of the Internet for personality and social psychology. Personality and Psychology Review, 4, 57-75.

Mehra, B., Merkel, C., & Bishop, A. P. (2004). The Internet for empowerment of minor- ity and marginalized users. New Media and Society, 6, 781-802.

Melde, C., & Esbensen, F. -A. (2011). Gang membership as a turning point in the life course. Criminology, 49, 513-552.

Melde, C., & Esbensen, F.-A. (2012). Gangs and violence: Disentangling the impact of 30 gang membership on the level and nature of offending. Journal of Quantitative Criminology. Online first. doi: 10.1007/s10940-012-9164-z

Miller, W. B. (2011). City Gangs. Retrieved from http://gangresearch.asu.edu/ wal- ter_miller_library/walter-b.-miller-book/city-gangs-book

Morselli, C., & Décary-Hétu, D. (2012). Crime facilitation purposes of social networking sites: A review and analysis of the “cyberbanging” phenomenon. Small Wars and Insurgencies, 23, Volume 24, PP: 152-170.

Mossberger, K., Kaplan, D., & Gilbert, M. A. (2008). Going online without easy access: A tale of three cities. Journal of Urban Affairs, 30, 469-488.

Moule, R. K.Jr, Decker, S. H., & Pyrooz, D. C. (2013). Social capital, the life-course, and gangs. In Gibson & M. Krohn (Eds.), Handbook of life-course criminology (pp. 143-158). New York, NY: Springer.

Newman, G. R., & Clarke, R. V. (2003). Superhighway robbery: Preventing e-commerce crime. Devon, UK: Willan.

CRIMINAL AND ROUTINE ACTIVITIES IN ONLINE SETTINGS 497

NGIC. (2011). 2011 National gang threat assessment—Emerging trends. National Gang Intelligence Center. Retrieved from http://www.fbi.gov/stats-services/publications/ 2011-national-gang-threat-assessment

NTIA. (2011). Digital nation: Expanding Internet usage. National Telecommunications and Information Administration. Retrieved from http://www.ntia.doc.gov/files/ntia/ publications/ntia_internet_use_report_february_2011.pdf

O’Deane, Matthew. (2011). Combating gangsters online. FBI Law Enforcement Bulletin, 80, 1-9.

O’Reilly, T. (2007). What is web 2.0: Design patterns and business models for the next generation of software. Communications and Strategy, 65, 17-37.

Osgood, D. W., McMorris, B. J., & Potenza, M. T. (2002). Analyzing multiple-item mea- sures of crime and deviance I: Item response theory scaling. Journal of Quantitative Criminology, 18, 267-296.

Papachristos, A. V. (2005). Gang world. Foreign Policy, March/April, 48-55. Pasek, J., More, E., & Romer, D. (2008). Realizing the social Internet? Online social net-

working meets offline civic engagement. Journal of Information Technology and Poli- tics, 6, 197-215.

Pew. (2012). Demographics of Internet users. Retrieved from http://pewinternet.org/ Trend-Data-%28Adults%29/Whos-Online.aspx

Pratt, T. C., Holtfreter, K., & Reisig, M. D. (2010). Routine online activity and internet fraud targeting: Extending the generality of routine activity theory. Journal of Research in Crime and Delinquency, 47, 267-296.

Pyrooz, D. C., & Decker, S. H. (2012). Delinquent behavior, gangs and violence in China. Journal of Quantitative Criminology, online first. doi: 10.1007/s10940-012-9178-6

Pyrooz, D. C., Decker, S. H., & Webb, V. J. (2010). The ties that bind: Desistance from gangs. Crime and Delinquency. Online First. doi: 10.1177/0011128710372191

Quinn, J. F., & Forsyth, C. J. (2005). Describing sexual behavior in the era of the Inter- net: A typology for empirical research. Deviant Behavior, 26, 191-207.

Raudenbush, S. W., Johnson, C., & Sampson, R. J. (2003). A multivariate, multilevel Rasch model with application to self-reported criminal behavior. Sociological Meth- odology, 33, 169-211.

Redmon, D. (2003). Playful deviance as an urban leisure activity: Secret selves, self-val- idation, and entering performances. Deviant Behavior, 24, 27-51.

Reid, E., & Chen, H. (2002). Internet-savvy US and Middle Eastern extremist groups. Mobilization: An International Quarterly Review, 12, 177-192.

Rogers, E. (2005). The diffusion of innovation (5th ed.). New York, NY: Simon and Schuster.

Rosenfeld, R. (2004). Terrorism and criminology. In M. Deflem (Ed.), Terrorism and counter-terrorism: Criminological perspectives (pp. 19-32). London: Elsevier.

Schafer, J. (2002). Spinning the web of hate: Web-based hate propaganda by extremist organizations. Journal of Criminal Justice and Popular Culture, 9, 69-88.

Sela-Shayovitz, R. (2012a). Gangs and the web: Gang members’ online behavior. Journal of Contemporary Criminal Justice. Online first. doi:10.1177/1043986212458191

Sela-Shayovitz, R. (2012b). The impact of globalization, migration, and social group pro- cesses on neo-nazi youth gangs. In F.-A. Esbensen & C. L. Maxson (Eds.), Youth gangs in international perspective: Results from the Eurogang program of research (pp. 211-224). New York, NY: Springer.

Selepak, A. (2010). Skinhead super mario brothers: An examination of racist and violent games on White supremacist web sites. Journal of Criminal Justice and Popular Cul- ture, 17, 1-47.

Short, J. F.Jr. (1985). The level of explanation problem. In R. Meier (Ed.), Theoretical methods in criminology (pp. 51-74). Beverly Hills, CA: Sage.

498 PYROOZ ET AL.

Short, J. F., & Strodtbeck, F. L. (1965). Group process and gang delinquency. Chicago: University of Chicago Press.

Skinner, W. F., & Fream, A. M. (1997). A social learning theory analysis of computer crime among college students. Journal of Research in Crime and Delinquency, 34, 495-518.

Smith, M. A., & Kollock, P. (1999). Communities in cyberspace. London: Routledge. Spaaij, R. (2010). The enigma of lone wolf terrorism: An assessment. Studies in Conflict

and Terrorism, 33, 854-870. Sweeten, G. (2012). Scaling criminal offending. Journal of Quantitative Criminology,

28, 533-557. Sweeten, G., Pyrooz, D. C., & Piquero, A. R. (2012). Disengaging from gangs and desist-

ing from crime. Justice Quarterly. Online first. doi: 10.1080/07418825.2012.723033 Thornberry, T. P., Krohn, M. D., Lizotte, A. J., Smith, C. A., & Tobin, K. (2003). Gangs

and delinquency in developmental perspective. Cambridge: Cambridge University Press.

Tita, G. E., Cohen, J., & Engberg, J. (2005). An ecological study of the location of gang ‘set space’. Social Problems, 52, 272-299.

Tyler, T. R. (2002). Is the Internet changing social life? It seems the more things change, the more they stay the same. Journal of Social Issues, 58, 195-205.

van Dijk, J. A. G. M. (2006). Digital divide research, achievements, and shortcomings. Poetics, 34, 221-235.

Waskul, D. D. (2003). Self-games and body-play: Personhood in online chat and cyber- sex. New York, NY: Peter Lang.

Waskul, D. D., & Martin, J. A. (2010). Now the orgy is over. Symbolic Interaction, 33, 297-318.

Weimann, G. (2004). www.terror.net: How modern terrorism uses the Internet. Wash- ington, DC: United States Institute of Peace.

Weimann, G. (2006). Virtual disputes: The use of the Internet for terrorist debates. Studies in Conflict & Terrorism, 29, 623-639.

Wellman, B., Haase, A. Q., Witte, J., & Hampton, K. (2001). Does the Internet increase, decrease, or supplement social capital? Social networks, participation, and commu- nity commitment. American Behavioral Scientist, 45, 436-455.

Wolfe, S. E., Higgins, G. E., & Marcum, C. A. (2007). Deterrence and digital piracy: A preliminary examination of the role of viruses. Social Science Computer Review, 26, 317-333.

Womer, S., & Bunker, R. J. (2010). Surenos gangs and Mexican cartel use of social net- working sites. Small Wars and Insurgencies, 21, 81-94.

Zhou, Y., Reid, E., Qin, J., Chen, H., & Lai, G. (2005). US domestic extremist groups on the web: Link and content analysis. IEEE Intelligent Systems, 20, 44-51.

CRIMINAL AND ROUTINE ACTIVITIES IN ONLINE SETTINGS 499