Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2008 Sep 11.
Published in final edited form as: J Hum Ecol. 2008;23(4):285–293. doi: 10.1080/09709274.2008.11906082

Problem Behaviors of Homeless Youth: A Social Capital Perspective

Denitza Bantchevska 1, Suzanne Bartle-Haring 1, Pushpanjali Dashora 1, Tatiana Glebova 1, Natasha Slesnick 1
PMCID: PMC2533425  NIHMSID: NIHMS52848  PMID: 18787647

Abstract

Homeless youth are one of the most marginalized groups in our society. Many researchers identify much higher levels of various problem behaviors among these youth compared to their non-homeless peers. The current study examined the utility of social capital in predicting problem behaviors among homeless youth. Overall, the theoretically derived social capital variable significantly predicted substance use frequency, sexual risk behavior, depression, delinquent behavior as well as number of days homeless. Thus, social capital was useful in understanding and predicting the current life situation among these youth and may be worthy of further study. Findings suggest that meaningful change should utilize interventions that go beyond the individual and are geared towards modifying the social context of individuals’ lives.

Keywords: Social Capital, Adolescence, Homelessness, Problem Behavior

INTRODUCTION

An estimated 500,000 to one million youth are homeless each year in the United States (Centre for Law and Social Policy, 2003). While definitions vary, the Department Of Health and Human Services (DHHS) defines homelessness as ‘a situation in which a youth has no place of shelter and is in need of services and a shelter where he or she can receive supervision and care.’ Homelessness among youth can lead to an acceleration of a number of psychosocial problems including substance abuse, high-risk behavior (e.g. survival sex), health and social problems. This paper examined the utility of social capital in predicting problem behaviors among homeless youth. Identifying predictors of problem behaviors among homeless youth may be particularly useful for identifying targets of prevention and intervention efforts.

Problem Behaviors

Compared to non-homeless youth, homeless youth have significant substance abuse problems and Human Immunodeficiency Virus(HIV) risk. Two studies found that between 69% and 71% of their sample of homeless youth met criteria for an alcohol and/or illicit drug abuse disorder (Baer et al., 2003; Kipke et al., 1997). Irregular use of condoms, early age of first intercourse, survival sex and multiple sex partners are extensively documented among homeless youth (Clements et al., 1997; Rotheram-Borus et al., 1991; Wagner et al., 2001). Moreover, two studies reported that between 10.2 and 11.5% of their sample of street living youth (between the ages of 12 and 24) were positive for HIV (Pfeiffer and Oliver, 1997; Woods et al., 2002).

In addition to the high rate of substance abuse and risky sexual behaviors among homeless youth, rates of depression and delinquency are also high in this population (Mundy et al., 1990; Unger et al., 1997). Between 29% and 83% of homeless youth meet criteria for clinical depression (Unger et. al., 1997; Yates et al., 1988) and Patel and Greydanus (2002) found that 75% of the homeless youth in their sample reported involvement in illegal activities including theft, drug dealing and assault.

Various studies suggest that alienation from broader social institutions leads to homelessness (Bahr, 1973; Peled and Spiro, 1998). Moreover, Ennett et al. (1999) reported that approximately 25% of their sample could not name anyone with whom they had a social relationship. These socially isolated youth were significantly more likely to engage in substance abuse and were almost eight times as likely to report survival sex as youth who reported having a network. This research suggests that a lack of connection to others is associated with problem behaviors among homeless youth.

Social Capital

The concept of social capital was brought into light in the nineteenth century and recently has been made popular by Pierre Bourdieu, James Coleman, and Robert Putnam. James Coleman (1988), a pioneer in the field, defined social capital as those aspects of a social structure including the network of relationships among individuals and between individuals, family members and the community that facilitate actions within the structure. Social capital includes trust and mutual aid (the feeling that people can be relied upon you and the belief that help is available when needed) leading to progressive, productive action (Putnam, 1993). Further, Helliwell and Putnam (1999) reported that education has substantial effects on trust and mutual aid, key indicators of social capital. It is a theoretical construct that has shown great promise for understanding the contextual factors associated with disenfranchised groups. It has also shown some utility in guiding intervention for reintegrating marginalized groups. For example, women in the social caste ‘untouchable’ in India were able to break taboos of untouchability, and change the quality of their lives through collective actions guided by the concept of social capital (see Narayan and Kapoor, 2005).

In his study on youth who dropped out from high school, Coleman (1988) identified five indicators of social capital which predicted dropping out among sophomores: 1) the presence of both parents in the household, 2) one versus four siblings, 3) fewer changes of school since fifth grade, 4) regular attendance at religious services and 5) mother’s high expectations for a child’s educational attainment. According to Coleman, if a family consists of a single parent or too many siblings, parents simply do not have enough time to give attention to their children, which can reduce the social capital among individual children.

Since Coleman’s (1988) seminal study, researchers have identified different indicators for social capital (Onyx and Bullen, 2000; Putnam, 2000; Runyan et al., 1998) in different contexts (Baron et al., 2000). However, much debate remains unresolved as to what social capital is, how it can be measured, and whether it can be measured (Grix, 2001). Partly, the controversies arise from the different functions for which this term has been used. In this study, we were primarily interested in social capital at the family level and used the theoretical concept provided by Coleman (1988). Some researches have converged on usual indicators of social capital that include the density of personal associations, interpersonal trust indicators, and sociological measures such as the size of the household (Callois and Aubert, 2005). These indicators of social capital were also used in the current study.

Social Capital and Problem Behaviors

To date, two studies have explored the role of social capital in understanding youth homelessness. Hagan and McCarthy (1997) identified a lack of social capital in the family and community which they assert precipitated homelessness among youth in Canada. Though not focused on criminal activities of homeless youth, they also found higher rates of youth crime in communities with less social capital. In an examination of factors contributing to homelessness among youth in Russia, Stephenson (2001) similarly concluded that deficiencies in the social network predict homelessness.

Lack of social capital has been identified as an important predictor of problem behaviors among adolescents. Research findings show that it can predict crime rates (Verner and Alda, 2004; Wright et al., 2001), risky sexual behaviors (Crosby et al., 2003), alcohol consumption (Weitzman and Kawachi, 2000) and adolescent depressive symptoms (Fitzpatrick et al., 2005).

Current Study

Building on the work of Hagan and McCarthy (1997), this study has attempted to explore the relationship between social capital and problem behaviors experienced by homeless youth including days homeless. It was hypothesized that a lower level of social capital would predict greater substance use frequency, sexual risk behavior, depression, delinquent behavior as well as days homeless. The findings from this study have the potential to increase our understanding of contextual factors affecting the well-being of homeless youth, which can be useful for treatment development efforts.

METHODS

Participants

Two hundred-seventy youth were engaged into the project. All youth were recruited through the only drop-in center for homeless youth in a southwest urban area as part of two larger studies examining therapy outcome with homeless youth (grants R01 DA13549 and TI 12503). Hence, all youth in this study agreed to the possibility of treatment. Inclusion criteria for youth were that they were (1) between the ages of 14–22, (2) had been living in the area for at least 3 months, with plans to remain for at least 6 months, (3) met Diagnostic and Statistical Manual of Mental Disorders, 4th edition, (DSM-IV) criteria for Alcohol or other Psychoactive Substance Use Disorders, as assessed by the computerized diagnostic interview schedule for children (CDISC, Shaffer, 1992), and (4) met DHHS criteria for homelessness. Youth not eligible for the project but who wished to participate in treatment continued with services provided by the drop-in center and were provided outside referrals.

Procedure

As noted, existing baseline data from two larger studies were examined. In these studies, potentially eligible youth were screened for participation at the drop-in center and those meeting criteria for participation in the study, and who provided written consent, continued with the assessment battery. The youth was the only data source for all cases, and all procedures were approved by the Institutional Review Board of the university through which this project was conducted. The assessment required approximately two hours to complete and was conducted in offices within the drop-in center. Assistance in completing forms was provided as needed by the research assistant and participants were offered a care package at the completion of their assessment which included toiletries, a blanket, socks, underwear, and food items.

Measures

A Demographic Questionnaire was used to access family composition, number of years raised by both parents, and educational status of youth and parents.

The Form 90, developed for National Institute on Alcohol Abuse and Alcoholism funded Project Match (Miller and Del Boca, 1994), provides an estimated frequency of drug and alcohol use in the 90 days prior to their last use. In addition, the semi-structured interview provides the percent days in the period of homelessness, working, education, Alcoholics Anonymous/Narcotics Anonymous attendance and religious service attendance. This tool has shown excellent test-retest reliability for indices of drug use in major categories (Westerberg et al., 1998) including with runaway substance abusing adolescents (Slesnick and Tonigan, 2004) with kappas for different drug classes ranging from .74 to .95. In this sample, alphas for different drug classes ranged from .55 to .94.

The National Youth Survey Delinquency Scale (NYSDS, Elliott and Huizinga, 1983) is a structured interview used as a measure of delinquent behavior. It includes 5 subscales: general theft, crimes against persons, index offenses, drug sales and total delinquency. Internal consistency alphas range between .65 and .92, and criterion correlations between self-report and police or parent data approach .40 (Moffitt, 1989). The overall total delinquency score was used as a dependent variable in the current study.

Depressive symptoms were assessed using the Beck Depression Inventory (BDI-II, Beck, 1996). The BDI-II is a 21-item self-report instrument for measuring depressive symptoms in adults and adolescents age 13 and above. In the current sample alpha = .91 for the total depression score.

The Health Risk Questionnaire incorporates items from the Health Risk Survey (Kann et al., 1989) and the Homeless Youth Questionnaire (Johnson et al., 1996). Lifetime HIV risk as well as risk within the past 3 months were used as dependent variables. Internal reliability for the HIV risk subscale for this sample was alpha = .73.

The Youth Self-Report (YSR, Achenbach and Edelbrock, 1982) assesses children’s behavior across a wide range of areas including problem behaviors and social competencies using a 120-item likert-type scale. The Coping Inventory for Stressful Situations (CISS, Endler and Parker, 1990) consists of 48 items assessing youths’ coping in stressful situations. The scale has been shown to be a valid multidimensional coping measure, and to have adequate construct validity with adolescent and clinical populations (Endler and Parker, 1990).

Social Capital Indicators

Based on the theoretical framework of Coleman (1988) and previous researchers, the indicators of social capital for this study included 1) mutual aid, 2) connection with social institutions, 3) family structure among participants’ parents (single versus two parent), 4) total number of siblings, 5) years that the participant was raised by both biological parents, 6) participant’s education, and 7) parent’s education. Mutual aid, defined as helping and getting help from others, was measured by a composite score of five items on the CISS and two items on the YSR (Cronbach’s alpha was .78). Connection with social institutions was defined as the number of days in the assessment period that the participant was involved in school, religious events, Alcoholics Anonymous/Narcotics Anonymous and employment. The sum of these days as reported by the participant on the Form 90 was used.

RESULTS

Demographic Characteristics of the Sample

The sample (N = 268) included 169 males and 99 females between ages of 14 and 22 (M = 18.6, SD = 2.3). The sample included Anglos (48%), Latinos (30%), and “other” or mixed race (22%). The youth reported on average 56.2 days homeless (SD = 34.2) during a 90 day period prior the assessment. Forty nine percent of youth reported having two parents in their home at the time they left for the shelter. On average, participants were raised by both biological parents for 5 years and had 3 siblings. The homeless youth’s parents’ years of education ranged from 7 to 23 years with a mean of 13.8 years of education (SD = 2.6).

Analyses

In order to test the model presented in Figure 1, we used structural equation modeling with causal indicators for social capital (Bollen and Lennox, 1991). We elected to use causal indicators because social capital is believed to be a cumulative latent variable, not an index. That is, the more indicators of social capital that are available, the more social capital the individual has. There was no expectation, for example, that if the individual had more siblings that they were also more likely to have lived with both of their biological parents longer. In other words, there was no expectation of internal consistency.

Fig. 1. Original Model: causal indicators of social capital, which in turn leads to trouble, homelessness and HIV risk as outcomes. (Asterisks indicate significant paths).

Fig. 1

Open in a new tab

In order to use causal indicators in structural equation modeling several assumptions are made in order for the model to be identified. Bollen (2006) makes several suggestions. First, the indicators of the construct are considered observed exogenous variables in the model, then the construct itself is considered a latent endogenous variable. The observed exogenous variables are freed to covary among themselves. Then the latent variable must have at least two emitted paths - it must have an effect on at least two other constructs. In the model for this study, the latent construct for social capital influences three latent constructs (trouble, homelessness/depression, and HIV risk). Bollen (2006) also suggests that one of the paths that the latent variable emits should be set to 1 in order to “scale” the latent variable. In this instance the social capital latent construct was scaled on the “trouble” latent construct which consisted of delinquency and substance abuse indicators.

We tested the original model as depicted in Figure 1 which included family structure, total number of siblings, years raised with biological parents, connection to institutions, youth’s education, parent or caregiver’s level of education and mutual aid as the causal indicators for social capital. Social capital, in turn, influenced the “trouble,” homelessness/depressive symptoms and HIV risk variables in the model. Using the Expectation-Maximization (EM) algorithm for missing data in Linear Structural Relations (LISREL 8.72) (Joreskog and Sorbom, 2005) with full information maximum likelihood estimation and the asymptotic covariance matrix, the original model fit the data fairly well (χ2 (59) =103.82; p<.001; Root Mean Square Error of Estimation (RMSEA) =.053; N=268). It should be noted that with full information maximum likelihood estimation, the only fit indices available in LISREL are the chi-square and the RMSEA. Examining the estimates for the loadings on the social capital construct (see Table 1), only the education variables were significant. Most of the paths from the observed variables to social capital were negative, with the exception of family structure and parent’s education. This suggests that the social capital latent construct is really assessing the lack of social capital. Social capital significantly predicted homelessness in terms of number of days homeless and depression scores, and significantly predicted HIV Risk. Both of these were in the positive direction, suggesting that lower social capital predicted increased homelessness/depressive symptoms and increased HIV risk. However, there were several inadmissible solutions in the model including negative variance accounted for, and negative error variances.

Table 1. Model estimates for original and trimmed model with standard errors in parentheses.

Paths to social capital Original Trimmed
Family Structure .96 (6.48) deleted
Total # of Siblings −.99 (8.45) deleted
Years Raised by Biological Parents −1.62 (1.45) −.23 (−1.49)
Connection to Institutions −.84 (−1.79) −.12 (.04)*
Youth education −4.64 (1.87)* −.44 (.12)*
Parent Education 13.96 (7.06)* .95 (.42)*
Mutual Aid −.55 (2.13) −3.17 (2.26)
Paths From Social Capital to Outcomes 1.0 5.30 (2.25)*
Trouble
Homelessness 1.0
BDI .19 (.08)* .21 (.10)*
HIV Risk .01 (.00)* .03 (.01)*
Open in a new tab
*

p <.05

We trimmed the model in light of these inadmissible solutions (see Figure 2). In the trimmed model, the causal indicators for social capital included years raised by at least one biological parent, connection to social institutions, years of education of youth and parent, and mutual aid. We also separated depression from the homeless latent variable, so homelessness was simply the observed number of homeless days in the period, and depressive symptoms was the BDI score, rather than a latent construct. We also “scaled” social capital on homeless days, rather than the delinquency and substance abuse variable to determine whether social capital significantly predicted substance abuse. This provided a better fit to the data (χ2(51)=64.40; p> .05; RMSEA =.032). In this model, connection to social institutions, youth’s education, mutual aid and years raised by at least one biological parent were negatively related to the social capital variable, and parents’ years of education was positively related to social capital. In turn, social capital predicted the “trouble” variable, depressive symptoms and HIV risk. Since the majority of the social capital observed variables negatively influenced social capital, the latent construct was the lack of social capital. Lower social capital was then related to more delinquency and substance abuse, more depressive symptoms and higher HIV risk.

Fig. 2. Trimmed model. (Asterisks indicate significant paths).

Fig. 2

Open in a new tab

Note: All exogenous variables were allowed to correlate (not pictured).

Finally, using this trimmed model, we added demographic variables as “control” variables to the model to answer the question whether or not age and ethnicity made a difference in the relationships between social capital and the outcome variables. The model also fit the data well (χ2(50)=66.69; p>.05; RMSEA =.035), but age and ethnicity were not related to social capital, nor were age and ethnicity related to the outcome variables.

DISCUSSION

This study examined the utility of social capital for predicting problem behaviors among homeless adolescents. Our theoretically driven model fit the data as expected; lower levels of social capital were associated with higher levels of delinquency, depression and HIV risk, substance use, and days spent on the street. Overall, findings support our initial hypothesis and support the utility of social capital theory for understanding problem behaviors from a broader contextual orientation. In fact, the findings could be counter-intuitive if not considered from a social capital perspective. For example, the association between higher parent education and more days homeless might be a negative result of parental disengagement from the community and social network, or what Coleman (1988) called a lack of closure. These findings contribute to the current, mostly international, research base that provides empirical support to the theoretical concept of social capital.

Social capital offers a useful framework for understanding and interpreting the struggles among youth who are homeless. Many therapy approaches, including the cognitive-behavioral orientation, focus on addressing individual maladaptive cognitions and behaviors through improving individuals’ skills - such as coping, communication, and emotion regulation. However, individual factors, though important, can be supplemented by understanding interactions among the youth and their environment. Research shows that levels of connection among the youth, their family, social services, schools, and the juvenile justice system contribute to marginalization, and as shown in this study, problem behaviors among those marginalized.

In sum, social capital requires that policy makers and interventionists look beyond the individual when trying to understand homelessness and risk behaviors, and consider the social and structural environment. In fact, structural interventions, interventions that aim at changing the risk behavior by altering the environment in which it occurs, have shown greater utility than individually-focused interventions for reducing HIV risk behaviors (Heise and Elias, 1995). Even though there is no evidence that social capital is a more significant predictor of outcomes compared to individual level factors, more research is needed to determine how social capital can direct intervention efforts with this disenfranchised and vulnerable population because social capital is potentially more malleable. Research is needed to determine how social capital can be increased and if such an increase impacts problem behaviors. Specifically, service providers may assist homeless youth by connecting them with social institutions, schools, religious organizations, and the workforce.

A critical issue facing service providers attempting to mainstream this marginalized population is trust (Bhugra, 1996), or more specifically the development of social capital that enables trusting connections in participant/provider relationships. Homeless youth often remain homeless despite organizations and others who offer services. Homeless youth often do not trust the motivations of these agencies, family members or outreach workers. At the same time, those who seek to reconnect homeless youth may lack trust in homeless youths’ motivations as well as lack the relevant information necessary to integrate them into their programs.

There are a number of limitations of this study that should be considered when interpreting the findings. The parent study was not designed to evaluate the utility of social capital in predicting homeless youths’ problem behaviors. Other indicators of social capital such as quality and quantity of friendship relationships, trust and family relations were not measured, and would likely also contribute to the findings. Yet, the measure of social capital used in this study predicted our outcomes, suggesting that social capital is a valid predictor. Our sample was relatively small and was a sample of convenience. Homeless, substance-abusing youth seeking services through a drop-in center are not representative of homeless youth who refuse services, and thus may be more disconnected from the system. Those with even greater disconnection might show a different pattern of results.

Overall, however, this study provides support to a promising theoretical framework and offers a broader ecological understanding of common problems experienced by homeless youth. Our findings contribute to a scarce body of research on a vulnerable, hard-to-study population, along with applying a novel statistical approach. And, while no standardized measures of social capital exist, this study further supports the use of typical indicators of social capital, and provides support for additional research in this area.

Acknowledgments

This work was supported by NIDA grant (R01 DA13549) and CSAT grant (TI12503).

References

  1. Achenbach TM, Edelbrock CS. Child Psychiatry. University of Vermont; Burlington, VT: 1982. Manual for the Child Behavior Checklist and Child Behavior Profile. [Google Scholar]
  2. Baer J, Ginzler J, Peterson P. DSM-IV alcohol and substance abuse and dependence in homeless youth. Journal of Studies on Alcohol. 2003;64:5–15. doi: 10.15288/jsa.2003.64.5. [DOI] [PubMed] [Google Scholar]
  3. Bahr M. Skid Row: An Introduction to Disaffiliation. Oxford University Press; New York: 1973. [Google Scholar]
  4. Baron S, Field J, Schuller T. Social Capital: Critical Perspectives. Oxford University Press; Oxford: 2000. [Google Scholar]
  5. Beck A. Beck Depression Inventory. 2. The Psychological Corporation; San Antonio: 1996. [Google Scholar]
  6. Bhugra D. Homelessness and Mental Health Studies in Social and Community Psychiatry. Cambridge University Press; Cambridge: 1996. [Google Scholar]
  7. Bollen K. SEM. Workshop with LISREL 8; Chicago, IL. September, 2006; 2006. [Google Scholar]
  8. Bollen K, Lennox R. Conventional wisdom on measurement: A structural equation perspective. Psychological Bulletin. 1991;110:305–314. [Google Scholar]
  9. Callois J, Aubert F. Towards indicators of social capital for regional development resulting: Regional growth diaries. Paper presented at International Conference of the Regional Studies Association; University of Aalborg. May, 2005; 2005. [Google Scholar]
  10. Center for Law and Social Policy. Leave No Youth Behind: Opportunities for Congress to Reach Disconnected Youth. Washington, DC: 2003. [Google Scholar]
  11. Clements K, Gleghorn A, Garcia D, Katz M, Marx R. A risk profile of street youth in Northern California: Implications for gender-specific human immunodeficiency virus prevention. Journal of Adolescent Health. 1997;20:343–353. doi: 10.1016/S1054-139X(97)00033-5. [DOI] [PubMed] [Google Scholar]
  12. Coleman J. Social capital in the creation of human capital. American Journal of Sociology. 1988;94:S95–S120. [Google Scholar]
  13. Crosby R, Holtgrave D, DiClemente R, Wingood G, Gayle J. Social capital as a predictor of adolescents’ sexual risk behavior: A state-level exploratory study. AIDS and Behavior. 2003;7:245–252. doi: 10.1023/a:1025439618581. [DOI] [PubMed] [Google Scholar]
  14. Elliot D, Huizinga D. Social class and delinquent behavior in a national youth panel. Criminology: An Interdisciplinary Journal. 1983;21:149–177. [Google Scholar]
  15. Endler N, Parker J. Coping Inventory for Stressful Situations (CISS): Manual. Multi-Health Systems; Toronto: 1990. [Google Scholar]
  16. Ennett S, Bailey S, Federman B. Social network characteristics associated with risky behaviors among runaway and homeless youth. Journal of Health and Social Behavior. 1999;40:63–78. [PubMed] [Google Scholar]
  17. Fitzpatrick K, Piko B, Wright D, LaGory M. Depressive symptomatology, exposure to violence, and the role of social capital among African American adolescents. American Journal of Orthopsychiatry. 2005;75:262–274. doi: 10.1037/0002-9432.75.2.262. [DOI] [PubMed] [Google Scholar]
  18. Grix J. Social capital as a concept in the social sciences: The current state of the debate. Democratization. 2001;8:189–210. [Google Scholar]
  19. Hagan J, McCarthy B. Mean Streets: Youth Crime and Homelessness. Cambridge University Press; Cambridge: 1997. [Google Scholar]
  20. Heise L, Elias C. Transforming aids prevention to meet women’s needs: A focus on developing countries. Journal of Social Science and Medicine. 1995;40:931–943. doi: 10.1016/0277-9536(94)00165-p. [DOI] [PubMed] [Google Scholar]
  21. Helliwell J, Putnam R. NBER Working Paper No. 7121. National Bureau of Economic Research; Cambridge: 1999. Education and Social Capital. [Google Scholar]
  22. Johnson T, Aschkenasy J, Herbers M, Gillenwater S. Self-reported risk factors for AIDS among homeless youth. AIDS Education and Prevention. 1996;8:308–322. [PubMed] [Google Scholar]
  23. Joreskog K, Sorbom D. Structural Equation Modeling Software. Scientific Software International; Chicago, IL: 2005. LISREL 8.72. [Google Scholar]
  24. Kann L, Nelson G, Jones J, Kolbe L. Establishing a system of complementary school-based surveys to annually assess HIV-related knowledge, beliefs, and behaviors among adolescents. Journal of School Health. 1989;59:55–58. doi: 10.1111/j.1746-1561.1989.tb05392.x. [DOI] [PubMed] [Google Scholar]
  25. Kipke M, Montgomery S, Simon T, Iverson E. Substance abuse disorders among runaways and homeless youth. Substance Use and Misuse. 1997;32:969–986. doi: 10.3109/10826089709055866. [DOI] [PubMed] [Google Scholar]
  26. Miller W, Del Boca F. Measurement of drinking behavior using the Form 90 family of instruments. Journal of Studies on Alcohol. 1994;12:112–118. doi: 10.15288/jsas.1994.s12.112. [DOI] [PubMed] [Google Scholar]
  27. Moffitt T. Accommodating self-report methods to a low-delinquency culture: Experience from New Zealand. In: Klein MW, editor. Cross-national Research in Self-Reported Crime and Delinquency. Kluwer Academic; Norwell, MA: 1989. pp. 43–66. [Google Scholar]
  28. Mundy P, Robertson M, Robertson J, Greenblatt M. The prevalence of psychotic symptoms in homeless adolescents. Journal of the American Academy of Child and Adolescent Psychiatry. 1990;29:724–731. doi: 10.1097/00004583-199009000-00008. [DOI] [PubMed] [Google Scholar]
  29. Narayan D, Kapoor S. Beyond ideologies: Creating wealth for the poor. Paper presented at the Arusha Conference, “New Frontiers of Social Policy”; Arusha, Tanzania. 2005. [Google Scholar]
  30. Onyx J, Bullen P. Measuring social capital in five communities. The Journal of Applied Behavioral Science. 2005;36:23–42. [Google Scholar]
  31. Patel D, Greydanus D. Homeless Adolescents in the United States: An overview for pediatricians. International Pediatrics. 2002;17:71–75. [Google Scholar]
  32. Peled E, Spiro S. Goal focused evaluation: Lessons from a study of a shelter for homeless youth. Evaluation. 1998;4:455–468. [Google Scholar]
  33. Pfeiffer R, Oliver J. A study of HIV seroprevalence in a group of homeless youth in Hollywood, California. Journal of Adolescent Health. 1997;20:339–342. doi: 10.1016/S1054-139X(97)00038-4. [DOI] [PubMed] [Google Scholar]
  34. Putnam R. The prosperous community: Social capital and public life. The American Prospect. 1993;13:35–42. [Google Scholar]
  35. Putnam R. Bowling Alone: The Collapse and Revival of American Community. Simon and Schuster; New York: 2000. [Google Scholar]
  36. Rotheram-Borus M, Koopman C, Haignere C, Davies M. Reducing HIV sexual risk behaviors among runaway adolescents. Journal of the American Medical Association. 1991;266:1237–1241. [PubMed] [Google Scholar]
  37. Runyan D, Hunter W, Socolar R, Amaya-Jackson L, English D, Landsverk J, Dubowitz H, Browne D, Bangdiwala S, Mathew R. Children who prosper in unfavorable environments: The relationship to social capital. Pediatrics. 1998;101:12–18. doi: 10.1542/peds.101.1.12. [DOI] [PubMed] [Google Scholar]
  38. Shaffer D. The Diagnostic Interview Schedule for Children - 2.3 Version. Columbia University; New York: 1992. [Google Scholar]
  39. Slesnick N, Tonigan J. Assessment of alcohol and other drugs used by runaway youths: A test-retest study of the Form 90. Alcoholism Treatment Quarterly. 2004;22:21–34. doi: 10.1300/J020v22n02_03. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Stephenson S. Street children in Moscow: using and creating social capital. The Editorial Board of the Sociological Review. 2001;49:530–547. [Google Scholar]
  41. Unger J, Kipke M, Simon T, Montgomery S, Johnson C. Homeless youths and young adults in Los Angeles: Prevalence of mental health and substance abuse disorders. American Journal of Community Psychology. 1997;25:371–394. doi: 10.1023/a:1024680727864. [DOI] [PubMed] [Google Scholar]
  42. Verner D, Alda E. Youth at risk, social exclusion, and intergenerational poverty dynamics: A new survey instrument with application to Brazil. 2004 Manuscript submitted for publication. [Google Scholar]
  43. Wagner L, Carlin L, Cauce A, Tenner A. A snapshot of homeless youth in Seattle: Their characteristics, behaviors, and beliefs about HIV protective strategies. Journal of Community Health. 2001;26:219–233. doi: 10.1023/a:1010325329898. [DOI] [PubMed] [Google Scholar]
  44. Weitzman E, Kawachi I. Giving means receiving: The protective effect of social capital on binge drinking on college campuses. American Journal of Public Health. 2000;90:1936–1939. doi: 10.2105/ajph.90.12.1936. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Westerberg V, Tonigan J, Miller W. Reliability of Form 90D: An instrument for quantifying drug use. Substance Use. 1998;19:179–189. doi: 10.1080/08897079809511386. [DOI] [PubMed] [Google Scholar]
  46. Woods E, Samples C, Melchiono M, Keenan P, Durrell J, Harris S the Boston HAPPENS Program Collaborators. Initiation of services in the Boston HAPPENS program: Human immunodeficiency virus-positive, homeless, and at-risk youth can access services. AIDS Patient Care and STDs. 2002;16:497–510. doi: 10.1089/10872910260351276. [DOI] [PubMed] [Google Scholar]
  47. Wright J, Cullen F, Miller J. Family social capital and delinquent involvement. Journal of Criminal Justice. 2001;29:1–9. [Google Scholar]
  48. Yates G, MacKenzie R, Pennbridge J, Cohen E. A risk profile comparison of runaway and non-runaway youth. American Journal of Public Health. 1988;78:820–821. doi: 10.2105/ajph.78.7.820. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES