Street Youth,Social Schemas,and Crime

Abstract

Utilizing 400 youths living on the street, the article examines the social schematic theory of crime developed by Simons and Burt. It explores the role homelessness, physical abuse, emotional neglect, violent victimization, and peers play in the development of criminogenic knowledge structures (CKSs). It then examines the associations between adverse experiences, the CKS, and crime. Results show that deviant peers mediate the relationships between physical abuse, homelessness, violent victimization, and the CKS, while emotional neglect is directly associated with the CKS. The CKS in turn is directly linked to crime along with peers, homelessness, and violent victimization. The CKS also mediates the relationships between deviant peers and offending, and emotional neglect and offending. The relationships between physical abuse, homelessness, violent victimization, and crime are mediated by deviant peers. Furthermore, these relationships are also serially mediated through deviant peers and the CKS. Avenues for future research and policy implications are discussed.

Keywords

social schematic theory street youth homelessness crime abuse victimization

Introduction

In their social schematic theory of crime (SSTC), Simons and Burt (2011; Simons, Burt, Barr, Lei, & Stewart, 2014) integrate important notions and research evidence from social-learning (Akers, 1998), strain (Agnew, 2006), culture (Anderson, 1999), self-control (Gottfredson & Hirschi, 1990), and social control perspectives (Hirschi, 1969) to provide a more comprehensive explanation for crime. Drawing from the literature that shows that adversity influences crime, Simons and Burt (2011) propose that chronic or repetitive adverse experiences provide “lessons” about relationships between people, social norms, and future expectations (p. 585). Rather than focusing on the form of learning as in other perspectives (Akers, 1998), Simons and Burt stress the content of what is learned. Poor parenting, community poverty, victimization, and racial discrimination offer messages about the predictability and supportiveness of the world. These disparate social factors communicate to individuals that their environment is a dangerous and hostile place where people are untrustworthy, rewards unpredictable, and social rules differentially applicable. These experiences become internalized as social schemas that focus on hostile views of relationships, cynical orientations toward conventional norms, and immediate gratification. As the negative situations provide similar lessons, the schemas emerge interrelated and operate together to form a criminogenic knowledge structure (CKS). This CKS encourages “criminogenic situational definitions” that induce or justify participation in criminal behavior (Simons & Burt, 2011; Simons et al., 2014).

Existing research (Barr & Simons, 2015; Burt & Simons, 2015; Simons & Barr, 2014; Simons & Burt, 2011; Simons et al., 2014) provides clear support for this integrated theoretical approach showing community levels of crime/victimization, poor parenting, racial discrimination, racial socialization, romantic relationships, and neighborhood levels of collective efficacy influence the development of the CKS. Furthermore, evidence suggests that the CKS is key in understanding crime, mediating the impact of various background factors. The CKS is also important to understanding participation in risky activities and criminogenic activity spaces which increase the likelihood of situational definitions supportive of crime and offending (Simons et al., 2014).

While informative, support for the theory to date relies on research from a single data set and has yet to explore the full range of negative experiences that may contribute to the development of the CKS. In particular, the roles of young people’s experiences with homelessness and key factors associated with this adverse history have yet to be examined in the development of a CKS. Furthermore, no work has investigated how these experiences and the CKS might together explain the criminal behavior of homeless youth.

Homeless street youth are a population immersed in extremely impoverished situations where food, shelter, clothing, and monetary funds are often absent and their provision unpredictable (Baron, 2004; Dachner & Tarasuk, 2002; Ferguson, Bender, & Thompson, 2015a, 2015b; Ferguson, Bender, Thompson, Maccio, & Pollio, 2012; Gaetz & O’Grady, 2002; Hagan & McCarthy, 1997; O’Grady & Gaetz, 2004; Tyler & Johnson, 2004). Street youth tend to be drawn from families characterized by neglect and/or abuse (Baron, 2003a; Bender, Thompson, Ferguson, Yoder, & Kern, 2014; Hagan & McCarthy, 1997; Kort-Butler, Tyler, & Melander, 2011; Thrane, Hoyt, Whitbeck, & Yoder, 2006; Tyler & Johnson, 2006; Whitbeck & Hoyt, 1999). They often experience high levels of violent victimization on the street (Baron, 2003a; Bender et al., 2014; Gaetz, 2004; Hoyt, Ryan, & Cauce, 1999; McCarthy, Hagan, & Martin, 2002; Tyler & Johnson, 2004; Tyler & Melander, 2015; Whitbeck & Hoyt, 1999), and they tend to share relationships with other youth who are involved in deviant activities (Baron, 2009; Ferguson et al., 2015a, 2015b; Ferguson et al., 2012; Hagan & McCarthy, 1997; McCarthy, 1996; McCarthy, Hagan, & Cohen, 1998; McCarthy et al., 2002; Tyler & Johnson, 2004; Whitbeck & Hoyt, 1999). These youth, therefore, have many negative social experiences that might provide the lessons that could be internalized to form a CKS encouraging or justifying crime.

To date, research on this population has tended to focus on how physical abuse, emotional neglect, homelessness, victimization, and deviant peers have direct relationships with crime (e.g., Baron, 2004; Hagan & McCarthy, 1997) with little attention being paid to the potential role of the CKS. Limited research suggests that street youth may have developed some of the schematic aspects of the CKS. Work has established that many street youth are detached from conventional norms (Baron, 2003b, 2004; Hagan & McCarthy, 1997; McCarthy, 1996; McCarthy & Hagan, 2001; McCarthy et al., 1998), but research has yet to explore the role hostile orientations toward the world (although see Baron, 2017, for street code) or immediate gratification might play in the generation of crime (although see Baron, Forde, & Kay, 2007, for alcohol use). Furthermore, no research has centered on the links between individual adverse experiences and the development of hostile orientations, immediate gratification, and a detachment from conventional values. This means that research has not determined whether these schemas exist in this population and are interrelated in a CKS that promotes crime. Finally, no scholarship has investigated whether the potential criminogenic effects of adverse backgrounds identified in prior research remain in the presence of this CKS. Without this information, it is possible that existing research may have misspecified the link between adverse experiences and crime in this population. Consequently, the current work provides an opportunity to explore the applicability of the social schematic theory to a different population and context, enhancing our understanding of a specific group’s criminal behavior, while extending the explanatory breadth of the perspective by examining adverse experiences not previously explored within the framework.

Social Schematic Theory

Central to the social schematic theory is the notion of social schemas. Social schemas can be described as “internalized representations of the patterns inherent in past social interactions” (Simons & Burt, 2011, p. 555; Simons et al., 2014). Schemas develop through the identification of repetitious or consistent aspects of experiences that then enable individuals to quickly identify these characteristics in similar future circumstances. These experiences educate on the importance of certain patterns, the expected behaviors within interactions, and the significance of behavioral options to these circumstances (Crick & Dodge, 1994; Simons & Burt, 2011). As they evolve from continual conditions and repetitive episodes, schemas become durable and generalizable (Simons & Burt, 2011; Simons et al., 2014).

Drawing from past empirical and theoretical work on crime, Simons and Burt (2011) observe that past research shows those who break the law tend to be drawn from settings that provide certain repetitive or continuous negative circumstances. Offenders have often experienced poverty, poor parenting, exposure to crime and victimization, and delinquent peer contact, all of which offer lessons about the world. Despite being seemingly dissimilar, these experiences all afford examples of similar lessons in different contexts about predictability and supportiveness in the world. In particular, these experiences encourage schemas that focus on a hostile view of relationships, immediate gratification, and a cynical view of conventional norms (Simons & Burt, 2011; Simons et al., 2014).

Individuals who develop a hostile view of the world have a contemptuous, cynical view about other people. These individuals are likely to perceive those they interact with as treacherous, selfish predators. This schema leaves people more likely to enter into interpersonal interactions with skepticism, animosity, and antagonism, as well as a heightened sensitivity to slights. These individuals come to believe that coercive tactics are required to deter potential or to penalize past harm, and should be invoked to secure items or identities perceived as deserved. Furthermore, these persons come to lack empathy—reducing reservations against victimizing others.

The immediate gratification schema focuses on the tendency to discount the future (Simons & Burt, 2011; see also Gottfredson & Hirschi, 1990). In particular, individuals who are socially excluded, view the world as unjust, and indicate a fatalistic future are more likely to focus on immediate rewards and behaviors. Their negative experiences demonstrate the unfairness and unpredictability of life and provide knowledge that others are manipulative or exploitative. These personal experiences provide overwhelming evidence that future rewards through delayed gratification are unlikely. Therefore, a focus on immediate gratification is a more practical and sensible response (Simons & Burt, 2011).

The schema centered on cynical views of conventional norms evolves from the range of negative environments individuals encounter that provide exposure to models and rules that promote contemptuous views toward conventional morals. These environments encourage a cynical view of conventional norms by providing additional evidence people are untrustworthy and self-interested and any promises of rewards are unreliable. These experiences demonstrate that society’s rules are not followed by all, suggesting conventional norms are not important and perhaps not in an individual’s best interests to respect. Together, these occurrences leave individuals more likely to become detached from conventional views against lawbreaking (Simons & Burt, 2011).

Simons and Burt (2011; Simons et al., 2014) outline that the three schemas that make up a person’s CKS will be associated, interlinked, and “mutually reinforcing.” Beyond this, the three schemas are also understood to function as a coherent unit.

People are expected to subliminally pool the rubrics of their schemas when reading, interpreting, and reacting to the situations they encounter. Therefore, it is the amalgamation of the three schemas, rather than the predominance of one, that allows understanding of an individual’s participation in crime.

This CKS will lead to criminogenic situational definitions. These definitions can lead to crime by influencing perceptions of threat, as well as shaping perceptions of opportunities for offending. These criminogenic situational definitions develop not only from adverse backgrounds but are also swayed by “contexts for action” or the way locational characteristics prompt behavior. The theory outlines that individuals with strong CKSs will be attracted to criminogenic activity spaces and risky behaviors which increase situational definitions for crime. Furthermore, those with high CKSs, who are involved in risky activities in criminogenic activity spaces, are also more likely to interpret aspects of the risky activities and criminogenic spaces as favoring crime, thereby increasing the likelihood of offending (Simons et al., 2014).

Past Research on Social Schematic Theory

Past work on the social schematic theory does provide support for the perspective. In the first test of the social schematic perspective, Simons and Burt (2011) found support for the basic theoretical relationships. They found the three criminogenic schemas, hostile views, immediate gratification, and cynical views, combined to form the CKS. The adverse background experiences, unsupportive parenting, racial discrimination, community crime, and collective efficacy were all associated with the CKS. Furthermore, discrimination, supportive parenting, and community crime/victimization were related to deviant peers, which in turn was associated with the CKS. The CKS had a strong relationship with crime and fully mediated the impact of all the other variables, but racial discrimination.

Simons et al. (2014) replicated the findings regarding the adverse backgrounds and further revealed that the CKS increased risky activities and exposure to criminogenic activity spaces, which in turn increased criminogenic definitions of situations. These situational variables fully mediated the relationship between the CKS and crime. Burt and Simons (2015) found that racial discrimination promoted crime primarily through the development of the CKS and that ethnic racial socialization and resilience reduced the likelihood of offending through their impact on the CKS. Finally, research has found that high romantic relationship quality is associated with declines in the CKS, thereby increasing desistance (Barr & Simons, 2015; Simons & Barr, 2014).

In sum, the research to date provides support for the perspective. However, all of the research revolves around a single data set and has not explored the full range of possible negative experiences, including homelessness and the adverse experiences associated with homelessness.

Application of the Theory to Street Youth

Research suggests that homeless street youth are exposed to many negative experiences. Past work documents that these youth are often drawn from homes where they experience emotional neglect and/or physical abuse (Baron, 2003a, 2004; Bender et al., 2014; Hagan & McCarthy, 1997; Kort-Butler et al., 2011; Thrane et al., 2006; Tyler & Johnson, 2006; Tyler & Melander, 2015; Whitbeck & Hoyt, 1999). Furthermore, the homelessness experience is characterized by a lack of reliable shelter, food, clothing, and financial support (Baron, 2004; Dachner & Tarasuk, 2002; Ferguson et al., 2015a, 2015b; Ferguson et al., 2012; Gaetz & O’Grady, 2002; Hagan & McCarthy, 1997; O’Grady & Gaetz, 2004; Tyler & Johnson, 2004; Whitbeck & Hoyt, 1999). Street youth also suffer high rates of violent victimization on the street as a result of their exposure to offenders and participation in deviant survival strategies (Baron, 2003a; Bender et al., 2014; Gaetz, 2004; Hoyt et al., 1999; McCarthy et al., 2002; Tyler & Johnson, 2004; Tyler & Melander, 2015; Whitbeck & Hoyt, 1999). Homeless youth often form friendships with criminal others who not only assist in survival and provide guardianship but also exploit and harm them (Baron, 2003a; Hagan & McCarthy, 1997; McCarthy & Hagan, 2001; McCarthy et al., 1998; McCarthy et al., 2002; Tyler, 2008; Tyler & Johnson, 2004; Whitbeck & Hoyt, 1999). Research suggests that homelessness, abuse, neglect, victimization, and peers are related to crime in this population (Baron, 2003a, 2004; Hagan & McCarthy, 1997). It also suggests that youth tend to move from one adverse experience to another. Thus, escaping a negative home environment means moving into another adverse environment—homelessness—that over time exposes youth to other negative experiences (Baron, 2009; Hagan & McCarthy, 1997; Whitbeck & Hoyt, 1999).

It may be that these various adverse experiences teach lessons that lead to the development of a CKS, increasing the likelihood of criminal behavior. For example, absent or inconsistent affection, severe physical punishment, a lack of adequate housing, unpredictable predatory peers, and violent victimization all provide evidence of a dangerous, untrustworthy, unreliable environment that could lead to the internalization of a schema characterized by mistrust and hostility. It is also the case that the lack of shelter, deviant peers, abuse, neglect, and experiences with victimization all demonstrate the unfairness and unpredictability of life. Rare affection, disproportionate punishments, and the inability to confidently determine who they can trust, when they will eat, and where they might sleep can all be seen to lead to a “here and now” orientation. Finally, emotionally neglectful or aggressive parents provide exposure to behaviors that promote a contemptuous view of conventional morals and experiences that lead to a detachment from conventional others. Homelessness and victimization can encourage a cynical view of conventional norms by providing an environment and experiences that demonstrate behaviors that violate broader conventions and by offering evidence that following rules may not always in one’s best interest. Furthermore, deviant peers promote a cynical view of conventional norms by delivering corroborative behavior that reveals the manipulative, exploitive nature of people and exposes contradictions in societal rules.

There is limited evidence of these types of schemas in this population. Although no research has examined hostile views of relationships among homeless youth, some work shows that some homeless youth adopt a street code (Baron, 2017), which contains aspects of a hostile view (see Anderson, 1999). Scholarship has yet to establish, however, if homelessness, victimization, neglect, abuse, or deviant peers are linked to the development of this type of orientation. Research also indicates that some homeless street youth have low self-control—a broad orientation which can include a focus on immediate gratification (Baron, 2003b, 2009, 2015; Kort-Butler et al., 2011). There is little existing investigation focusing on immediate gratification itself (see Baron et al., 2007). Available information suggests that aggregate measures of adverse experiences, including homelessness, abuse, and street victimization (Baron, 2009, 2015), as well as neglect (Baron, 2015), have some association with the development of low self-control. There is a lack of inquiry, however, into the links between individual adverse experiences and immediate gratification specifically. Research has also identified a disengagement from conventional norms (Baron, 2003b, 2004; Hagan & McCarthy, 1997; McCarthy, 1996; McCarthy & Hagan, 2001; McCarthy et al., 1998). Minimal work has explicitly explored links between most of the adverse experiences and disengagement, but available studies demonstrate deviant peers decrease support for conventional norms (McCarthy, 1996).

The sparse available scholarship has identified relationships between the above factors and crime in this population. Research shows that hostile orientations (Baron, 2017), low self-control/immediate gratification (Baron, 2003b, 2009, 2015; Baron et al., 2007; Kort-Butler et al., 2011), and low support for conventional norms (Baron, 2003b, 2004; Hagan & McCarthy, 1997; McCarthy, 1996; McCarthy & Hagan, 2001; McCarthy et al., 1998) have some association with criminal activity. While promising, the overall lack of attention to these factors separately, as well as in tandem, has meant that research has not been undertaken to establish if these schemas come together to form a CKS and if the CKS is related to crime, and if it mediates the impact of the adverse experiences.

The prior research examining the relationships between adverse experiences and crime, and the individual criminogenic schemas and crime has tended to contextualize the work within a range of theoretical perspectives. For example, research exploring the link between adverse experiences and crime has often applied strain and social control theories (see Baron, 2004; Hagan & McCarthy, 1997) as well as coercion perspectives (Baron, 2009, 2015) which have been argued to be specific applications of strain approaches (Baron, 2015). The limited examinations of the individual criminogenic schemas have utilized cultural (Baron, 2017), self-control (e.g., Baron, 2003b; Kort-Butler et al., 2011), coercion (Baron, 2009, 2015), and learning theories (e.g., McCarthy, 1996), and these schemas have generally been examined in isolation of one another. Thus, the types of adverse experiences and potential presence of criminogenic schemas in this population, along with the disparate theoretical explanations and empirical conceptual inquiries, would seem to make the current population a prime candidate for application of the type of theoretical integration provided by the social schematic theory.

Hypotheses

From the literature reviewed above, the following hypotheses can be explored. The social schematic theory outlines that adverse social conditions will lead to the development of a CKS. Past research also outlines a link between the adverse social conditions and the CKS through the association with deviant peers (Simons & Burt, 2011). The adverse conditions increase the likelihood of moving to other negative social environments that increase the likelihood of developing a CKS. Therefore, it is predicted that the adverse social conditions of physical abuse, emotional neglect, homelessness, and violent victimization on the street will have positive direct associations with deviant peers and the CKS. It is also expected that the adverse social conditions will have indirect associations with the CKS through their relationships with deviant peers, which is anticipated to be positively associated with the CKS.

Simons and Burt (2011) make clear the CKS will be associated with crime. Furthermore, they outline that it will mediate the relationship between the adverse social circumstances, deviant peers, and crime. Consequently, it is anticipated that the CKS will be positively associated with crime. Furthermore, it is projected that the relationships between the adverse social circumstances (including physical abuse, emotional neglect, homelessness, and violent victimization), deviant peers, and crime will be mediated by the CKS. In addition, it is expected that the relationships between the adverse social circumstances and crime will be mediated in a serial process through deviant peers and the CKS (see Simons & Burt, 2011).

Method

Participants

The data for this study were collected between June 2009 and August 2010 in a large Canadian city. In total, 400 youths were interviewed in a range of locations, including the streets (and nearby parks, alleys, and businesses), shelters, and drop-in centers. The researcher approached potential respondents and those identified as possible participants were alerted to the project. Youth who communicated interest in contributing to the project were then screened further to determine their eligibility for inclusion. To be considered for participation, youth had to be between the ages of 16 and 24 at the time of the interview, not attending school, unemployed, and reporting no fixed address during the prior 12 months. The minimum age requirement utilized by the research is consistent with the minimum age requirement established by social service agencies. Youth who met the eligibility criteria and indicated interested in participating in the study were then read a letter of information. This document outlined the study more completely, including procedures, the rights of the participants within the interview, and other contact information. Those individuals who agreed to participate were then asked to provide informed consent and were provided with copies of the information and consent form. Participants received $30 (CND) in food coupons for their participation in the interview.

The mean age of the sample used in the study is 20 years. Consistent with the gender composition in previous street youth research, males comprise 64.0% of the youths in sample (Hagan & McCarthy, 1997). Caucasians made up the largest racial/ethnic group (47.5%), 19.5% are African Canadian, 8.0% are Aboriginal, and 17.7% are Asian Canadian, Indo-Canadian, Hispanic, or Biracial/Multiracial. Almost 8.0% of the sample indicated that they were simply Canadian or Other. The average length of time on the streets is 6 months out of the previous 12 months.

Procedure

All participants were interviewed using a structured self-report questionnaire. As the literacy level in the population is limited, the questions were read aloud from the questionnaire to the participants. This ensured that participants would be able to clearly understand what information was being requested as well as providing the opportunity to clarify concerns. The participants’ responses were then recorded onto the questionnaire. All of the participants were asked the same questions in the same order.

Measures

Adverse Experiences

Simons and Burt (2011) argue that the CKS is created by a number of adverse social conditions. In this research, the focus is on homelessness, victimization, physical abuse, and emotional neglect. First, length of no fixed address (NFA), or length of homelessness, was determined by asking respondents how many months out of the prior 12 they had been without shelter or a fixed address. Street victimization was measured by asking participants how many times in the past year they had physical force used against them to get money or things; been attacked by someone with a weapon or fists, injuring them so badly they needed a doctor; been physically attacked for no apparent reason. These three items were summed and then logged to reduce skewness. The physical abuse measure is drawn from the Childhood Trauma Questionnaire (CTQ) and is made up of five items asking respondents if they had experienced the following when growing up: “You got hit so hard by someone in your family that you had to see a doctor or go to the hospital; people in your family hit you so hard that it left you with bruises or marks; you were punished with a belt, a board, a cord, or some other hard object; you got hit or beaten so badly it was noticed by someone like a teacher, neighbor, or doctor; you believe that you were physically abused” (1 = never, 5 = always; α = .90; see Bernstein & Fink, 1998; Scher, Stein, Asmundson, McCreary, & Forde, 2001). Emotional neglect was developed by participants’ responses to five additional questions drawn from the CTQ. The respondents were asked when they were growing up how true was it that: “There was someone in your family who helped you feel important or special; you felt loved; people in your family looked out for each other; people in your family felt close to each other; your family was a source of strength and support” (1 = always, 5 = never; α = .87). While the abusive and neglectful experiences are retrospective reports, research shows that the CTQ generally, and within the homeless population specifically, is a viable instrument for screening for abuse (Forde, Baron, Scher, & Stein, 2012; Scher et al., 2001).

The theory suggests the negative social experiences will leave people more likely to associate with deviant peers. A measure of deviant peers was created by asking respondents, “Of your closest 10 friends how many have shoplifted; stolen a car; broken into a place to steal something; hit someone in a fight; used drugs; sold drugs?” (1 = none, 5 = all; α = .85).

Criminogenic Schemas

Simons and Burt (2011; see also Simons et al., 2014) stress the importance of three different factors in the development of CKSs: hostile view of the environment, immediate gratification (time discounting the future), and low commitment to social conventions. Hostile view of the environment is constructed from seven measures drawn from Simons et al. (2014). Respondents were asked to agree or disagree with the statements: “When people are friendly, they usually want something from you; you have often been lied to; people do not respect a person who is afraid to fight physically for his or her rights; people tend to respect a person who is tough and aggressive; people will take advantage of you if you don’t let them know how tough you are; it is important to show others that you cannot be intimidated; and if you don’t let people know that you will defend yourself, they will think you are weak and take advantage of you” (1 = strongly disagree, 4 = strongly agree; α = .82).

The second component of the CKS, immediate gratification, is drawn from five items from the Grasmick, Tittle, Bursick, and Arneklev (1993) scale. Participants were asked to agree or disagree with the statements: “I often act on impulse without stopping to think; I often do whatever brings me pleasure here and now even at the cost of some distant goal; I’m more concerned with what happens to me in the short run than in the long run; I don’t give much thought to the future (reverse coded); and excitement and adventure are more important than security” (1 = strongly disagree, 4 = strongly agree; α = .60).

The third component of the CKS outlined by Simons and Burt (2011)—low commitment to social conventions—is determined by asking the respondents about four different offenses. Participants were asked, “How morally acceptable is it to steal, use drugs, fight, sell drugs?” (1 = never acceptable, 5 = always acceptable; α = .74).

Simons and Burt (2011; Simons et al., 2014) argue that the three criminogenic schemas come together to form a CKS. To explore for this, the procedure utilized by Simons and Burt was followed where each of the three schemas was standardized and then the three schemas were summed to create a composite measure. An exploratory factor analysis confirmed, similar to past work (see Burt & Simons, 2015; Simons & Barr, 2014; Simons & Burt, 2011; Simons et al., 2014), that the three schemas load onto a single factor with all loadings over .66. Nunnally’s (1978) formula to calculate the reliability of a scale constructed with linear combinations of concepts measuring different aspects within a broader concept, taking into account the reliabilities of the subconcept variables, and the covariances among them, showed a reliability score of .81.

Dependent Variable

Crime was a variety measure. Respondents were asked how many times in the prior 12 months they had broken into a car; broken into a building; sold marijuana or other prescription drugs; used physical force to get money or things from someone; attacked someone with a weapon or fists hurting them badly they probably needed a doctor; got into a fight with someone just for the hell of it; damaged or destroyed, on purpose, property that did not belong to you; taken something worth less than $50 which did not belong to you; taken something worth more than $50 which did not belong to you; broken into a structure to sleep; stolen food because you were hungry; been paid to have sex; taken part in a fight where a group of your friends were against another group; taken a car that did not belong to you or your family without permission of the owner; set fire to someone’s property on purpose? Any item where the respondent indicated participation in the past year was coded 1. Reports of nonparticipation were coded 0. The 15 items were then added together to create the variety crime variable (α = .84).

Analytical Strategy

The analysis begins by examining the relationships between the adverse social circumstances and deviant peers. It then proceeds to explore how adverse social circumstances lead to the development of the CKS, paying attention to the manner in which deviant peers can mediate these influences. The link between adverse social circumstances, criminal peers, the CKS, and crime is then investigated. Here, attention will focus not only on the direct effects of the CKS but also its mediating influence on the relationships between adverse social circumstances and crime and the more complex relationship where the association between adverse social circumstances and crime flows in a serial process through deviant peers and then the CKS.¹ The analysis of the mediation effects exploits a technique developed by Hayes (2013) for ordinary least squares (OLS) regression. The mediation effect of deviant peers on the relationship between negative experiences and the CKS is determined through a procedure where deviant peers are regressed against a particular negative experience (with the other negative experiences and the control variables used as covariates), followed by the CKS regressed against the adverse experiences and deviant peers (along with the covariates), where the deviant peers variable is identified as the mediator. This procedure was replicated with each adverse experience. Estimating each equation within the same execution and designating that confidence intervals (CIs) be calculated from the same random samples allow the procedure to produce results that are mathematically the same as if they had been estimated simultaneously (Hayes, 2013). This technique provides effect sizes and 95% bootstrap CIs for the indirect relationships. The CIs are bias-corrected based on 10,000 bootstrap samples. If the intervals do not contain a zero (if both lower and upper level confidence levels are above 0 or below 0), then the indirect relationship (or mediation) is significant. This approach is preferred over a normal theory approach, where the indirect effect is divided by the estimated standard error to derive a p value, because assumptions of distribution normality are unrealistic with indirect effects (Hayes, 2013). This mediation approach was also utilized examining the indirect effect of deviant peers on crime through the CKS.

The above process was then generalized to explore the process of multiple mediation predicting crime. For each adverse effect, there are three specific indirect effects, as well as the direct effect on crime. First, deviant peers is regressed against an adverse effect (with the other adverse experiences and control variables used as covariates), then the CKS is regressed against the adverse experience and deviant peers (along with the covariates), and finally crime is regressed against the adverse experience, deviant peers, and the CKS. In this process, peers are identified as the first mediator, theoretically preceding the CKS, which is identified as the second mediator. This process provides 95% bias-corrected CIs based on 10,000 samples for (a) the mediating effect of peers on the relationship between the adverse experience on crime, (b) the mediation effect of the CKS on the effect of the adverse experience on crime, and (c) the indirect effect of the adverse experience on crime through deviant peers and the CKS. This procedure was replicated with each adverse experience. Estimating each equation within the same execution and ensuring that the bootstrap samples are drawn from the same 10,000 random samples again allows the Hayes procedure to produce results that are mathematically the same as if they had been estimated simultaneously (Hayes, 2013).

Results

Table 1 provides a summary of the descriptive statistics. It indicates variation in the negative social experiences of the sample with none, beyond homelessness, showing a tendency to be, on average, extremely severe. Seventy-eight respondents reported no incidents of physical abuse but 56 reported very high scores (averaged across the five experiences examined here). Emotional neglect was totally absent for 36 of the participants, while 63 reported consistently severe experiences. About 37% of the sample indicated no experience with violent victimization, but more than half of the sample had experienced multiple victimizations. Most of the youth had associations with deviant peers, with only nine respondents indicating no friendships with someone who had committed at least one of the offenses examined here. However, another 45 indicated only a few of their peers engaged in one these offenses. Most of the respondents had engaged in at least one of the offenses examined here. Only 56 participants (14%) indicated they had not engaged in any offending over the prior year. Offending variety tended to be relatively spread out in the sample with 42, 48, and 41 participants indicating 1, 2, and 3 types of crime participation, respectively. Similarly, 32, 40, and 33 respondents indicated participating in 4, 5, or 6 of the offending types.

Table 1:

Descriptive Statistics

Variable	M	SD	Minimum	Maximum
Age	20.06	2.36	16	24
Gender	0.36	0.48	M = 0	F = 1
Emotional neglect	2.70	1.10	1	5
Physical abuse	2.28	1.23	1	5
Homelessness	6.16	4.00	1	12
Violent victimization	1.04	1.11	0	4
Deviant peers	3.12	0.91	1	5
Hostile view	2.83	0.58	1	4
Immediate gratification	2.64	0.53	1	4
Social convention	2.35	1.01	1	5
Criminogenic knowledge structure	0.01	2.09	−6.59	6.32
Crime	4.51	3.66	0	15

A review of Table 2 shows that many of the theoretically expected relationships are present at the bivariate level. First, the adverse experiences of homelessness, physical abuse, and violent victimization are associated with having deviant friends. Emotional neglect is the only adverse experience that does not have an association with deviant peers. Gender and age were not significantly associated with deviant peers.

Table 2:

Zero-Order Correlation Matrix

Variable	1	2	3	4	5	6	7	8	9
1. Age	—
2. Gender	−.32*	—
3. Emotional neglect	−.17*	.10	—
4. Physical abuse	−.15*	.13*	.43*	—
5. Homelessness	.03	.05	.18*	.07	—
6. Violent victimization	.12*	.16*	.07	.23*	−.02	—
7. Deviant peers	.04	−.07	.05	.16*	.10*	.20*	—
8. Criminogenic knowledge structure	−.07	−.07	.18*	.11*	−.05	.12*	.47*	—
9. Crime	−.11*	−.07	.17*	.16*	.18*	.19*	.42*	.57*	—

p < .05 (two-tailed test).

Three of the adverse experiences also have bivariate relationships with the CKS. Physical abuse, emotional neglect, and violent victimization all have statistically significant relationships with the CKS. Only the length of homelessness shows no significant association here. Deviant peers also shows an association with the CKS. Neither age nor gender show any relationship with the CKS.

The CKS has a moderately strong association with crime. Furthermore, all four adverse experiences, physical abuse, emotional neglect, violent victimization, and homelessness have associations with crime. Finally, deviant peers are related to greater participation in crime. Gender has no relationship with participation in crime but younger respondents are more likely to report participating in a greater range of crimes.

Overall, the patterns are encouraging. The adverse background factors appear to be related to deviant peers. Deviant peers appear to be related to the CKS, and the CKS has an association with crime. The analysis moves now to the multivariate structural model.

The social schematic perspective begins by outlining that adverse social experiences will lead to associations with deviant peers. In turn, deviant peers provide additional lessons enhancing the development of the CKS. An examination of Table 3 shows that three of the four adverse background factors are associated with deviant peers. Those with greater periods of homelessness, more violent victimization, and severe histories of physical abuse are more likely to associate with deviant peers. Emotional neglect, however, does not show an association with deviant peers. Gender is also associated with deviant peers. Males are more likely to associate with deviant peers.

Table 3:

Linear Regression Models of Crime

Variable	Deviant peers		CKS		Crime
Variable	b	SE	b	SE	b	SE
Age	.025	.020	−.084	.042*	−.128	.067
Gender	−.198	.099*	−.410	.208*	−.622	.331
Emotional neglect	−.010	.047	.301	.097*	.106	.156
Physical abuse	.098	.042*	−.067	.088	.119	.140
Homelessness	.025	.011*	−.009	.024	.124	.038*
Violent victimization	.159	.042*	.077	.089	.325	.141*
Deviant peers	—		1.041	.108*	.642	.191*
CKS	—		—		.802	.082*
Adjusted R²	.068		.237		.382

Note. CKS = criminogenic knowledge structure.

p < .05 (two-tailed test).

The theory outlines that adverse social experiences and deviant peers increase the likelihood of developing the CKS. Table 3 indicates that deviant peers are strongly associated with the development of the CKS. Of the adverse social experiences, only emotional neglect has a significant direct relationship with the CKS. Gender and age also have significant direct relationships with the CKS. Males and younger respondents show stronger relationships with this variable.

While three of the four adverse social experiences do not have a direct relationship with the CKS, the theory and past research outline that deviant peers mediate the relationships between the adverse social experiences and the CKS. Table 4 provides information about the indirect relationships the adverse social experiences have with the CKS and the mediating role deviant peers play. Importantly, it reveals several significant indirect relationships. In particular, the indirect relationships physical abuse, homelessness, and violent victimization have with the CKS are significant and mediated by deviant peers. The indirect effect of physical abuse on the CKS through deviant peers is small (b = .1019). The bias-corrected CI for the indirect effect based on 10,000 bootstrap samples, however, did not contain a zero (95% CI = [0.0145, 0.1994]), indicating a significant effect. The indirect effect of homelessness on the CKS through deviant peers is smaller (b = .0262, 95% CI = [0.0029, 0.0533]). The adverse experience with largest indirect effect on the CKS through deviant peers is violent victimization (b = .1653, 95% CI = [0.0749, 0.2710]). While supportive of the presence of indirect effects, the findings suggest that the indirect impact of the adverse experiences on the development of the CKS, through deviant peer association, is substantively modest.

Table 4:

Bias-Corrected CIs for Mediation Effect of Deviant Peers and the CKS on the Development of Crime (10,000 Bootstrap Samples)

Indirect effect	b	95% CI
Indirect effect	b	Lower	Upper
Indirect effects of adverse experiences on CKS through deviant peers
Emotional neglect → Deviant peers → CKS	−.0103	−0.1081	0.0941
Physical abuse → Deviant peers → CKS	.1019*	0.0145	0.1994
Homelessness → Deviant peers → CKS	.0262*	0.0029	0.0533
Violent victimization → Deviant peers → CKS	.1653*	0.0749	0.2710
Indirect effects of adverse experiences on crime through CKS
Emotional neglect → CKS → Crime	.2417*	0.0893	0.4233
Physical abuse → CKS → Crime	−.0536	−0.1969	0.0981
Homelessness → CKS → Crime	−.0075	−0.0444	0.0293
Violent victimization → CKS → Crime	−.0621	−0.0709	0.1928
Deviant peers → CKS → Crime	.8353*	0.6156	1.0922
Indirect effects of adverse experiences on crime through deviant peers
Emotional neglect → Deviant peers → Crime	−.0064	−0.0745	0.0589
Physical abuse → Deviant peers → Crime	.0628*	0.0095	0.1579
Homelessness → Deviant peers → Crime	.0161*	0.0023	0.0399
Violent victimization → Deviant peers → Crime	.1019*	0.0372	0.2095
Indirect effects of adverse experiences on crime through deviant peers and CKS
Emotional neglect → Deviant peers → CKS → Crime	−.0083	0.0891	0.0737
Physical abuse → Deviant peers → CKS → Crime	.0817*	0.0123	0.1667
Homelessness → Deviant peers → CKS → Crime	.0210*	0.0027	0.0440
Violent victimization → Deviant peers → CKS → Crime	.1326*	0.0595	0.2095

Note. CI = confidence interval; CKS = criminogenic knowledge structure.

Significant mediation effect: p < .05.

Moving to an assessment of the relationships between adverse conditions, deviant peers, the CKS, and crime, an inspection of Table 3 shows that the CKS structure has a significant association with crime. Homelessness, violent victimization, and deviant peers are also directly related to crime.

Simons and Burt (2011) outline that the relationship between deviant peers and crime will be mediated by the CKS. Table 4 again provides the information on the mediating relationships. First, the findings show that the indirect relationship between deviant peers and crime mediated by the CKS is quite large (b = .8353). The bias-corrected CI for the indirect effect based on 10,000 bootstrap samples did not contain zero (95 % CI = [0.6156, 1.0922]). This supports the notion that deviant peers are indirectly related to crime through the association with the CKS. Simons and Burt also outline that the CKS will mediate the relationships between the adverse experiences and crime. Table 4 shows little support for this process. There is only one case of direct mediation. There is a significant indirect effect between emotional neglect and crime through the CKS that is moderate in strength (b = .2417, 95% CI = [0.0893, 0.4233]).

Table 4 suggests, however, that deviant peers mediate the relationship between adverse experiences and crime. This table shows that physical abuse has a small indirect relationship with crime through deviant peers (b = .0628, 95% CI = [0.0095, 0.1579]). The relationships between homelessness (b = .0161, 95% CI = [0.0023, 0.0399]) and crime, as well as violent victimization (b = .1019, 95% CI = [0.0372, 0.2095]) and crime were also mediated by deviant peers. Again, these latter two indirect relationships appear to be quite modest.

The theory outlines that the relationship between adverse background factors and crime is complicated and can be understood by a process of serial mediation where the relationship between adverse social circumstances and crime is indirect through deviant peers and then the CKS. Table 4 shows this to be the case for three of the adverse histories. Physical abuse (b = .0817, 95% CI = [0.0123, 0.1667]), homelessness (b = .0210, 95% CI = [0.0027, 0.0440]), and violent victimization (b = .1326; 95% CI = [0.0595, 0.2095]) have significant indirect relationships with crime through deviant peers and the CKS. While significant, these serially mediated relationships were not strong.

Discussion

This article set out to examine the social schematic theory’s applicability in understanding the criminal activity of street youth, paying particular attention to the overlooked adverse experience of homelessness and the negative social environments that tend to be associated with it: emotional neglect, physical abuse, violent victimization, and deviant peer association. The findings provide more support for the theory. The social schematic theory provides a valuable framework for contextualizing how this group of adverse social circumstances can lead to lessons that promote crime. The results here show that homelessness, physical abuse, and violent victimization promote association with deviant peers—a social environment which provides access to additional lessons on trust, reliability, morality, and future rewards, increasing the likelihood of developing a CKS. In turn, the CKS is associated with higher levels of criminal activity.

The findings show that this CKS mediates the relationships between emotional neglect, deviant peers, and crime. Furthermore, the relationship between most of the negative social experiences and crime is indirect through a serial mediation process. Here, adverse experiences (physical abuse, homelessness, victimization) were associated with deviant peers, which contributed to the CKS, which in turn encouraged crime. Furthermore, these adverse experiences had indirect associations with crime through their relationships with deviant peers.

The CKS has the strongest relationship with crime, providing additional support for the importance of this concept in understanding crime. However, homelessness, violent victimization, and deviant peers also had direct relationships with crime. This suggests that their impact on illegal behavior cannot be explained entirely by the lessons delivered by these experiences. Perhaps these relationships fit best with explanations offered in extensions to the theory where homelessness, victimization, and deviant peers offer activity settings, routine activities, and contexts for action that encourage and promote situational definitions favoring crime (Simons et al., 2014).

The results reveal that adverse experiences may not all follow the same path to crime. In particular, emotional neglect did not seem to follow the mold of the other negative social environments. Its relationship with the CKS appears to be more direct instead of mediated by deviant peers like the other adverse experiences. Furthermore, its relationship with crime was through the CKS directly, and not through deviant peers or the more complex deviant peers/CKS serial mediation pathway. This suggests more work is required to explore how different adverse experiences may follow different pathways to the development of the CKS and crime.

The adverse experiences link to crime through deviant peers would also appear to provide some challenge to the theory. While Simons and Burt (2011) acknowledge that this is consistent with a number of theoretical perspectives (e.g., Agnew, 2006; Akers, 1998), the strong direct effect of peers unmediated by the CKS, and this mediated link, suggests that possibly a broader interpretation of the role peers play in crime within the theory is required. Perhaps the results here are simply selection and facilitation effects. Alternatively, these results may indicate additional effects emphasized in other perspectives. Perhaps peers are also providing “tutelage” in the tools and skills of crime that go beyond the content of “lessons” that support using crime, as well as an introduction to criminal networks that enable offenders with CKSs to be more successful in their endeavors (Hagan & McCarthy, 1997; McCarthy, 1996; McCarthy & Hagan, 2001).

While the findings add to the literature examining the social schematic theory and provide a more comprehensive understanding of how street youth become involved in crime, the results should be interpreted with caution. The data are cross-sectional making the ability to determine causality limited. It may be that the CKS and crime increase the likelihood of associating with deviant peers, as well as perhaps fostering some of the adverse experiences, including homelessness and victimization. Deviant peers can also contribute to victimization. Furthermore, participation in crime may serve to strengthen the CKS providing further evidence that supports this perspective on the world. Future work may want to probe the reciprocal fashion in which some of these processes might work.

It is also the case that many of the indirect, or mediating, effects are quite modest. This is applicable to the roles the adverse experiences play in the development of the CKS through deviant peers, as well as the more complex relationship with crime through deviant peers and the CKS. Simons and Burt (2011), however, emphasize that it is the lessons taught in multiple environments that lead to the development of the CKS. It may therefore be important to clarify that while each adverse experience on its own may make only a small contribution to this knowledge structure, the lessons replicated across a number of domains can additively lead to the development of a crime predictor that is more powerful than other traditional measures. Furthermore, the adverse experience that emerged as the strongest in each of the indirect paths was violent victimization, suggesting that this experience in particular is very harmful. Violent victimization destroys trust, breaks bonds with conventional values, creates a focus on immediate rewards, and promotes association with deviant peers and crime. This suggests that while the prior observation that negative experiences in multiple domains are important in the creation of the CKS, some adverse experiences may be more influential than others in the process (Agnew, 2006). Future work may seek to clarify those experiences that have a greater impact on the CKS and crime.

The current work also could not examine the importance of criminogenic situational definitions, nor could it explore the criminogenic settings linked with risky activities and criminogenic activity spaces. While, broadly speaking, homelessness is a risky lifestyle, there may be a variety of activity spaces that homeless youth can access. These settings may be more or less subject to both formal and informal social control, more or less supportive of norms encouraging deviance, and dissimilar in amounts of deviance. For example, some youth may spend more of their time in public locations that are used by populations characterized by high proportions of offenders, public drug use and distribution, and inconsistent police presence. Here, one would expect little social control and a greater proportion of youth committed to perspectives that support a CKS. Other youth may spend less time in these types of locations—instead using social services for homeless youth. Here, youth are not on the street, there is greater informal social control, less support for criminal activity, and less deviant activity present (see Hoyt et al., 1999). Hagan and McCarthy (1997) detail that street youth who spend their time in areas where more social services are available have lower participation in crime, while Hoyt et al. (1999) show that street youth who access social service resources have lower victimization risk.

Some of the measures are also subject to criticism. The items for the immediate gratification schema are drawn from the Grasmick et al. (1993) scale. There is a debate regarding the appropriateness of dividing this scale into its subitems (see Baron et al., 2007; Delisi, Hochstetler, & Murphy, 2003; Forde & Kennedy, 1997; Longshore, Turner, & Stein, 1996; Piquero & Rosay, 1998). Furthermore, the reliability of this particular scale is somewhat low and should be interpreted with caution. Nevertheless, recent research suggests the importance of examining aspects of low self-control separately (Burt, Sweeten, & Simons, 2014). Future research should attempt to examine immediate gratification with a stronger measure perhaps incorporating items from a number of established scales (see Simons & Burt, 2011).

There are a number of important policy implications that emerge from the findings surrounding the need to tackle adverse experiences, deviant peer associations, and criminogenic ways of viewing and responding to the world. Scholars note that typically interventions with this population have adopted harm-reduction approaches (Ferguson, Bender, Thompson, Xie, & Pollio, 2011). This strategy involves providing homeless youth with information and education focusing on risk reduction, addressing personal and employment skill deficits, attempting to link them with prosocial peers, and focusing on decreasing substance use and illegal activities (Ferguson, Bender, & Thompson, 2016). However, Ferguson and Xie (2008) note that three important issues emerge in this population. First, there is a failure to fully utilize the services that are available (Barman-Adhikari & Rice, 2014; De Rosa et al., 1999; Ferguson, 2012; Ferguson & Xie, 2008; Kipke, Unger, O’Connor, Palmer, & LaFrance, 1997). For example, De Rosa et al. (1999) found that while 75% of youth used drop-in facilities, only 40% used shelters and less than 10% of the sample accessed the mental health or substance abuse programs. Second, these services are often offered in isolation from one another potentially decreasing their effectiveness. As the current research has identified, there are multiple factors that must be addressed within the client population and these factors tend to be interrelated and therefore require more holistic tactics (De Rosa et al., 1999; Ferguson, 2012; Kipke et al., 1997). Third, this silo approach to service delivery is seen as potentially contributing to continued participation in illegal subsistence strategies (Ferguson, 2012; Ferguson & Xie, 2008). There is some suggestion that the provision of basic resources without addressing the more complex factors associated with crime allows youth to continue their alternative economic strategies on the street (Ferguson, 2012; Ferguson & Xie, 2008).

It is therefore key to first engage youth in the services that are available. Ferguson and Xie (2008) argue that outreach workers need to target youth who are “service aversive” or who do not have the available knowledge or social capital to access services. These contacts would provide the vital initial linkages to services and resources that are being underutilized. Second, the services that are offered need to be delivered in a more comprehensive and integrated manner that allows youth to move toward legal approaches to survival. Ferguson and Xie (see also Ferguson, 2012) outline that current programming needs to be reorganized to provide multiple service portals for youth. Programs which combine vocational training and mental health support are an example of attempts to provide a more comprehensive service approach (Ferguson, 2012; Ferguson & Xie, 2008). This strategy endeavors to identify adverse experiences and reorient their impact, while addressing issues of self-sufficiency. This type of approach has shown to produce a range of positive outcomes (Ferguson, 2012; Ferguson & Xie, 2008). Furthermore, peer-based prevention programs that exploit social networking technology can be adopted to link homeless youth with nonstreet peers, supportive adults, and service providers to alter the nature of peer associations and expectations (Ferguson et al., 2016; Rice, Milburn, & Monro, 2011). Therefore, changing the way current programming is offered, exploiting changing technologies and forms of communication, and enticing youth to exploit available services are key strategies to tackle the complex array of factors that lead to criminal activity in this population.

In sum, this research is important because it is the first to preliminarily establish the importance of extreme poverty in the social schematic theoretical approach to crime. This work extends our understanding of the process beyond a single sample. It provides a more comprehensive integrated understanding of street youth offending. The study also establishes a framework for future research to examine other adverse experiences, explorations of causal evolutions, and reciprocal causal processes. It is through additional work that we can more clearly establish how the social schematic theory proposed by Simons and Burt helps us understand criminal participation in marginal groups and locations.

Footnotes

Acknowledgements

The author would like to acknowledge the financial support of the Social Sciences and Humanities Research Council of Canada and recognize the contributions of the anonymous reviewers. Special thanks also to Nic Swager and Fiona Kay for their assistance.

Notes

Stephen W. Baron is a professor in the Department of Sociology at Queen’s University. His research focuses on homeless street youth and crime and substance abuse. His work has appeared in Criminology, Justice Quarterly, Journal of Research in Crime and Delinquency, Journal of Criminal Justice, and Deviant Behavior.

References

Agnew

R. A.

(2006). Pressured into crime: An overview of general strain theory. Los Angeles, CA: Roxbury.

Akers

R. L.

(1998). Social learning and social structure: A general theory of crime and deviance. Boston, MA: Northeastern University Press.

Anderson

(1999). Code of the street. New York, NY: W.W. Norton.

Barman-Adhikari

Rice

(2014). Social networks as understanding the context for understanding employment services utilization among homeless youth. Evaluation and Program Planning, 45, 90-101.

Baron

S. W.

(2003a). Self-control, social consequences, and criminal behavior: Street youth and the general theory of crime. Journal of Research in Crime and Delinquency, 40, 403-425.

Baron

S. W.

(2003b). Street youth violence and victimization. Trauma, Violence, & Abuse, 4, 22-44.

Baron

S. W.

(2004). General strain, street youth, and crime: Testing Agnew’s revised theory. Criminology, 42, 457-483.

Baron

S. W.

(2009). Differential coercion, street youth, and violent crime. Criminology, 47, 239-268.

Baron

S. W.

(2015). Differential social support, differential coercion, and organized criminal activities. Justice Quarterly, 32, 1089-1117.

10.

Baron

S. W.

(2017). It’s more than the code: Exploring the factors that moderate the street code’s relationship with violence. Justice Quarterly, 34, 491-516.

11.

Baron

S. W.

Forde

D. R.

Kay

F. M.

(2007). Self-control, risky lifestyles, and situation: The role of opportunity and context in the general theory. Journal of Criminal Justice, 35, 119-136.

12.

Barr

A. B.

Simons

R. L.

(2015). Different dimensions, different mechanisms? Distinguishing relationship status and quality effects on desistance. Journal of Family Psychology, 29, 360-370.

13.

Bender

K. A.

Thompson

S. J.

Ferguson

K. M.

Yoder

J. R.

Kern

(2014). Trauma among street-involved youth. Journal of Emotional and Behavioral Disorders, 22, 53-64.

14.

Bernstein

D. P.

Fink

(1998). Childhood Trauma Questionnaire: A retrospective self-report manual. San Antonio, TX: The Psychological Corporation.

15.

Burt

C. H.

Simons

R. L.

(2015). Interpersonal racial discrimination, ethnic-racial socialization, and offending: Risk and resilience among African American females. Justice Quarterly, 32, 532-570.

16.

Burt

C. H.

Sweeten

Simons

R. L.

(2014). Self-control through emerging adulthood: Instability, multidimensionality, and criminological significance. Criminology, 52, 450-487.

17.

Crick

N. R.

Dodge

K. A.

(1994). A review and reformulation of social information processing mechanisms in children’s social adjustment. Psychological Bulletin, 115, 74-101.

18.

Dachner

Tarasuk

(2002). Homeless “squeegee kids”: Food insecurity and daily survival. Social Science & Medicine, 54, 1039-1049.

19.

Delisi

Hochstetler

Murphy

D. S.

(2003). Self-control behind bars: A validation study of the Grasmick et al. scale. Justice Quarterly, 20, 241-263.

20.

De Rosa

C. J.

Montgomery

S. B.

Kipke

M. D.

Iverson

J. L.

Unger

J. B

. (1999). Service utilization among homeless and runaway youth in Los Angeles, California: Rates and reasons. Journal of Adolescent Health, 24, 449-458.

21.

Ferguson

K. M.

(2012). Merging the fields of mental health and social enterprise: Lessons from abroad and cumulative findings from research with homeless youth. Community Mental Health, 48, 490-502.

22.

Ferguson

K. M.

Bender

Thompson

S. J.

(2015a). Gender, coping strategies, homelessness stressors, and income generation among homeless young adults. Social Science & Medicine, 135, 47-55.

23.

Ferguson

K. M.

Bender

Thompson

S. J.

(2015b). Risk and resilience factors associated with formal and informal income generation among homeless young adults in three U.S. cities. Youth & Society. Advance online publication. doi:10.1177/0044118X15600722

24.

Ferguson

K. M.

Bender

Thompson

S. J.

(2016). Predicting illegal income generation among homeless male and female young adults: Understanding strains and responses to strain. Children and Youth Services Review, 63, 101-109.

25.

Ferguson

K. M.

Bender

Thompson

S. J.

Maccio

E. M.

Pollio

(2012). Employment status and income generation among homeless young adults: Results from a five-city, mixed methods study. Youth & Society, 44, 385-407.

26.

Ferguson

K. M.

Bender

Thompson

S. J.

Xie

Pollio

(2011). Correlates of street-survival behaviors in homeless young adults in four U.S. cities. Journal of Orthopsychiatry, 81, 401-409.

27.

Ferguson

K. M.

Xie

(2008). Feasibility study of the Social Enterprise Intervention with homeless youth. Research on Social Work Practice, 18, 5-19.

28.

Forde

D. R.

Baron

S. W.

Scher

C. D.

Stein

M. B.

(2012). Factor structure and reliability of the Childhood Trauma Questionnaire and prevalence estimates of trauma for male and female street youth. Journal of Interpersonal Violence, 27, 364-379.

29.

Forde

D. R.

Kennedy

L. W.

(1997). Risky lifestyles, routine activities, and the general theory of crime. Justice Quarterly, 14, 265-294.

30.

Gaetz

(2004). Safe streets for whom? Homeless youth, social exclusion and criminal victimization. Canadian Journal of Criminology and Criminal Justice, 46, 423-456.

31.

Gaetz

O’Grady

(2002). Making money: Exploring the economy of young homeless workers. Work, Employment and Society, 16, 433-456.

32.

Gottfredson

Hirschi

(1990). A general theory of crime. Stanford, CA: Stanford University Press.

33.

Grasmick

H. G.

Tittle

C. R.

Bursick

R. J.

Jr. Arneklev

B. J.

(1993). Testing the core empirical implications of Gottfredson and Hirschi’s general theory of crime. Journal of Research in Crime & Delinquency, 30, 5-29.

34.

Hagan

McCarthy

(1997). Mean streets: Youth crime and homelessness. Cambridge, MA: Cambridge University Press.

35.

Hayes

A. F.

(2013). Introduction to mediation, moderation, and conditional process analysis: A regression based approach. New York, NY: The Guilford Press.

36.

Hirschi

(1969). Causes of delinquency. Berkley: University of California Press.

37.

Hoyt

D. R.

Ryan

K. D.

Cauce

A. M.

(1999). Personal victimization in a high-risk environment: Homeless and runaway adolescents. Journal of Research in Crime & Delinquency, 36, 371-392.

38.

Kipke

M. D.

Unger

J. B.

O’Connor

Palmer

R. F.

LaFrance

(1997). Street youth, their peer group affiliation, and differences according to residential status, subsistence patterns, and use of services. Adolescence, 32, 655-669.

39.

Kort-Butler

L. A.

Tyler

K. A.

Melander

L. A.

(2011). Childhood maltreatment, parental monitoring, and self-control among homeless young adults. Criminal Justice and Behavior, 38, 1244-1264.

40.

Longshore

Turner

Stein

J. A.

(1996). Self-control in a criminal sample: An examination of construct validity. Criminology, 34, 209-227.

41.

McCarthy

(1996). The attitudes and actions of others: Tutelage and Sutherland’s theory of differential association. British Journal of Criminology, 36, 135-147.

42.

McCarthy

Hagan

(2001). When crime pays: Capital, competence, and criminal success. Social Forces, 79, 1035-1060.

43.

McCarthy

Hagan

Cohen

L. W.

(1998). Uncertainty, cooperation, and crime: Understanding the decision to co-offend. Social Forces, 77, 155-176.

44.

McCarthy

Hagan

Martin

M. J.

(2002). In and out of harm’s way: Violent victimization and the social capital of fictive street families. Criminology, 40, 831-865.

45.

Nunnally

J. C.

(1978). Psychometric theory (2nd ed.). New York, NY: McGraw-Hill.

46.

O’Grady

Gaetz

(2004). Homelessness, gender and subsistence: The case of Toronto street youth. Journal of Youth Studies, 7, 397-416.

47.

Piquero

A. R.

Rosay

A. B.

(1998). The reliability and validity of Grasmick et al.’s self-control scale. A comment on Longshore et al. Criminology, 36, 157-173.

48.

Rice

Milburn

N. G.

Monro

(2011). Social networking technology, social network composition, and reductions in substance use among homeless adolescents. Prevention Science, 12, 80-88.

49.

Scher

Stein

M. B.

Asmundson

G. J. G.

McCreary

D. R.

Forde

D. R.

(2001). The Childhood Trauma Questionnaire in a community sample: Psychometric properties and normative data. Journal of Traumatic Stress, 14, 843-857.

50.

Simons

R. L.

Barr

A. B.

(2014). Shifting perspectives: Cognitive changes mediate the impact of romantic relationships on desistance from crime. Justice Quarterly, 31, 793-821.

51.

Simons

R. L.

Burt

C. H.

(2011). Learning to be bad: Adverse social conditions, social schemas, and crime. Criminology, 49, 553-598.

52.

Simons

R. L.

Burt

C. H.

Barr

A. B.

Lei

Stewart

(2014). Incorporating routine activities, activity spaces, and situational definitions into the social schematic theory of crime. Criminology, 52, 655-687.

53.

Thrane

L. E.

Hoyt

D. R.

Whitbeck

L. B.

Yoder

K. A.

(2006). Impact of family abuse on running away, deviance, and street victimization among homeless rural and urban youth. Child Abuse & Neglect, 30, 1117-1128.

54.

Tyler

K. A.

(2008). Social network characteristics and risky sexual and drug related behaviors among homeless young adults. Social Science Research, 37, 673-685.

55.

Tyler

K. A.

Johnson

K. A.

(2004). Victims and offenders: Accounts of payback, invulnerability, and financial gain among homeless youth. Deviant Behavior, 25, 427-449.

56.

Tyler

K. A.

Johnson

K. A.

(2006). A longitudinal study of the effects of early abuse on later victimization among high-risk adolescents. Violence and Victims, 21, 287-306.

57.

Tyler

K. A.

Melander

L. A.

(2015). Child abuse, street victimization, and substance use among homeless young adults. Youth & Society, 47, 502-519.

58.

Whitbeck

L. B.

Hoyt

D. R.

(1999). Nowhere to grow. New York, NY: Aldine de Gruyter.