Predictors of Transience Among Homeless Emerging Adults

Abstract

This study identified predictors of transience among homeless emerging adults in three cities. A total of 601 homeless emerging adults from Los Angeles, Austin, and Denver were recruited using purposive sampling. Ordinary least squares regression results revealed that significant predictors of greater transience include White ethnicity, high school degree or equivalent, homeless residence in the 6 months prior to the study, longer period of homelessness, history of arrest and juvenile detention, earning income through informal sources, history of physical abuse, alcohol/drug addiction, and resilience characteristics. Quantitative findings were expanded upon with data from in-depth interviews with 145 of these homeless emerging adults regarding their reasons and motivations for transience. Identifying predictors of transience will facilitate customizing interventions that, when needed, aim to stabilize homeless emerging adults, prevent their mobility, and/or offer mobile services to them on the road.

Keywords

homeless emerging adults mobility transience risk behavior resilience

Prior research suggests that homeless emerging adults are a highly mobile population (Ferguson, Jun, Bender, Thompson, & Pollio, 2010; Hyde, 2005; Rew, Taylor-Seehafer, Thomas, & Yockey, 2001) with two thirds identifying themselves as “travelers” (Sanders, Lankenau, Jackson-Bloom, & Hathazi, 2008). In one study of 139 homeless urban nomads in New York City, participants reported visiting approximately 600 different cities/towns; each averaged 31 unique trips to different cities over a 3-year period (Des Jarlais, Perlis, & Settembrino, 2005). In another study of 146 homeless emerging adults in Austin, TX and Los Angeles, CA, participants averaged 129 and 52 intercity moves, respectively, since leaving home. In both cities, those who had been homeless longer reported greater transience (Ferguson et al., 2010).

Common terms used to describe transient homeless young people include travelers (Finley & Finley, 1999; Hyde, 2005; Lankenau et al., 2007; Sanders et al., 2008), transgressive youth (Mallett, Rosenthal, Myers, Milburn, & Rotheram-Borus, 2004), migrants, and geographically mobile urban nomads (Des Jarlais et al., 2005). Other references used among traveling homeless young people include crusty, gutter punk, and squatter (Finley & Finley, 1999). Travelers are nomadic in that they often do not remain in a city for longer than a few weeks or months and commonly move around the United States by train, bus, and car. Some travel in pursuit of adventure, work, drugs, relationships, and more comfortable weather; others seek to avoid law enforcement and the legal consequences associated with their behaviors (e.g., warrants, gang activity, violence; Lankenau et al., 2007). Yet given their often lengthy episodes of traveling and street involvement, it is difficult to assist them in transitioning off the streets and into housing. Stringent shelter and transitional-living program eligibility criteria that residents abstain from drug/alcohol use, maintain employment, and enroll in educational programs require travelers to undergo major changes in their attitudes and behaviors as a precursor to housing (Hyde, 2005).

Collectively, travelers are a unique subgroup of homeless individuals who can be distinguished demographically and behaviorally from the more geographically stable homeless locals. Demographically, traveling homeless emerging adults are more likely to be male, White, heterosexual, and high school graduates or equivalent (Sanders et al., 2008). They are less likely to be formally employed, earning income instead through informal sources, such as under-the-table work and day labor (Hahn, Page-Shafer, Ford, Paciorek, & Lum, 2008; Sanders et al., 2008). Research also has found a greater reliance on survival behaviors (drug dealing, theft, prostitution) among travelers, since these activities offer immediate economic returns, can be used in any city, and require fewer commitments, relationships and responsibilities than traditional employment (Ferguson, Bender, Thompson, Xie, & Pollio, 2011).

Behavioral risk factors of travelers include high rates of drug/alcohol abuse, high-risk sexual activity, and criminal activity (Hahn et al., 2008; Sanders et al., 2008). For instance, travelers who are addicted to substances may move to new cities in search of their drug supply (Sanders et al., 2008). Travelers are frequently disconnected from supportive adults and institutions; instead, they associate with small groups of like-minded traveling peers in locations such as the streets, public spaces, railway/bus stations, and abandoned buildings (Mallett et al., 2004). The traveling lifestyle includes high levels of drug use, informal and often illegal work, and high-risk behaviors, all of which can become normalized among this population (Sanders et al., 2008).

The association between transience and myriad negative outcomes has been established, such as increased risk for health problems, including sexually transmitted infections, and HIV/AIDS (Des Jarlais et al., 2005). Travel is also associated with attrition from health and mental health treatment and disruption in service engagement, both of which hinder the delivery of effective or prescribed doses of treatment (Hahn et al., 2008). This traveling lifestyle—characterized by transient locations, few stable relationships, and loose connections to supportive adults and institutions—places these emerging adults at increased risk for trauma exposure, victimization (e.g., physical and sexual violence and exploitation on the streets), and self-harm (e.g., drug use, survival sex, criminal activity) compared to sheltered and precariously housed peers (Bender, Ferguson, Thompson, Komlo, & Pollio, 2010; Mallett et al., 2004). Inconsistent health and mental health treatment in this population can also result in an increased risk of psychopathology (e.g., posttraumatic stress disorder) following victimization (Bender et al., 2010).

In contrast, transience can also be used as a coping or adaptive behavior as well as a means through which to accomplish broader life goals. In these cases, frequent moves may be related with more prosocial outcomes (Rew et al., 2001). However, these associations are less well documented in the extant literature on homeless emerging adults (Ferguson et al., 2010; Rew et al., 2001). To illustrate, homeless emerging adults involved in abusive relationships at home, with intimate partners, or in gangs may move to different cities to escape violence or ensure their personal safety. Others may relocate to new cities to pursue their educational and employment goals. In the literature on housed emerging adults, transience is a common characteristic during emerging adulthood, which enables young people to accomplish their personal and professional goals. Transience is associated with leaving home to become independent of parents, attending or completing college or vocational school, entering or leaving intimate relationships, and starting or ending a job (Arnett, 2004). Given both the negative and positive outcomes associated with transience, as well as limited knowledge of the motivations for transience among homeless emerging adults, it is crucial to better understand who traveling homeless emerging adults are and why they travel. Findings in turn will facilitate customizing interventions that, when needed, aim to stabilize them, prevent their mobility, and/or offer mobile services to them on the road.

Theoretical Framework

The risk and resilience framework is useful in explaining how intrapersonal and environmental factors promote and inhibit positive development (Jenson & Fraser, 2010). Risk factors are those intrapersonal and environmental factors in the lives of emerging adults that can increase the likelihood of the occurrence of future problem behaviors and negative outcomes (Fraser, Galinsky, & Richman, 1999; Hawkins, Catalano, & Miller, 1992). In contrast, resiliency refers to the social and psychological competence of emerging adults to cope with and adapt to adversity, stressors, and risk in their environment (Wagnild & Young, 1993).

Risk and resilience are important concepts in emerging adulthood (late teens through the mid to late 20s), as this is a developmental period in which emerging adults engage in identity exploration and development in an effort to cement the foundations for their personal and professional lives as adults (Arnett, 2004). Characterized as a period of instability and transition, and at the same time, opportunities and possibilities, emerging adults often experience anxiety and fear as well as optimism and resilience regarding their choices and decisions (Arnett, 2007). Understanding how risk and resilience factors interact among homeless emerging adults can help these young people successfully navigate this developmental period and make choices that enable them to thrive in their occupations, relationships, and responsibilities as adults.

As evidenced in extant research with homeless emerging adult travelers, studies have largely adopted a deviance lens to identify the health, mental health, and behavioral risks associated with traveling. Although important to understand from an assessment and intervention standpoint, problem-based depictions of this population and their risk behaviors fail to recognize and build upon their entrepreneurial skills and resiliency (Bridgman, 2001). For instance, through informal social networks, seasoned travelers offer advice and support to those less experienced travelers regarding specific cities that are considered receptive and accommodating to homeless populations (Bender, Thompson, McManus, Lantry, & Flynn, 2007). Additionally, substance use, high-risk sexual behaviors, criminal activity, and continual relocation may be effective coping mechanisms used by travelers to survive and thrive amid their transient lifestyle (Ferguson et al., 2011; Rew et al., 2001). These examples suggest that emerging adult travelers may possess resilient characteristics, such as independence, self-reliance, and self-confidence in their abilities and in coping with life adversities (Rew et al., 2001). Furthermore, perceiving these emerging adults as deviants from the norm attributes individual blame for their circumstances and thus negates larger systemic, structural, and environmental forces that influence their decisions, circumstances, and outcomes. For example, parental abuse is frequently cited among the primary reasons why homeless emerging adults leave home (Thompson, McManus, & Voss, 2006; Whitbeck, 2009). This may suggest that emerging adults who leave home possess additional resilient characteristics such as an inner sense of life meaning or greater purpose for them in their lives beyond their dysfunctional family system (Rew et al., 2001).

The present study addresses this gap by examining both risk and resilience predictors of transience. Quantitative study findings are complemented with rich descriptions of homeless emerging adults’ motivations for traveling. Grounded theory was selected for the qualitative study since limited empirical research exists to understand the intersection among transience, risk, and resiliency in homeless emerging adults, in particular regarding their motivations for why they travel—whether to escape life responsibilities, to pursue broader life opportunities, and/or to cope with life adversity. Furthermore, no studies of which we are aware have identified predictors of transience from a risk and resilience perspective. Thus, this mixed-methods study investigates the intricacies of transience among homeless emerging adults by exploring their demographic characteristics (age, gender, ethnicity, and education); risk factors (being homeless or in a temporary shelter, length-of-time homeless, criminal history, income generation from informal sources, victimization, and substance abuse); and resilience characteristics (self-reliance, meaning, equanimity, perseverance, and existential aloneness). Developing a more comprehensive understanding of traveling homeless emerging adults’ resiliency may inform future study of how their strengths can buffer the effects of risk and negative outcomes.

Based on this literature, seven hypotheses guided this study. Greater transience is expected to be reported by those who:

Hypothesis 1a (H1a): are White.

Hypothesis 1b (H1b): have a high school degree (or equivalent).

Hypothesis 2a (H2a): are currently homeless or in temporary shelter.

Hypothesis 2b (H2b): have been on the streets longer.

Hypothesis 3a (H3a): have histories of juvenile detention.

Hypothesis 3b (H3b): have histories of arrests.

Hypothesis 4 (H4): earn income from informal sources.

Hypothesis 5a (H5a): report prior physical abuse.

Hypothesis 5b (H5b): report prior emotional abuse.

Hypothesis 6 (H6): meet criteria for alcohol/drug addiction.

Hypothesis 7 (H7): report higher levels of resilience.

Method

Research Settings

This cross-sectional, comparative study of homeless emerging adults included participation from agencies providing services to homeless youth in Los Angeles, CA; Austin, TX; and Denver, CO. Agencies were selected based on their existing relationships with researchers and their commitment to host the study. Participating agencies were multiservice, nonprofit organizations that offer homeless, runaway, and at-risk young people a comprehensive system of care including street outreach, meals, shelter, health care, counseling, and educational and employment services. Human subjects’ approval was received by each investigator’s university.

Participants and Recruitment Procedures

Emerging adult participants (ages 18-24) were recruited from host agencies providing services to homeless/runaway youth in the three cities. Recruitment of participants took place from March 2010 to July 2011. Using purposive sampling, researchers and trained research assistants in each city recruited 200 emerging adult participants from the host agencies using similar methods (201 participants were recruited in Denver). Purposeful, maximum variation sampling was used to recruit a diverse array of information-rich participants or those homeless emerging adults from whom researchers can learn a lot about the central issues worthy of in-depth inquiry in this study (Patton, 1990). Researchers recruited homeless emerging adults from among three different service programs within the host agencies: (a) street outreach/drop-in centers (nonresidential), (b) residential short-term and mid-length shelters (30 days to up to 6 years), and (c) transitional housing (long-term, subsidized housing for homeless youth). Recruitment occurred on weekdays during morning, afternoon, and evening services in the host agencies’ shelters, drop-in centers, and street-outreach services.

To participate in the study, emerging adults had to (a) be 18 to 24 years of age, (b) have spent at least 2 weeks away from home in the month before the interview, and (c) provide written informed consent. An eligibility screening form was used to determine the participants’ ages and length of time away from home. Emerging adults were excluded if they were incapable of comprehending the consent form because of cognitive limitations or if they were noticeably intoxicated or high at the time of the interview. In the latter case, they were asked to return for an interview at a time when they were not intoxicated. Host-agency staff made the determination whether clients were eligible for recruitment based on their knowledge of them and their current level of sobriety. Eligible participants were referred to interviewers who explained the study procedures, secured written consent, and administered a two-part interview with them.

Quantitative and Qualitative Design and Data Collection

A QUANT → qual (i.e., qualitative follow-up) design was used in this study in which a smaller qualitative study helped evaluate and interpret the initial results from a largely quantitative study (Morgan, 1998). A two-part quantitative/qualitative interview was used. The first part of the interview included a 45-minute quantitative retrospective questionnaire that sought information on transience, trauma history, coping strategies, and a host of risk and resilience factors. Interviewers read questions and response options aloud to participants, who responded verbally. For one sensitive section of the survey focusing on childhood trauma, participants were given the option of reading and answering survey questions themselves. All interviews were conducted within the host agencies in private rooms. Participants were compensated with a US$10 gift card for their involvement in the quantitative interview. A total of 601 emerging adults participated in these quantitative interviews across the three cities.

The second part of the interview involved the first 50 of 200 participants in each city, who were asked semistructured qualitative questions. If an emerging adult declined this interview, subsequent young people were offered the opportunity until 50 interviews were achieved in each city (45 interviews were completed in Austin). Based on the researchers’ prior qualitative research experience with homeless youth, this number of interviews was deemed acceptable to meet qualitative redundancy requirements (Seidman, 1991). The qualitative interview took 30 to 60 minutes and participants were compensated an additional US$10 gift card. Interview questions covered four broad topics, focusing loosely on emerging adults’ perceptions of (a) trauma (e.g., When you hear the word “trauma,” what do you think of?); (b) protection and coping strategies to prevent and manage trauma (e.g., What do you think are your greatest strengths that help you cope with street life? What do you do to protect yourself or keep yourself safe?); (c) substance use (e.g., In what ways have you or friends you know used substances to deal with feelings related to experiences that make you feel distressed or upset? How do you think using substances helps with these feelings? How can it be unhelpful or harmful?); and (d) transience (e.g., What are the benefits and risks involved in traveling? Describe an example of a time in which you or people you’ve known left town to escape a dangerous situation. How about when you or they have left town for an opportunity?). All interviews were audio recorded and subsequently transcribed.

Quantitative Measures

Dependent Variable

Transience was measured as the total number of times the emerging adult had moved between cities since leaving home for the first time. Responses were quantified by counting the number of cities (new or repeated) to which the emerging adult had moved since he/she first left home. Given the large skewness of the original variable (10.96), a log transformation was performed (Chin & Lee, 2008).

Predictors of Transience

Demographic information included age, gender (0 = female, 1 = male), ethnicity (0 = other, 1 = White), and education (0 = dropped out, suspended, or still enrolled or 1 = high school degree or General Educational Development [GED]).

Risk factors included various indicators. Two homelessness variables were used. Primary residence over the past 6 months was measured as 0 = homeless (i.e., homeless or in a temporary shelter) or 1 = housed (i.e., with family, relatives, or in foster care, jail, youth detention, or residential housing). Length of time homeless was determined by subtracting the number of months since the participant had left home for good from the date of the interview. Criminal history was comprised of four variables that measured whether they had ever been in juvenile detention or arrested (0 = no, 1 = yes) as well as the number of times for each. The informal income variable was measured by asking whether they had received income in the past 6 months from informal sources (i.e., selling self-made items, your possessions, bottles/cans, or your blood; dealing drugs; panhandling, trading sexual favors, gambling, or stealing; coded 0 = no or 1 = yes).

The two victimization variables were derived from the Childhood Trauma Questionnaire (CTQ; Bernstein & Fink, 1998), a 28-item retrospective, self-report questionnaire designed to assess types of negative childhood experiences. This instrument has demonstrated reliability and validity, including test-retest reliability coefficients ranging from .79 to .86. Physical abuse included five CTQ items such as “I was punished with a belt, a board, a cord, or some other hard object.” Emotional abuse included five CTQ items such as “People in my family said hurtful or insulting things to me.” Participants rated the truth of each statement on a 1 to 5 scale, from 1 = never true when they were growing up to 5 = very often true when they were growing up. For both the physical and emotional abuse variables, whether or not participants had experienced any of the five types of abuse was dichotomized as 0 = never true (no) and 1 = all else (yes). Dichotomized items were added so that the physical and emotional abuse variables represent the total number of times participants reported experiencing that type of abuse (range 0-5).

The one substance abuse variable was determined using the Mini-International Neuropsychiatric Interview (MINI; Sheehan et al., 1998). The MINI asks dichotomous (no/yes) screening and symptom questions for drug abuse/dependence. Affirmative answers to screening questions and a sufficient number of positive responses to symptom questions results in meeting criteria for diagnosis. Alcohol and/or drug addiction was conceptualized as meeting the criteria for either abuse or dependence and was coded as 0 = does not meet criteria, 1 = meets criteria for either alcohol or drug addiction, and 2 = meets the criteria for both alcohol and drug addiction.

There were five resilience characteristics measured by the Resilience Scale (Wagnild & Young, 1993). This 25-item scale identifies five characteristics of resilience: (a) self-reliance (i.e., belief in oneself and one’s capabilities); (b) meaning (i.e., realization that life has a purpose); (c) equanimity (i.e., balanced perspective of one’s life and experiences); (d) perseverance (i.e., act of persistence despite adversity or discouragement); and (e) existential aloneness (i.e., realization that although some life experiences are shared, some are completely unique and must be dealt with alone). Participants rated their feelings on a scale, from 1 (strongly disagree) to 7 (strongly agree). The Resilience Scale has shown good-to-strong reliability with α coefficients ranging from 0.72 to 0.94 across a review of completed studies with a variety of individuals from diverse socioeconomic groups, ages, and educational backgrounds (Wagnild, 2009), including strong internal consistency (α = 0.91) in a study with homeless adolescents (Rew et al., 2001).

Quantitative Data Analysis

Global empirical analyses including means, and standard deviations were used to describe emerging adults’ demographics and characteristics. All variables had fewer than 2% missing data. Chi-square tests were used to examine similarities and differences in transience across cities. Review of the variance inflation factors (VIFs) revealed that multicollinearity between independent variables would not significantly affect analyses. All VIF statistics were below 2.52.

Ordinary least squares (OLS) regression was conducted in SPSS to analyze associations between predictors and transience across three cities. Two dummy variables reflecting three cities were included to control for city differences. It is not realistic to isolate the hypothesized variables in the analyses, since risk and resilience factors among homeless emerging adults interrelate in real-world settings to influence transience. Rather, it is more appropriate to conceptualize these variables as cumulative influences on emerging adults’ behavior. Thus, demographic, risk, and resilience variables were entered separately in blocks using seven hierarchical regression models: (a) demographic (age, gender, ethnicity, and education) and city dummy variables; (b) Block 1 + two homelessness variables; (c) Blocks 1-2 + four criminal justice variables; (d) Blocks 1-3 + one informal income variable; (e) Blocks 1-4 + two victimization variables; (f) Blocks 1-5 + one substance abuse variable; (g) Blocks 1-6 + five resilience variables. In this way, the final model included all predictor variables, highlighting the effects for each category of risk and resilience, as well as their cumulative effects on transience.

Qualitative Measures

To develop the qualitative interview guide, investigators across three cities reviewed the literature on homeless emerging adults’ informal (i.e., self-protection and coping) and formal strategies (i.e., service use) to prevent and manage trauma, substance use, and transience. Using an iterative process, investigators drafted and circulated questions in each of these categories for review. Once the final interview guide was established, investigators trained their respective research assistants by conducting initial qualitative interviews during which the assistants observed. Subsequently, the investigators observed their respective assistants conducting one or two interviews. The qualitative interviewing technique described by Patton (1990) was used in which the interview guide served as a flexible outline of topics. Investigators instructed the research assistants to use the interview guide loosely with participants and to encourage them to speak openly and at length about topics of their choice related to the study categories.

Qualitative Data Analysis

Data were collected and analyzed inductively using grounded theory and following constructivist assumptions, recognizing limitations in knowing beforehand the multiple realities that would emerge from the in-depth interviews (Rodwell, 1998). In generating grounded theory, Glaser and Strauss (1967) emphasize that data should be concurrently collected, coded, and analyzed. Because the in-depth interviews were conducted over the 16-month study, this approach was useful for refining and reshaping the categories identified in earlier discussions by collecting additional data from participants in subsequent interviews. For instance, emerging adults commonly noted their preferred forms of transience, including riding trains, hitchhiking, rubber tramping (i.e., traveling and living out of a vehicle, such as a van, or car), and traveling with large groups of peers. As these topics inductively emerged from the interviews, researchers asked follow-up questions on the participant’s experiences using identified forms of travel.

The constant comparative method was used to analyze participants’ transcripts from the in-depth interviews (Glaser & Strauss, 1967). Data coding was initiated by the lead investigator in one of the three cities using the Maxqda software (Kuckartz, 2001). First, to generate initial, low-inference codes (i.e., open codes), the investigator identified key words from the interview guide and participants’ transcripts, including in vivo language used by the participants themselves (Strauss & Corbin, 1998). Next, the investigator adopted a high-inference coding process (i.e., axial coding) in which similar codes were grouped into broader categories. Within each primary code, subcategories were created by using codes from the original list, breaking down complex codes into subcategories. Finally, the investigator selected central or core coding categories (i.e., selective coding), which were systematically related to other categories and validated as unique. Memoing (i.e., personal note writing) was used during this final coding stage to summarize and elaborate on the categories among codes (Lofland & Lofland, 1995). The lead investigator developed a full coding template for transience and discussed with the other two researchers all codes, categories and conclusions that emerged from the analysis. Minor revisions to the coding template resulted from ongoing consultation among the full research team.

Results

Quantitative Findings

Table 1 presents the demographic, risk and resilience characteristics of the full sample.

Table 1.

Full and City-Level Sample Characteristics of Homeless Emerging Adults.

	Full sample (N = 601)		Los Angeles (n = 200)		Denver (n = 201)		Austin (n = 200)		χ²/ANOVA
Variables	N	%	n	%	n	%	n	%
Gender^a
Male	385	64	128	64	129	64	128	64
Female	216	36	72	36	72	36	72	36
Race/ethnicity (n = 600)									χ² = 207.9***
White	240	40	18	9	70	35	152	76
Black	152	25	92	46	54	27	6	3
Latino	107	18	52	26	37	18	18	9
Other	101	17	38	19	39	20	24	12
Primary residence (past 6 months)									χ² = 19.1***
Homeless	303	50	87	44	90	45	126	63
Housed	298	50	113	56	111	55	74	37
Highest education level									χ² = 57.7***
Quit	143	24	47	23	31	15	65	32
Enrolled: High school or college	114	19	36	18	63	31	15	8
General Educational Development	114	19	17	9	55	27	42	21
Graduated high school	172	29	57	29	49	25	66	33
Other	58	9	43	21	3	2	12	6
Criminal-justice history (ever)
Juvenile detention	213	35	57	29	79	39	77	39	χ² = 6.3*
Arrested	419	70	100	50	157	78	162	81	χ² = 55.6***
Informal income sources^b	455	76	125	63	144	72	186	93	χ² = 53.3***
Meet diagnostic criteria for
Drug addiction	359	60	92	46	118	59	149	75	χ² = 39.3***
Alcohol addiction	299	50	81	41	78	39	140	70	χ² = 50.3***
Both drug & alcohol	248	41	64	32	64	32	120	60	χ² = 70.0***
	Mean	SD	Mean	SD	Mean	SD	Mean	SD	ANOVA^d
Transience (# intercity moves)	3.5	3.7	2.6	3.2	1.9	2.8	5.9	3.7	F = 85.3***
Age	20.5	1.6	19.4	1.1	19.9	1.6	20.8	1.7	F = 40.6***
Length of time homeless (months)^c	32.4	31.0	22.9	26.0	31.6	29.6	42.8	33.9	F = 22.1***
Victimization
Physical abuse	2.5	1.8	2.4	1.9	2.6	1.8	2.4	1.8
Emotional abuse	3.6	1.8	3.2	1.8	3.4	1.7	3.5	1.7
Resiliency
Self-reliance	28.3	4.5	28.0	4.6	28.4	4.3	28.4	4.5
Meaning	27.4	4.6	28.1	4.2	27.6	4.6	26.4	4.8	F = 7.4**
Equanimity	27.5	4.6	27.3	4.4	27.7	4.8	27.4	4.6
Perseverance	27.7	4.6	28.2	4.7	28.1	4.5	26.8	4.6	F = 5.6**
Existential aloneness	29.2	4.3	29.4	4.2	29.3	4.5	28.9	4.2

p < .05. **p < .01. ***p < .001.

N = 601 for all variables unless indicated. ^bIn the past 6 months, did you get any money or resources from informal sources? (i.e., selling self-made items, your possessions, bottles/cans, or your blood; dealing drugs; panhandling, trading sexual favors, gambling, or stealing). ^cLength of time homeless (in months) was calculated as the date of interview minus the date on which emerging adult left home for good. ^dPost hoc analyses of ANOVA tests indicate Austin emerging adults significantly differed in transience, age, and months homeless compared to those in Los Angeles and Denver.

Multivariate Analyses of Risk and Resilience Factors Predicting Transience

Table 2 displays the seven OLS regression models after all demographic, risk and resilience factors were entered separately. Each subsequent model, including all variables from previous models, the location to control for city-level differences, and one additional set of risk (or resilience) factors, was a significant improvement over previous models. The seventh model with all predictor variables was a significant improvement over previous models (ΔF [5, 552] = 3.070, ΔR² = 0.016, p < .05, Adjusted R² = 0.386). The improved change statistics indicate support for the cumulative effects of risk and resilience factors on transience. In effect, the full model of determinants was significantly associated with homeless emerging adults’ transience (F = 18.128, p < .001).

Table 2.

Ordinary Least Squares Regressions of Predictors of Transience from Hierarchical Regressions.

	Step 1			Step 2			Step 3			Step 4
Variables	B (Beta)^a	SE	t	B (Beta)^a	SE	t	B (Beta)^a	SE	t	B (Beta)^a	SE	t
Demographic factors
Age	.02 (.06)	.02	1.50	−.03 (−.07)	.02	−1.58	−.03 (−.08)⁺	.02	−1.94	−.03 (−.06)	.02	−1.56
Gender^b	.08 (.06)⁺	.05	1.65	.06 (.05)	.05	1.32	.02 (.01)	.05	0.41	.01 (.01)	.05	0.31
Ethnicity^c	.20 (.15)***	.06	3.46	.20 (.15)***	.06	3.56	.20 (.15)***	.06	3.66	.20 (.15)***	.06	3.56
High school degree^d	.06 (.05)	.05	1.23	.07 (.05)⁺	.05	1.54	.09 (.07)*	.05	1.94	.08 (.06)*	.05	1.72
Austin dummy^e	.37 (.26)***	.07	4.99	.30 (.21)***	.07	4.22	.21 (.15)**	.07	3.01	.17 (.12)*	.07	2.36
Denver dummy^e	−.29 (−.21)***	.06	−5.12	−.30 (−.22)***	.06	−5.27	−.37 (−.27)***	.06	−6.46	−.38 (−.28)***	.06	−6.66
Risk (homelessness) factors
Primary residence^f				−.18 (−.14)***	.05	−3.88	−.17 (−.13)***	.05	−3.79	−.16 (−.12)***	.04	−3.59
Length of time homeless (months)				.01 (.27)***	.00	6.84	.01 (.24)***	.00	5.79	.01 (.22)***	.00	5.53
Risk (criminal history) factors
History of juvenile detention							−.10 (−.08)*	.06	−1.83	−.10 (−.07)*	.06	−1.71
Number of detentions							.01 (.07)*	.00	1.77	.01 (.07)*	.00	1.89
History of arrest							.19 (.14)***	.06	3.29	.17 (.12)**	.06	2.94
Number of arrests							.01 (.10)**	.00	2.43	.01 (.09)**	.00	2.36
Risk (employment) factors
Informal sources of income										.18 (.12)***	.05	3.34
Change-score statistics	ΔR² = 0.255, ΔF [6, 567] = 32.32, p < .001, Adjusted R² = 0.247			ΔR² = 0.072, ΔF [2, 565] = 30.07, p < .001, Adjusted R² = 0.317			ΔR² = 0.029, ΔF [4, 561] = 6.40, p < .001, Adjusted R² = 0.342			ΔR² = 0.013, ΔF [1, 560] = 11.15, p < .01, Adjusted R² = 0.354
	Step 5			Step 6			Step 7
Variables	B (Beta)	SE	t	B (Beta)	SE	t	B (Beta)	SE	t
Demographic factors
Age	−.02 (−.06)	.02	−1.39	−.02 (−.06)	.02	−1.36	−.02 (−.06)	.02	−1.47
Gender	−.00 (−.00)	.05	−0.06	−.01 (−.01)	.05	−0.27	.00 (.00)	.05	−0.01
Ethnicity	.21 (.16)***	.06	3.84	.21 (.16)***	.05	3.78	.20 (.15)***	.05	3.72
High school degree	.08 (.06)*	.04	1.83	.08 (.06)*	.04	1.86	.08 (.06)*	.04	1.91
Austin dummy	.18 (.13)*	.07	2.57	.16 (.11)*	.07	2.19	.16 (.11)*	.07	2.21
Denver dummy	−.39 (−.28)***	.06	−6.79	−.38 (−.28)***	.06	−6.76	−.38 (−.28)***	.06	−6.74
Risk (homelessness) factors
Primary residence	−.15 (−.12)***	.04	−3.42	−.15 (−.11)***	.04	−3.32	−.15 (−.12)***	.04	−3.52
Length of time homeless (months)	.00 (.20)***	.00	4.99	.00 (.21)***	.00	5.17	.01 (.22)***	.00	5.48
Risk (criminal history) factors
History of juvenile detention	−.10 (−.07)*	.06	−1.82	−.10 (−.07)*	.06	−1.81	−.09 (−.07)⁺	.06	−1.62
Number of detentions	.01 (.08)*	.00	2.09	.01 (.08)*	.00	2.09	.01 (.08)*	.00	2.16
History of arrest	.17 (.12)**	.06	3.04	.15 (.11)**	.06	2.63	.15 (.11)**	.06	2.63
Number of arrests	.00 (.08)*	.00	1.92	.00 (.07)*	.00	1.68	.00 (.06)⁺	.00	1.53
Risk (income) factor
Informal sources of income	.19 (.12)***	.06	3.40	.15 (.10)**	.06	2.76	.16 (.11)**	.06	2.84
Risk (victimization) factors
Physical abuse	.06 (.16)***	.02	3.55	.05 (.15)***	.02	3.45	.05 (.15)***	.02	3.39
Emotional abuse	−.05 (−.13)**	.02	−2.83	−.05 (−.14)**	.02	−3.04	−.05 (−.13)**	.02	−2.94
Risk (substance abuse) factor
Alcohol and/or drug addiction				.08 (.10)**	.03	2.76	.09 (.11)**	.03	2.98
Resilience factors
Self-reliance							.02 (.13)**	.01	2.52
Meaning							.01 (.08)*	.01	1.66
Equanimity							−.02 (−.11)*	.01	−2.24
Perseverance							.00 (.01)	.01	0.13
Existential aloneness							−.01 (−.08)*	.01	−1.70
Change-score statistics	ΔR² = 0.015, ΔF [2, 558] = 6.75, p < .01, Adjusted R² = 0.367			ΔR² = 0.008, ΔF [1, 557] = 7.60, p < .01, Adjusted R² = 0.374			ΔR² = 0.016, ΔF [5, 552] = 3.07, p < .05, Adjusted R² = 0.386

Note: ⁺p < .10. *p < .05. **p < .01. ***p < .001.

Standardized coefficients (Betas) are in parentheses. ^bAge, gender, and two city control variables used significance 0.05 two-tailed test; all other predictors used significance 0.05 one-tailed test. ^c0 = Black/Latino/Other; 1 = White. ^d0 = dropped out, suspended, or still enrolled in school; 1 = high school degree or General Educational Development. ^eFor Austin dummy variable: 0 = Los Angeles and Denver; 1 = Austin. For Denver dummy variable: 0 = Los Angeles and Austin; 1 = Denver). ^f0 = homeless (i.e., homeless or in a temporary shelter); 1 = housed (i.e., with family or relatives, or in foster care, jail, youth detention, or residential housing) in the 6 months prior to the interview.

Demographic Factors

White emerging adults reported greater transience in comparison to their minority peers to (β = 0.15, p < .001). Those who had their high school or GED degree also reported greater transience than those with lower educational levels (β = 0.06, p < .05).

Risk Factors

Emerging adults who were homeless or in temporary shelters reported greater transience than their housed peers (β = −0.12, p < .001). Those who had been on the streets for a longer period of time also reported greater transience (β = 0.22, p < .001). Homeless emerging adults who had a greater number of juvenile-detention episodes (β = 0.08, p < .05) and who had been arrested (β = 0.11, p < .01) reported greater levels of transience. Earning an income through informal sources also was significantly associated with increased transience (β = 0.11, p < .01). Furthermore, those who had experienced greater episodes of physical abuse prior to leaving home reported higher transience (β = 0.15, p < .001); however, greater occurrences of emotional abuse predicted lower transience (β = −0.13, p < .01). Lastly, homeless emerging adults who met the criteria for both alcohol and drug dependency reported greater transience than their peers who met the criteria for only one addiction as well as those who had no addiction (β = 0.11, p < .01).

Resilience Factors

Homeless emerging adults who scored higher on the subscales of self-reliance (β = 0.13, p < .01) and meaning (β = 0.08, p < .05) reported greater transience. Conversely, scoring lower on the subscales of equanimity (β = −0.11, p < .05) and existential aloneness (β = −0.08, p < .05) was associated with greater transience.

Qualitative Findings

Participants

The 145 participants in the qualitative interviews were slightly younger than the full sample (M = 19.9 years, SD = 1.4). Roughly the same proportion was male (70%). Whites comprised 32% of the qualitative sample (vs. 40% of the full sample), Blacks 31% (vs. 25% of the full sample), and Latinos 23% (vs. 18% of the full sample). Qualitative participants had been away from home for about 2 fewer months (M = 30.8, SD = 28.8) and reported slightly fewer intercity moves than the full sample (M = 3.0, SD = 3.5). A greater proportion (61%) reported being homeless or in a temporary shelter in the prior 6 months (vs. 50% of the full sample).

In the qualitative interviews, common reasons for transience were similar to the risk and resilience factors included in quantitative analyses, including (a) education, (b) homelessness history, (c) criminal/juvenile justice history, (d) income generation, (e) victimization, (f) substance abuse, and (g) resiliency. The full coding template can be obtained from the lead author; yet the findings below are limited to those most closely linked to the quantitative results.

Education

Quantitative findings suggest that emerging adults with higher educational levels were more geographically mobile. Qualitative interviews described moving to new cities to pursue school and educational opportunities, such as GED programs, community colleges, and university extension programs. One emerging adult described the experience as:

I left Miami to go to New York because I felt like there was more opportunity in New York. So I went back, went to Job Corps, got my GED, graduated from Job Corps, and from there I went to Boston because I heard what they, um, have a university. They have an extension program. So I went to Boston.

Homelessness History

Quantitative findings reveal that participants who were homeless or in a temporary shelter reported greater transience than their more stably housed peers. Their testimonies suggest that they moved frequently between cities to locate safe places to sleep. Some could not afford local rent payments; others were expelled from shelters or housing programs and needed to find a safe place to sleep. They used various modes of transportation as safe places to sleep, such as riding trains or buses at night in lieu of sleeping in shelters or on the streets. They also described transience associated with their caregivers’ homelessness, instability, or death. For example, several noted:

I left town because I had nowhere to sleep. Then I was tired of sleeping on the same spot all the time. So I just went to go out and explore, you know?

I lived with my grandparents from when I was 4 until when I was 10. And then when she passed away, that was when I literally started being on my own.

Criminal/Juvenile Justice History

History of criminal-justice involvement was a predictor of transience in the quantitative findings. Qualitative testimonies suggest that emerging adults moved frequently to escape legal problems. One individual said: “I had to flee town a couple of times because the cops were looking for me. I just got drunk and was just being stupid. Shit I shouldn’t be doing.” Others moved to avoid consequences related to their prior gang activity or violent acts. Another participant shared:

I got involved in pretty much gang warfare and I wasn’t supposed to be involved, but I ended up getting in the middle of it and ended up taking sides and I ended up hearing rumors going all around town from people I don’t even know that maybe that one of the gangs wanted me because of the side I helped and so I had to get out of town so I didn’t take any harm to myself or I would have died.

Income Generation

Quantitative findings indicate that emerging adults who earned an income through informal sources reported greater transience. Qualitative testimonies suggest that their mobility was often a result of their pursuit of income—both formal and informal sources—as well as their inability to locate formal employment. One noted: “We stay where it’s convenient, where we can make money, and where we can cop our dope.” Another reported:

I stayed at my friend’s house. I tried to work things out over there, trying to find a job and couldn’t find one. I was in Newark and just couldn’t find a job up there. So I came back, moved to the Valley for like a week and stayed at my brother’s.

Victimization

Physical abuse was a significant predictor of transience in the quantitative analysis. In the qualitative interviews, emerging adults described their transience resulting from experiencing physical abuse in their intimate relationships as well as in their neighborhoods for their sexual orientation. They commonly reported growing up in abusive biological and foster families until they could no longer handle the abuse. They noted:

I left because of my step-, my dad, my stepdad, my dad. He used to beat me. I couldn’t do it no more. I left and I left town. I got on the Greyhound bus and I left. I, I left. I couldn’t handle it anymore.

I was always scared, you know. She’d abuse. I used to be like, it was like, it was like almost like a routine every day, you know. She’d abuse me. I guess somehow I just took the pain and just moved on the next day . . . and there was nothing I could do. I had no family to run to. My brothers couldn’t do anything . . . That’s why I ran away sometimes, you know.

Substance Abuse

Substance dependency was positively associated with transience in the quantitative analysis. In the qualitative interviews, emerging adults described their drug use as a motivator for transience. In some cases, they escaped a particular city to evade danger brought on by their drug habits. In other cases, they recognized that they could not become and/or stay sober in a particular city due to the availability of drugs and proximity to substance-using peers. In these cases, homeless emerging adults moved to other cities in an effort to end their drug use. One emerging adult noted that “if I would have stayed, I would have ended up just like everybody else, until I was dead.” Another stated:

Just leaving town because you consistently use too much drugs and, if they are too available for you in town, sometimes it is in your best interest just to go somewhere else where you don’t know where to find them.

Resiliency

Qualitative testimonies provide initial support for the relationship between resiliency and transience. For instance, those who were independent, self-reliant, and adventurous described how their mobility was beneficial in helping them develop intellectually and emotionally. Consistent with quantitative findings, transient emerging adults demonstrated strong beliefs in their own capabilities, recognized their own strengths, and were highly self-dependent. One homeless emerging adult described himself in this way:

I’ve basically been all over the United States. So, I feel that it taught me a lot. Like, I’m more smart and if I do end up like, having to be on my own, I know what I need to do. I know how to use buses; I know how to use Greyhounds; I know how to use planes; I know how to use trains; I know how to hitchhike, you know, all of that. So . . . I just benefit from that. It’s better to explore. The more you stay in your little shelter, the more that your skin’s gonna dry.

Emerging adults also noted the importance of moving away from homelessness and toward personal and professional goals and opportunities to start anew. As revealed in the quantitative findings, highly transient emerging adults believed that their lives had a purpose, valued their own life contributions, and possessed a strong sense of the meaning of life. As one participant remarked, “I just have the motivation and courage and desire to keep moving forward, you know. I don’t let nothing stop me. I don’t let nothing hold me back from what I’m trying to do and what I’m trying to succeed in life, you know. I just keep it movin’.” Another said:

It would probably be tying me down too much to stay here. I feel like life on the road is right. The people that I meet and the people that I impact are you know, it was meant to be. If I’m stationary, I don’t have that option of meeting some kid that’s having a hard time. So if I feel like I’m supposed to meet somebody in Colorado, and I’m staying here because I want to help the same kids over and over, and they’re gonna hear what I have to say over and over and over again, it’s going to get old to them. And the person that needs to hear it isn’t going to hear it.

Negative associations between equanimity and transience in the quantitative findings are also illustrated in the qualitative testimonies. Highly transient emerging adults were those who expressed having a lower tolerance for the adversities of life on the streets and day-to-day annoyances. For instance, one noted:

When it gets really distressing every day constantly, constantly, every day, just like people just do stupid stuff just to antagonize you and bug you and it’s just like you get really tired of it and just you don’t know what to do and you’re just like, ‘(sigh) I gotta get out of here.’ So you just get away for a while and then you go back.

Likewise, qualitative reports support quantitative findings that those who had lower levels of existential aloneness reported greater transience, as some described searching for their unique path in life or being unable to face many of life’s challenges alone. For example, one stated: “I’ve been back and forth, back and forth, like all over, not all over the country but just moving everywhere, just bouncing trying to find the right place to be.” Another noted: “I was just avoiding my own problems because I was, you know, a runaway. I didn’t wanna be here, didn’t want to deal with that. And so, we hopped in the car and we went.”

Discussion

Findings provide greater understanding of the risk and resilience behaviors of traveling homeless emerging adults as well as their motivations for traveling. Quantitative results suggest a profile of their risk and resilience characteristics. The qualitative interviews largely supported these characteristics, yet raise new questions regarding how transience functions as both a maladaptive and adaptive behavior, depending upon emerging adults’ motivations for travel. The use of the risk and resilience framework in this study had conceptual relevance since previous research typically has viewed transience to be a problem (i.e., indicating instability and associated with other negative outcomes). However, our findings that transience can be both a problem and a positive solution bring into question whether the risk and resilience framework is overly simplistic for understanding transience as a behavior among homeless emerging adults.

In the event that transience is considered to be maladaptive and something to be prevented, it is useful to understand the risk factors identified in this study. Namely, transient homeless emerging adults are more likely to be White (Sanders et al., 2008), homeless or in a temporary shelter (vs. in a longer-term residential shelter) in the 6 months prior to the interview (Ferguson et al., 2010), and to earn income through informal—often illegal—means (Hahn et al., 2008; Sanders et al., 2008). They are also more likely to have a criminal-justice history (Hahn et al., 2008), to have been abused/victimized (Mallett et al., 2004), to be addicted to a variety of substances (Sanders et al., 2008), and to be highly self-reliant (Rew et al., 2001). In an effort to situate study participants’ experiences within emerging adulthood as a developmental period, this explanation is consistent with viewing emerging adulthood for homeless young people as an age of instability (Arnett, 2004). Developmentally, some homeless emerging adults may experience further delays due to cognitive limitations and mental illness (Cauce et al., 2000). Faced with instability, indecision and a host of interacting risk factors, making decisions during this period to lay the foundation for their adult lives may be outweighed by a greater need for daily survival (as in living on the streets or engaging in survival behaviors), meeting their basic needs (as in committing illegal activity), and coping with the adversity of street life (as in using substances).

In contrast, if transience is considered to be an adaptive behavior enabling emerging adults to ensure their personal safety and pursue their life goals, then the resilience characteristics identified here offer novel descriptions of their strengths. Rew et al. (2001) suggest that homeless youth may develop characteristics of resilience (e.g., self-reliance, independence, self-care) in response to their limited formal adult and institutional supports. Many homeless young people prefer to rely on themselves and their peers, given their history of frequent negative interactions with adults and distrust of institutions (De Rosa et al., 1999). If indeed traveling is a strategy used by homeless emerging adults to cope with or adapt to environmental adversity and stress, then the most transient homeless young people may be those who are highly self-reliant, seek meaning and opportunity beyond their present situations, demonstrate low equanimity in dealing with day-to-day challenges, and struggle to face and to deal with life’s challenging experiences alone. This explanation is consistent with viewing emerging adulthood as a developmental period marked by identity exploration and open possibilities (Arnett, 2004). During this period, homeless emerging adults with high levels of resiliency, who have been independent from their families and institutions for some time, may take advantage of their freedom to explore their identity (as in traveling around the country to enhance their personal experiences and knowledge). Likewise, they may travel in search of opportunities to transform their lives (as in relocating to a new city to achieve/maintain sobriety, or to pursue a job or educational degree).

Resilience characteristics of traveling homeless emerging adults, which are largely missing from extant literature, may contribute to a more complete understanding of why this population chooses to run, and what long-term outcomes are associated with transience. To date, it remains unclear whether they use transience to avoid life challenges and their own responsibilities (e.g., warrants for their arrest), which may lead to negative health, mental health and functional outcomes. Quantitative and qualitative findings here suggest that emerging adults with lower reported equanimity and existential aloneness were those who were less able to cope with life adversity and used transience to avoid day-to-day annoyances and their own perceived problems. Alternatively, traveling homeless emerging adults may use transience to escape from abusive environments (e.g., family or intimate partners), which may lead to more positive health, mental health and functional outcomes. Qualitative narratives reveal that emerging adults frequently noted that their transience was motivated by a need to be free from violent and abusive relationships, as well as local drug cultures that interfered with them becoming and staying sober. An additional speculation is that traveling homeless emerging adults may use transience to move toward greater meaning or purpose in their lives (e.g., work and education), which may lead to more positive health, mental health and functional outcomes. Findings suggest that emerging adults with higher levels of self-reliance and life meaning used transience to help themselves grow intellectually and emotionally, as well as to pursue desired personal and professional goals. Future study of the relationships among transience, resilience, and long-term outcomes in homeless emerging adults will help elucidate whether they are running from their responsibilities, from dangerous environments, and/or toward greater opportunities.

Limitations

This study had several limitations. First, although the quantitative data collection was similar across cities, the samples may not be representative of the population in any of the cities. Because homeless emerging adults are transient and difficult to locate, probability sampling is often not feasible, and purposive sampling in street locations and agencies is commonly used in empirical investigations (Clatts, Davis, & Atillasoy, 1995). Recruiting a purposive sample of service-seeking study participants likely increased their motivation to participate. Findings thus may not be generalizable (or transferable) to other nonservice-using homeless emerging adults. Furthermore, the choice of the three cities was based on feasibility, rather than representativeness of the types of homeless emerging adults. Thus, it is impossible to determine whether other cities replicate these findings, as the city-level differences detected here may be due to samples collected from agencies serving homeless emerging adults in smaller versus larger cities.

Additionally, the targeted age range of our participants (18-24 years old) and the cross-sectional nature of this study limit our ability to draw conclusions on the trajectory of transience during the larger developmental period of emerging adulthood (late teens through the mid to late 20s). We are unable to determine whether transience is concentrated among emerging adults and whether this population begins to desire greater stability as they enter the subsequent period of adulthood. We are also unable to determine the outcomes associated with homeless emerging adults’ motivations to travel given that data collection occurred at one time point only. Subsequent studies using a longitudinal design with multiple follow-ups among a broader age range of homeless emerging adults will elucidate some of these unanswered questions.

Furthermore, the transience and high-risk behavior variables used in the analyses were limited in several important ways. The complex histories of transience present a challenge to any study using recall-based data. Although the development and use of the study instrument represented an attempt to maximize the reliability and validity of geographic mobility data, it is clear that recall-based information will always represent a limitation. It is equally possible that responses to our questions about stigmatized risk behaviors (e.g., prostitution, survival sex, drug use, criminal activity) may have been underreported due to social desirability. Despite these limitations, our findings on the predictors and descriptions of transience facilitate greater understanding of the risk and resilience characteristics of traveling homeless emerging adults.

Implications for Research and Practice

Identifying the risk and resilience predictors of transience in homeless emerging adults as well as understanding their motivations for why they travel will facilitate customizing interventions that prevent geographic mobility when transience is considered a maladaptive behavior, such as when homeless emerging adults are escaping legal consequences associated with their criminal activity. In contrast, when transience is used as an adaptive behavior or coping mechanism to pursue personal safety and broader life opportunities, interventions are needed that make travel safer—and services more accessible—for this population on the road.

To address the interrelated risk and resilience factors associated with transience, program and policy responses need to depart from knowledge of who traveling homeless emerging adults are and why they travel. Namely, transitional-living and supportive-housing programs are needed that recognize the coping mechanisms (e.g., drug use) and survival strategies (e.g., theft, survival sex, and drug dealing) of traveling emerging adults and adopt a harm-reduction (vs. abstinence) approach to help them gradually diminish their high-risk behaviors (Ferguson et al., 2011). These supportive environments will reach the broader population of traveling emerging adults who are ready to transition off the streets, yet who are restricted from receiving support in existing housing programs with stringent eligibility criteria (Hyde, 2005). Likewise, to effectively engage and retain this population in services, providers should consider how emerging adults’ transience can be a resource for spreading agencies’ prevention and service-use messages to other travelers via informal peer networks. In this way, travelers’ strengths, skills, and resiliency are the starting point of services, not their needs, risks, or deficits (Hyde, 2005).

Lastly, for those who continue traveling, service providers and researchers should consider adopting innovative technologies (e.g., cell phone, texting, email, social media, electronic debit cards) as modalities for providing mobile services, treatments, and monetary incentives and reimbursements to traveling emerging adults (Bender et al., 2010; Des Jarlais et al., 2005). Furthermore, establishing intercity and interstate “migrant” networks of youth-serving agencies and providers in the locations where they are most likely to frequent will help them receive services on the road (Hahn et al., 2008). Each of these efforts can help enhance access, continuity, and adherence to treatment, resources, and services for traveling homeless emerging adults.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Funding for this study was provided in Los Angeles by the University of Southern California (USC), School of Social Work Hamovitch Research Center; in Denver by the University of Denver, Graduate School of Social Work; and in Austin by a Faculty Development Grant from the University of Texas at Austin and the Center for Social Work Research. We would like to acknowledge Connie Chung from the Graduate School of Education at Harvard University, Kimberly Biddle and Jina Sang from the USC School of Social Work, Jamie Yoder and Chelsea Komlo from the University of Denver, and Tiffany Ryan, Katherine Montgomery, and Angie Lippman from the University of Texas at Austin for their involvement in the study as research assistants.

Author Biographies

Kristin M. Ferguson, Ph.D., is an Associate Professor at the CUNY Silberman School of Social Work. Her research interests include homeless and street-living youth, social capital, and criminal behavior among street youth. Dr. Ferguson’s recent work is related to testing the effectiveness of vocational interventions for street youth.

Kimberly Bender, Ph.D., is an Assistant Professor at the University of Denver, Graduate School of Social Work, where she studies youth problem behavior with a particular focus on victimization and related mental health problems among homeless youth.

Sanna J. Thompson, Ph.D., is an Associate Professor at the University of Texas at Austin, School of Social Work. Her research focuses on high-risk youth and families, with special emphasis on runaway/homeless youth populations. She has conducted several externally funded studies of homeless youth focused on substance abuse and trauma.

References

Arnett

J. J.

(2004). Emerging adulthood: The winding road from the late teens through the twenties. Cary, NC: Oxford University Press.

Arnett

J. J.

(2007). Suffering, selfish, slackers? Myths and reality about emerging adults. Journal of Youth and Adolescence, 36, 23-29. doi:10.1007/s10964-006-9157-z

Bender

Ferguson

Thompson

Komlo

Pollio

(2010). Factors associated with trauma and posttraumatic stress disorder among homeless youth: The importance of transience. Journal of Traumatic Stress, 23(1), 161-168. doi:10.1093/swr/33.4.219

Bender

Thompson

S. J.

McManus

H. H.

Lantry

Flynn

P. M.

(2007). Capacity for survival: Exploring strengths of homeless street youth. Child and Youth Care Forum, 36, 25-42. doi:10.1007/s10566-006-9029-4

Bernstein

D. P.

Fink

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

Bridgman

(2001). I helped build that: A demonstration employment training program for homeless youth in Toronto, Canada. American Anthropologist, 103, 779-795. doi:10.1525/aa.2001.103.3.779

Cauce

A. M.

Paradise

Ginzler

J. A.

Embry

Morgan

C. J.

Lohr

Theofelis

(2000). The characteristics and mental health of homeless adolescents: Age and gender differences. Journal of Emotional and Behavioral Disorders, 8, 230-239. doi:10.1177/106342660000800403

Chin

Lee

B. Y.

(2008). Principles and practice of clinical trial medicine. San Diego, CA: Elsevier.

Clatts

M. C.

Davis

W. R.

Atillasoy

(1995). Hitting a moving target: The use of ethnographic methods in the development of sampling strategies for the evaluation of AIDS outreach programs for homeless youth in New York City. In Lambert

E. Y.

Ashery

R. S.

Needle

R. H.

(Eds.), Qualitative methods in drug abuse and HIV research (pp. 117-135). Rockville, MD: U.S. Department of Health and Human Services.

10.

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. doi:10.1016/S1054-139X(99)00040-3

11.

Des Jarlais

D. C.

Perlis

T. E.

Settembrino

J. M.

(2005). The use of electronic debit cards in longitudinal data collection with geographically mobile drug users. Drug and Alcohol Dependence, 77, 1-5. doi:10.1016/j.drugalcdep.2004.06.010

12.

Ferguson

Bender

Thompson

Xie

Pollio

(2011). Correlates of street-survival behaviors in homeless young adults in four U.S. cities. American Journal of Orthopsychiatry, 81, 401-409. doi:10.1111/j.1939-0025.2011.01108.x

13.

Ferguson

Jun

Bender

Thompson

Pollio

(2010). A comparison of addiction and transience among street youth: Los Angeles, California, Austin, Texas, and St. Louis, Missouri. Community Mental Health Journal, 46, 296-307. doi:10.1007/s10597-009-9264-x

14.

Finley

(1999). Sp’ange: A research story. Qualitative Inquiry, 5, 313-337. doi:10.1177/107780049900500302

15.

Fraser

M. W.

Galinsky

M. J.

Richman

J. M.

(1999). Risk, production, and resilience: Toward a conceptual framework for social work practice. Social Work Research, 23(3), 131-143. doi:10.1093/swr/23.3.131

16.

Glaser

B. G.

Strauss

A. L.

(1967). The discovery of grounded theory: Strategies for qualitative research. Hawthorne, NY: Aldine Publishing Company.

17.

Hahn

J. A.

Page-Shafer

Ford

Paciorek

Lum

P. J.

(2008). Traveling young injection drug users at high risk for acquisition and transmission of viral infections. Drug and Alcohol Dependence, 93(1-2), 43-50. doi:10.1016/j.drugalcdep.2007.08.016

18.

Hawkins

J. D.

Catalano

R. F.

Miller

J. Y.

(1992). Risk and protective factors for alcohol and other drug problems in adolescent and early adulthood: Implications for substance abuse prevention. Psychological Bulletin, 112(1), 64-105. doi:10.1037/0033-2909.112.1.64

19.

Hyde

(2005). From home to street: Understanding young people’s transitions into homelessness. Journal of Adolescence, 28, 171-183. doi:10.1016/j.adolescence.2005.02.001

20.

Jenson

J. M.

Fraser

M. W.

(2010). Social policy for children and families. A risk and resilience perspective (2nd ed.). Thousand Oaks, CA: SAGE.

21.

Kuckartz

(2001). Maxqda: Qualitative data analysis. Berlin, Germany: Verbi Software.

22.

Lankenau

S. E.

Sanders

Jackson-Bloom

Hathazi

Alarcon

Tortu

Clatts

M. C.

(2007). First injection of ketamine among young injection drug users (IDUs) in three U.S. cities. Drug and Alcohol Dependence, 87(2-3), 183-193. doi:10.1016/j.drugalcdep.2006.08.015

23.

Lofland

L. H.

(1995). Analyzing social settings. Belmont, CA: Wadsworth.

24.

Mallett

Rosenthal

Myers

Milburn

Rotheram-Borus

M. J.

(2004). Practising homelessness: A typology approach to young people’s daily routines. Journal of Adolescence, 27, 337-349. doi:10.1016/j.adolescence.2003.11.014

25.

Morgan

D. L.

(1998). Practical strategies for combining qualitative and quantitative methods: Applications to health research. Qualitative Health Research, 8, 362-376. doi:10.1177/104973239800800307

26.

Patton

M. Q.

(1990). Qualitative research and evaluation methods (2nd ed.). Newbury Park, CA: SAGE.

27.

Rew

Taylor-Seehafer

Thomas

N. Y.

Yockey

R. D.

(2001). Correlates of resilience in homeless adolescents. Journal of Nursing Scholarship, 33(1), 33-40. doi:10.1111/j.1547- 5069.2001.00033.x

28.

Rodwell

M. K.

(1998). Social work constructivist research. New York, NY: Garland Publishing, Inc.

29.

Sanders

Lankenau

S. E.

Jackson-Bloom

Hathazi

(2008). Multiple drug use and polydrug use amongst homeless traveling youth. Journal of Ethnicity in Substance Abuse, 7(1), 23-40. doi:10.1080/1533264002081893

30.

Seidman

I. E.

(1991). Interviewing as qualitative research: A guide for researchers in education and the social sciences. New York, NY: Teachers College Press.

31.

Sheehan

D. V.

Lecrubier

Sheehan

K. H.

Amorim

Janavs

Weiller

Dunbar

G. C.

(1998). The Mini-International Neuropsychiatric Interview (MINI): The development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. Journal of Clinical Psychiatry, 59, 22-33.

32.

Strauss

Corbin

(1998). Basics of qualitative research: Techniques and procedures for developing grounded theory (2nd ed.). Thousand Oaks, CA: SAGE.

33.

Thompson

S. J.

McManus

Voss

(2006). PTSD and substance abuse among youth who are homeless: Treatment issues and implications. Brief Treatment and Crisis Intervention, 6, 206-217. doi:10.1093/brief-treatment/mhl002

34.

Wagnild

G. M.

(2009). The Resilience Scale user’s guide. In Guinn

P. E.

(Ed.), The Resilience Scale user’s guide: For the U.S. English version of the Resilience Scale and the 14-item Resilience Scale (RS-14) (pp. 1-101). Worden, MT: Resilience Center.

35.

Wagnild

G. M.

Young

H. M.

(1993). Development and psychometric evaluation of the Resilience Scale. Journal of Nursing Measurement, 1(2), 165-178.

36.

Whitbeck

L. B.

(2009). Mental health and emerging adulthood among homeless young people. New York, NY: Psychology Press.