Student Homelessness in High School: Prevalence,Individual Characteristics,and Profiles of Risk and Multidomain Functioning

Abstract

We estimated the prevalence of high school homelessness and examined associations among homelessness, demographic factors, victimization experiences, and poor functioning using variable- and person-centered approaches. Anonymous self-report survey data from the 2019 Youth Risk Behavior Surveillance System represented approximately 6.5 million high school students in 29 regions of the United States. An estimated 9.32% of students experienced homelessness. Path analysis found homelessness was related to poor functioning, accounting for victimization and demographic factors. Latent class analysis found the highest risk profiles had high levels of homelessness, victimization, and poor functioning and an overrepresentation of multiracial, Native, and Hispanic/Latino students, and students with LGBT identities. Interventions to prevent, identify, and respond to youth homelessness should address sources of marginalization and adversity.

Keywords

homelessness and housing subjects of practice sexual identity identity/identities adolescence development across the lifespan race

Adolescent homelessness is associated with poor mental health, physical health, and academic functioning (Armstrong et al., 2018; Cutuli et al., 2020; Hodgson et al., 2013; Medlow et al., 2014) and is linked to higher rates of experiencing violence and victimization (Moore et al., 2019). Homelessness disproportionately affects multiple marginalized groups, including Black, Indigenous, and Hispanic/Latino youth as well as those who identify as lesbian, gay, bisexual, or transgender (LGBT; Corliss et al., 2011; Evangelist & Shaefer, 2020; Hatchimonji et al., 2021; Morton et al., 2018; Smith-Grant et al., 2022). This confluence of marginalization exacerbates inequity in education and health outcomes. Social workers are well-positioned to ameliorate inequities at different levels of practice and across settings through direct service to youth experiencing homelessness. Social workers are also called on to advocate on their behalf to encourage more equitable policies and social service systems (NASW, 2021). Yet, efforts to address adolescent homelessness are hindered by systems that have made limited progress in accurately identifying youth and the challenge of prioritizing service delivery for youth with myriad simultaneous social needs. To inform these efforts, we estimate homelessness prevalence in high schools using anonymous self-report surveys from the 2019 Youth Risk Behavior Surveillance System (YRBSS) and examine associated adverse experiences and indicators of poor functioning.

Estimating Prevalence of Student Homelessness

The lack of a consistent definition of homelessness and the varied methodologies and age groupings employed by different agencies hinder responses to youth homelessness. Official prevalence estimates of student homelessness are broadly recognized as underestimates (DiPerro & Mitchell, 2022; Morton et al., 2018). School districts identified about 2.5% (379,027) of all high school students as lacking a fixed, regular, and adequate nighttime residence during the 2018-2019 school year, a broad definition that includes unsheltered, sheltered, and shared housing (“doubled-up”) situations, among others (National Center for Homeless Education, 2021). However, past analyses of 2019 YRBSS data from 24 states suggest that during the 2018-19 school year state education agencies failed to identify about two-thirds of high school students experiencing homelessness (Hatchimonji et al., 2021). Meanwhile, the U. S. Department of Housing and Urban Development (HUD) reports that 171,670 people in families with children under 18 years old and 35,038 unaccompanied youth under 25 years old were homeless on a single night in 2019 (HUD, 2020). Recent findings using the YRBSS echo general themes across nearly a decade of research that is more restricted geographically or in content and analyzed with only variable-focused approaches: Official counts of student homelessness in education settings vastly underestimate the true prevalence of student homelessness and likely obscure its relation to indicators of student functioning (Cutuli, 2018; Cutuli et al., 2015, 2017; Perlman et al., 2014).

The widely recognized insufficiencies in understanding the prevalence of youth homelessness have interfered with efforts by social workers and advocates across disciplines (e.g., education, health, mental health care, child development) to establish means to adequately identify and serve these students (Cutuli et al., 2020; Morton et al., 2018). Thus, continued research and attention are needed to identify prevalence rates of youth homelessness that are more accurate than current official estimates and more reliably inform needed health and human services.

Promising methods for more accurately estimating the prevalence of adolescent homelessness are those that reduce social stigma. While stigma related to housing status may be a concern at any point in the lifespan, concern for how others perceive their homelessness is particularly salient for adolescents (Kidd, 2007). Thus, adolescents, more than other groups, may be more likely to share their housing status in anonymous surveys over other mechanisms that could draw attention to their housing instability. Beyond concerns about social stigma, many adolescents may prefer to avoid agency involvement or believe that support is not available (Cutuli et al., 2020). These adolescents may intentionally avoid identification by education and social service systems, necessitating strategies for identifying youth who are outside the traditional roles of homeless liaisons and homeless services.

Homelessness and Related Risks to Adolescent Functioning

The association between homelessness and poor functioning for adolescents is well-established. Like much of the risk and resilience literature, the literature examining associations with youth homelessness relies on variable-centered methods (e.g., Smith-Grant et al., 2022). These types of analyses consider how the experience of homelessness relates to other risk factors or outcomes across individuals and often consider the added risk of homelessness above and beyond other risks.

Research using variable-centered approaches has established that homelessness is associated with poor functioning across mental health, sexual health, and education domains. Homeless youth demonstrate higher rates of substance use and alcohol abuse (Medlow et al., 2014) and higher rates of sexually transmitted infections and risky sexual behavior (e.g., no condom use; Medlow et al., 2014; Solorio et al., 2008) compared with their housed peers. In addition, homelessness is associated with poor mental health compared with other children/adolescents living in poverty and the general population (Bassuk et al., 2015; Edidin et al., 2012). Findings from Armstrong and colleagues (2018) support a continuum of risk for mental health problems: Students currently experiencing homelessness are at the greatest risk, high school students with a history of homelessness are at the next greatest risk, and both of these groups are at greater risk than those who have never experienced homelessness. These results are similar to findings related to academic achievement in which youth who experienced homelessness or high housing mobility underperformed more stably housed and low-income peers on reading and math achievement, especially during years in which homelessness was occurring compared with years when it was not (Cutuli et al., 2013).

Homelessness is also linked to experiences of adversity that may place youth at additional risk for poor mental health and education outcomes. Transitional-age youth experiencing homelessness in San Francisco endorsed an extremely high rate of potentially traumatic events, with 77% of participants reporting four or more Adverse Childhood Experiences (Dawson-Rose et al., 2020). Students experiencing homelessness commonly report having been a victim of sexual, physical, or emotional violence (Heerde et al., 2015; Meltzer et al., 2012; Moore et al., 2019). Experiences of victimization are likely to both predate experiences of homelessness and recur during episodes of homelessness. Adolescents who are homeless or have run away from home have often experienced previous physical and sexual abuse (Tyler, 2006), and once they leave home they are at risk of further victimization (Whitbeck et al., 2001).

Homelessness represents a risk to adolescent development, likely through direct contributions to poor functioning and through its association with other experiences of adversity. However, few studies have considered how different experiences of adversity tend to cooccur as profiles of risk among students experiencing homelessness and how particular configurations of risk relate to different problems, such as emotional/behavioral problems, problematic substance and alcohol use, and others. In contrast to the prevailing literature based on variable-focused approaches, person-focused analyses make the person the unit of analysis and try to capture how variables tend to co-present in the lives of young people. A small literature has used person-centered methods to develop profiles of youth experiencing homelessness according to mental health (Adlaf & Zdanowicz, 1999; Hodgson et al., 2015), service needs (Bucher, 2008), and pathways to homelessness (Martijn & Sharpe, 2006). These studies corroborate variable-centered findings: Homeless youth experience high levels of trauma and mental health concerns and are not adequately served by current systems. Person-centered findings also illustrate how configurations of adverse experiences and indicators of poor functioning are related in a way that both acknowledges heterogeneity within the population of high school students and surfaces likely implications for service policy interventions.

Current Study

Homelessness is more likely to occur among marginalized groups and with a range of chronic and acute adversities, all of which have important implications for education, health, and human services as well as broader efforts to address structural sources of inequity (Cutuli, 2018). Less is known about how homelessness co-occurs with experiences of marginalization, victimization, and violence within individuals to impact health and education outcomes. Person-centered approaches are better suited to uncover such patterns of adversities and other factors with implications for outcomes.

To better understand the experiences of adolescents experiencing homelessness, we used data from the Centers for Disease Prevention and Control Youth Risk Behavior Surveillance System (YRBSS), a population-representative survey, to examine how high school student profiles of co-occurring experiences and characteristics relate to poor functioning. First, we aimed to estimate the self-reported prevalence of adolescent homelessness using this anonymous survey. Second, we used variable-centered path analysis to test the association of homelessness with poor functioning, controlling for three kinds of victimization experiences and demographic covariates. Consistent with the literature described earlier, we expected that homelessness would be linked to poor functioning in terms of suicidality, substance and alcohol use, risky sexual behavior, and academic outcomes. Third, we used person-centered latent class analysis to examine how student profiles of co-occurring experiences and characteristics within individuals related to indicators of poor functioning. Given the lack of current literature using this methodology to explore associations between teen homelessness and poor outcomes, we did not establish a priori hypotheses about the profiles that would emerge.

Method

Participants and Procedure

Data are from 29 non-overlapping geographies (states and school districts) that produced representative data about housing status on the 2019 YRBSS: Alaska, Arkansas, California, Cleveland (Ohio), Connecticut, Hawaii, Idaho, Illinois, Kansas, Kentucky, Louisiana, Maine, Maryland, Massachusetts, Michigan, Montana, New Hampshire, New Mexico, New York City (New York), North Carolina, North Dakota, Palm Beach (Florida), Pennsylvania, Rhode Island, Seattle (Washington), South Carolina, South Dakota, Vermont, and Virginia. Participants were selected using a two-stage cluster sample design and completed an anonymous paper-and-pencil survey. This resulted in 153,250 observations that represent an estimated population of 6,511,773.98 (95% confidence interval [CI]: [4,937,666.46, 8,085,881.50]) public high school students. See additional details elsewhere (Underwood et al., 2020). The Institutional Review Board of Nemours Children’s Health determined the study was exempt.

Measures

Indicators of poor functioning were severe behaviors that are considered clinically severe enough to warrant intervention, indexed dichotomously (absent/present). We detail item wording and operationalization in Table 1.

Table 1.

Operationalization of Variables From the 2019 YRBSS.

Variable	Item(s)	Operationalization
Sexual Victimization	1. Have you ever been physically forced to have sexual intercourse when you did not want to? 2. During the past 12 months, how many times did anyone force you to do sexual things that you did not want to do? 3. During the past 12 months, how many times did someone you were dating or going out with force you to do sexual things that you did not want to do?	1. “Yes” to any instance of forced sexual intercourse or 2 & 3. “1 or more times” of forced sexual experiences from anyone or a dating partner
Bullying Victimization	1. During the past 12 months, have you ever been bullied on school property? 2. During the past 12 months, have you ever been electronically bullied?	Any “yes”
Physical Victimization	1. During the past 12 months, how many times has someone threatened or injured you with a weapon such as a gun, knife or club on school property? 2. During the past 12 months, how many times did someone you were dating or going out with physically hurt you on purpose?	1 or more times to either of these experiences
Homeless	1. During the past 30 days, did you ever sleep away from your parents or guardians because you were kicked out, ran away, or were abandoned? 2. During the past 30 days, where did you usually sleep?	1. Any response besides “In my parent’s or guardian’s house” or “Somewhere else.” or 2. Any “yes”
Severe Suicidality	1. During the past 12 months, did you make a plan about how you would attempt suicide? 2. During the past 12 months, how many times did you actually attempt suicide? 3. If you attempted suicide during the past 12 months, did any attempt result in an injury, poisoning, or overdose that had to be treated by a doctor or nurse?	1. “Yes” or 2. “1 or more times” or 3. “Yes”
Hard Drug Use	1. During your life, how many times have you taken prescription pain medicine without a doctor’s prescription or differently than how a doctor told you to use it? 2. During your life, how many times have you used any form of cocaine, including powder, crack, or freebase? 3. During your life, how many times have you sniffed glue, breathed the contents of aerosol spray cans, or inhaled any paints or sprays to get high? 4. During your life, how many times have you used heroin (also called smack, junk, or China White)? 5. During your life, how many times have you used methamphetamines (also called speed, crystal meth, crank, ice, or meth)? 6. During your life, how many times have you used a needle to inject any illegal drug into your body?	1 or more times to any of these six questions
Problematic Alcohol Use	1. During the past 30 days, how many times did you drive a car or other vehicle when you had been drinking alcohol? 2. During the past 30 days, on how many days did you have 4 or more drinks of alcohol in a row, that is, within a couple of hours (if you are female) or 5 or more drinks of alcohol in a row, that is, within a couple of hours (if you are male)? 3. During the past 30 days, what is the largest number of alcoholic drinks you had in a row, that is, within a couple of hours?	1. “1 or more times” or 2. “1 or more days in a row” or 3. “Drank 5 or more drinks”
Risky Sexual Behavior	1. During your life, with how many people have you had sexual intercourse? 2. During the past 3 months, with how many people did you have sexual intercourse? 3. The last time you had sexual intercourse, did you or your partner use a condom? 4. The last time you had sexual intercourse, what one method did you or your partner use to prevent pregnancy?	1 & 2. “4 or more” sexual partners or 3. “No” Condom use or 4. “No method was used to prevent pregnancy”
Poor Grades	1. During the past 12 months, how would you describe your grades in school?	1. Received “D’s, F’s”

Note. YRBSS = Youth Risk Behavior Surveillance System.

Independent Variables

Independent variables were demographic and adversity factors, modeled to statistically predict indicators of poor functioning. Causal inference is limited by the cross-sectional nature of the survey. Demographic factors were age (treated as a quasi-continuous variable), race/ethnicity (coded five mutually exclusive dichotomous variables with “Non-Hispanic/Latino, White” as the reference group), LGBT identification (any reported identity as gay/lesbian, bisexual, or transgender), and sex (binary as asked by the standard YRBSS question). For race/ethnicity, we used group categories that reflect the survey items and acknowledge there are limitations to these categories and labels: Asian (Asian or Asian American, non-Hispanic/Latino), Black (Black or African American, non-Hispanic/Latino), Hispanic/Latino (Hispanic/Latino ethnicity of any race or multiple races), Native (non-Hispanic/Latino Native American/Hawaiian/Alaskan or American Indian or Pacific Islander), White (non-Hispanic), and Multiracial (more than one race selected, non-Hispanic/Latino). We deeply respect individuals’ right to self-describe their identities and encourage future efforts to offer ever-better ways to accomplish that in research, such as through the YRBSS. Given the importance of self-identification, we chose to use the labels provided on the YRBSS which the participants chose to endorse.

We operationalized homelessness using the McKinney-Vento definition federally required of education agencies (McKinney-Vento Homeless Assistance Act, 2015). Aligned with this definition, students were considered homeless if they reported that they (a) usually slept in a homeless situation in the last 30 days or (b) were abandoned, ran away, or were kicked out of their parent/guardian’s house. Students indicated a history of physical victimization if they were threatened or injured with a weapon at school or had been physically hurt by a dating partner in the past 12 months. Three items indexed sexual victimization, including having been physically forced to have unwanted sex, forced by a dating partner to have unwanted sex in the past 12 months, or forced by anyone to do sexual things in the past 12 months. Bullying victimization refers to having been bullied at school or electronically in the past 12 months.

Dependent Variables

We coded problematic alcohol use as present if the student indicated they had driven while using alcohol or engaged in binge drinking (4-5 drinks in a row; see Supplemental Materials for full items). Hard drug use was present if the student ever used cocaine, inhalants, methamphetamines, or heroin, misused pain medication, or injected an illicit substance. Poor grades were indicated if the student reported receiving “Mostly Ds” or “Mostly Fs.” We coded risky sexual behavior using two standards of risky sexual behavior common in past work: (a) no method to prevent the transmission of sexually transmitted disease during intercourse or (b) a high frequency of partners (4 or more sexual partners in life or past 3 months; Silverman et al., 2001). Severe suicidality was present if the student made a plan to attempt suicide, attempted suicide, or was injured as part of an attempted suicide.

Analyses

We accounted for stratification, clustering, and unequal selection probabilities related to the complex sampling design in all analyses using Mplus version 8.3 (Muthén & Muthén, 2017). Missing data appeared to conform to assumptions of Missing at Random. We accounted for missing data using FIML estimation (Collins & Lanza, 2009).

Variable-Centered Analysis

Just-identified path analysis models estimated relations between independent and dependent variables and the covariance between independent variables and between dependent variables, respectively. We tested for associations using separate direct effects of each adversity indicator and estimated relations between each demographic and adversity variable and dependent variable as separate, simultaneous predictors.

Person-Centered Analysis

We derived latent profiles of independent and dependent variables. We determined the best solution for the number of profiles by estimating a set of models with different numbers of latent classes (Masyn, 2013; Nylund-Gibson & Masyn, 2016). The best-fitting and most parsimonious solution met the following criteria: lowest Bayes Information Criterion (BIC) to justify increased complexity relative to model fit, overall entropy for the degree of differentiation among classes above 0.80, and smallest class size above 3%. Among the model solutions meeting these criteria, we preferred the most parsimonious model that best characterized the variability in rates of homelessness. We determined factors that differentiated each class by comparing within-class probabilities to the rates across all classes: We categorized factors as being different from the population rate if the within-class probability was less than 50% or greater than 150% of the across-class population rate (Shaw et al., 2019).

Results

Aim 1: Prevalence of Adolescent Homelessness

Prevalence of homelessness among high school students was 9.32% of the population (95% CI: [8.05%, 10.76%]), corresponding to an estimated 568,540 students across these 29 geographies (95% CI: [402,628, 734,453]). See Supplemental Table 1 for additional descriptive statistics.

Aim 2: Association of Homelessness With Poor Functioning (Variable-Centered)

Homelessness was significantly (p’s < .001) related to a higher likelihood of each indicator of poor functioning: suicidality (B = 0.10), substance abuse (B = 0.14), risky sexual behavior (B = 0.08), and poor grades (B = 0.09) when accounting for victimization and demographic factors. Correlations between independent variables indicate significant, positive associations between homelessness and each of the following: age, Black race, Native race, Hispanic/Latino ethnicity, LGBT identity, and each form of victimization. Homelessness was not associated with multiracial race and was negatively associated with Asian race.

Most covariates were also linked to indicators of poor functioning. Each form of victimization (physical, sexual, and bullying) had a significant positive association (p’s < .05) with each indicator of poor functioning, with two exceptions: Having been bullied was not associated with problematic alcohol use (p = .56), and sexual victimization was not significantly related to poor grades (p = .053). LGBT identity was significantly and positively associated with each dependent variable, except for problematic alcohol use. We provide estimates for other demographic factors in Table 2.

Table 2.

(A) Path Analysis Main Effect Estimates, and (B) Correlations.

(A)
	Severe suicidality		Hard drug use		Risky sexual behavior		Problematic alcohol use		Poor grades
Main effects	Beta	p	Beta	p	Beta	p	Beta	p	Beta	p
Adversity variables
Homelessness	.10	<.001	.14	<.001	.11	<.001	.08	<.001	.09	<.001
Physical Victimization	.13	<.001	.17	<.001	.12	<.001	.07	<.001	.051	<.001
Sexual Victimization	.20	<.001	.17	<.001	.15	<.001	.09	<.001	.05	.05
Bullying Victimization	.15	<.001	.11	<.001	.02	.004	.01	.56	.03	.03
Demographic variables
LGBT Identity	.14	<.001	.06	<.001	.05	<.001	−.01	.47	.03	.02
Age	−.01	.20	.01	.38	.22	<.001	.13	<.001	−.015	.12
Sex	−.01	.59	.01	.13	.03	.00	−.01	.52	.05	<.001
Race/ethnicity^a
Hispanic/Latino	.07	<.001	.09	<.001	.03	.04	−.06	<.01	.09	<.001
Multiracial	.01	.26	.01	.39	.01	.21	−.02	.03	.00	.73
Asian	.02	.04	−.00	.71	−.04	<.001	−.05	<.001	−.02	.03
Black	.03	<.001	.04	<.001	.04	<.001	−.11	<.001	.03	.01
Native	.03	<.001	.01	.04	.00	.60	−.02	.01	.01	.02
(B)
Dependent variables	Severe suicidality		Hard drug use		Risky sexual behavior		Problematic alcohol use		Poor grades
Dependent variables	Estimate	p	Estimate	p	Estimate	p	Estimate	p	Poor grades
Severe Suicidality	—	—
Hard Drug Use	.16	<.001	—	—
Risky Sexual Behavior	.06	<.001	.12	<.001	—	—
Problematic Alcohol Use	.02	.06	.13	<.001	.21	<.001	—	—
Poor Grades	.09	<.001	.07	<.01	.06	<.001	.04	<.01
Model R²	.20**		.18**		.14**		.06**		.04**
Independent variables	Age		Sex		LGBT Id.		Phys. Vict.	Sex. Vict.	Bullying Vict.	Homeless
Adversity Variables
Homelessness	.03*		.06***		.10***		.21***	.20***	.12***	—
Physical Victimization	.00		.02		.09***		—	.31***	.26***	21***
Sexual Victimization	.03		−.13***		.17***		.31***	—	.29***	.20***
Bullying Victimization	−.05***		−.09***		.13***		.26***	.29***	—	.12***
Demographic variables
LGBT Identity	.01		−.16***		—		.09***	.17***	.13***	.26***
Age	−		.02*		.01		.00	.03	−.05***	.03*
Sex	.02*		—		−.16***		.02	−.13***	−.09***	.06***
Race/ethnicity^a
Hispanic/Latino	−.05		−.02		.02		.09	.12**	.02	.08**
Multiracial	−.04***		−.01		.02*		−.00	.02	.01	−.01
Asian	−.02		−.00		−.02*		−.06	−.06***	−.04***	−.05***
Black	.02		−.00		.03**		.00	−.04**	−.05***	.07***
Native	−.01		.02**		.01		.01	.00	.00	.02**

Note. LGBT = Lesbian, Gay, Bisexual, or Transgender. *p < .05; **p < .01; ***p < .001.

‘Hispanic/Latino’ includes any endorsed race category. Each race category is non-Hispanic/Latino. White, non-Hispanic/Latino was the reference group.

Aim 3: Profiles of Co-Occurring Experiences and Poor Functioning (Person-Centered)

The solution with seven classes provided the best fit for the data based on having the lowest BIC among solutions with an entropy value above .8, a smallest class above 3%, and classes that delineate high-risk profiles in a way that is relevant to student homelessness (Table 3). Table 4 reports item response probabilities for each factor within each class for the 7-class solution.

Table 3.

Model Solution.

#Classes	#Parameters	AIC	BIC	Entropy	Smallest class (%)
1	18	2,331,330	2,331,509
2	36	2,232,765	2,233,122	0.786	20.0
3	55	2,212,250	2,212,796	0.874	16.8
4	74	2,195,338	2,196,074	0.816	12.7
5	93	2,184,716	2,185,640	0.837	11.8
6	112	2,176,441	2,177,555	0.838	6.1
7	131	2,169,537	2,170,839	0.803	5.8
8	150	2,164,621	2,166,112	0.777	4.9
9	169	2,160,778	2,162,458	0.786	4.9

Note. Accepted solution in boldface. BIC = Bayes Information Criterion; AIC = Akaike Information Criterion.

Table 4.

Probabilities for Each of Seven Latent Profiles.

Variable	Profile							Population base rate
	A	B	C	D	E	F	G
	Low problem	Average	Low problem	High risk	Average	Sex/alcohol risk	High risk
Problem indicators
Severe Suicidality	7.03%	10.29%	9.58%	58.50%	11.31%	13.79%	78.95%	18.90%
Hard Drug Use	5.28%	12.13%	8.43%	52.36%	15.06%	29.39%	77.59%	19.84%
Risky Sexual Behavior	8.28%	10.49%	1.24%	42.89%	21.19%	70.84%	42.12%	18.52%
Prob. Alcohol Use	7.48%	5.16%	1.76%	25.05%	2.01%	46.28%	17.38%	11.20%
Poor Grades	1.97%	6.49%	2.58%	10.92%	5.01%	6.28%	20.88%	5.61%
Adversity indicators
Homelessness	2.77%	4.44%	2.86%	23.47%	11.11%	13.08%	40.60%	9.32%
Physical Victimization	2.53%	4.56%	4.41%	38.99%	5.96%	11.47%	56.62%	11.59%
Sexual Victimization	4.49%	7.74%	6.64%	55.27%	5.81%	21.99%	78.57%	17.09%
Bullying Victimization	15.42%	13.47%	24.92%	67.25%	10.90%	19.09%	72.14%	25.55%
Demographic variables
LGBT Identity	7.82%	8.85%	7.19%	35.83%	10.75%	9.62%	29.55%	12.64%
Age (M [centered], SE)	0.677	−0.113	−1.174	0.087	0.374	1.482	0.008	0 (0.04)
Sex (Female)	55.15%	49.67%	53.50%	35.08%	52.62%	55.69%	48.60%	51.06%
Race/ethnicity^a
Hispanic/Latino	0.00%	100.00%	0.00%	0.00%	0.00%	25.74%	100.00%	26.09%
Multiracial	6.19%	0.00%	7.05%	8.24%	0.00%	2.77%	0.00%	4.14%
Asian	12.91%	0.00%	10.33%	3.60%	0.00%	1.50%	0.00%	6.01%
Black	0.00%	0.00%	8.20%	19.10%	100.00%	2.25%	0.00%	14.34%
Native	3.07%	0.00%	3.08%	5.01%	0.00%	1.36%	0.00%	2.08%
White^b	77.83%	0.00%	71.34%	64.05%	0.00%	66.37%	0.00%	47.33%
Percentage of Population	28.3%	18.1%	17.6%	11.1%	10.6%	8.4%	5.9%

Note. Boldface statistics denote >150% of the population base for factors. Italicized statistics denote <50% of the population base rate. LGBT = Lesbian, Gay, Bisexual, or Transgender.

Hispanic/Latino’ includes any endorsed race category. Each race category is non-Hispanic/Latino.

Non-Hispanic/Latino, White reference group.

High-Risk Profiles

There were two high-risk profiles (labeled “D” and “G” in Table 4), differentiated from other profiles by high levels of homelessness, each form of victimization, and all five problem indicators. These profiles each had a higher probability of LGBT-identifying students. They differed from each other based largely on race/ethnicity. Profile D (11.1% of the population) did not contain any Hispanic/Latino students, contained above-average shares of multiracial and Native students, and average probabilities for each other racial group. Profile G (5.9% of the population) was exclusively Hispanic/Latino students. It also appeared to have a higher probability of suicidality, substance abuse, poor grades, homelessness, and all forms of victimization.

Average Levels of Functioning Profiles

Two profiles indexed average or below average levels of homelessness and victimization and largely average levels of problems. These were also differentiated by race and ethnicity. Profile E (10.6% of the population) exclusively described Black students, had average probability of homelessness, average or below average probabilities of victimization, of LGBT-identifying students, and each problem indicator. Profile B (18.1% of the population) described Hispanic/Latino students exclusively and similarly had average or below-average probabilities for each indicator.

Low-Problem Profiles

Two profiles indexed below-average probability of homelessness, largely below-average probabilities of victimization and of problems, and the average likelihood of LGBT-identifying students. One of these low-problem profiles (“C”; 17.6% of the population) exclusively described non-Hispanic students, with an overrepresentation of multiracial, Asian, and White students. The other low-problem profile (“A”; 28.3% of the population) described students who were neither Hispanic/Latino nor Black, with above-average probabilities of multiracial, Asian, and White students.

High-Risk Behaviors Related to Sex and Alcohol

Finally, Profile F (8.4% of the population) described an average probability of homelessness and most other variables and an above-average likelihood of risky sexual behavior and problematic alcohol use. The average age for this group appeared higher than the population overall. Racial groups and Hispanic/Latino ethnicity occurred at rates that approximated the population base rates, except that the likelihood of Asian race was below average for this profile.

Discussion

When using anonymously self-reported, population-representative surveys from 25 states and 4 districts, public high school student homelessness was much more prevalent than indicated by official estimates. These findings underscore the insufficiency of prevalence estimates from federal agencies and expand the evidence base of student homelessness. The high prevalence of homelessness in high school indicates an urgent need to improve prevention strategies for homelessness. Findings from person- and variable-centered analyses align with previous research: Homelessness is related to several indicators of poor functioning and tends to co-occur with victimization and marginalization based on LGBT identification, race, and ethnicity. The current findings underscore the need to improve identification and intervention related to both homelessness and accompanying adversities, particularly for youth and families experiencing identity-based marginalization.

Aim 1: Self-Reported Prevalence of Student Homelessness

Anonymous, self-reported rates of homelessness for high school students are consistently much higher than those reported by the Department of Education or HUD (Cutuli et al., 2020; Hatchimonji et al., 2021). In these 29 regions alone, the current analyses suggest 568,540 high schoolers experienced homelessness in the prior 30 days. This is well beyond the 379,027 high school students reported by education agencies across the entire United States during the whole of the 2018-2019 school year (National Center for Homeless Education, 2021). The anonymous, self-report methodology likely helped overcome common challenges in estimating prevalence. Specifically, adolescents who wish to avoid stigma and agency involvement may be more likely to respond honestly to an anonymous survey. Furthermore, the YRBSS items ask about specific homeless situations that adolescents may not consider to be “homeless.” In addition, the use of a complex sampling design allows the survey respondents to represent the entire population of public high school students without needing to identify every individual student experiencing homelessness. The true scope of student homelessness outstrips the capacities of federal and many local education, health, and human service response systems specifically charged with identifying and responding to students’ needs (DiPerro & Mitchell, 2022; United States Government Accountability & Office, 2014, 2020)

Aim 2: Association of Homelessness With Poor Functioning (Variable-Centered)

Consistent with past research, our variable-centered path analysis showed homelessness to be significantly related to substance use and alcohol abuse (Edidin et al., 2012), suicidality (Perlman et al., 2014), risky sexual behavior (Halverson et al., 2022), and poor educational outcomes (Obradović et al., 2009), beyond the effects of victimization and some forms of likely marginalization indexed by LGBT, racial, and ethnic identities. The field needs additional longitudinal research to better understand the mechanisms through which homelessness is associated with these outcomes. Nevertheless, the current variable-centered findings warrant efforts to prevent and address homelessness based on reducing poor outcomes and increasing public health and educational equity.

Aim 3: Profiles of Co-Occurring Experiences and Poor Functioning (Person-Centered)

Homelessness is a salient threat to health, mental health, and education in ways that are evident through our person-centered analyses. Not only does adolescent homelessness appear to contribute to poor outcomes, but it often occurs in the contexts of victimization and marginalization, underscoring the seriousness of failing to identify and effectively serve these students and their families. Results from the person-centered latent class analysis emphasize the patterns of co-occurring adverse experiences within individuals. Profiles with higher rates of homelessness also had the highest rates of each indicator of poor functioning, and, importantly, homelessness co-occurred with physical, sexual, and bullying victimization. These profiles should inform efforts to comprehensively meet the needs of students who experience homelessness.

Findings echoed previous literature indicating disproportionately high rates of homelessness among LGBT, Indigenous, Black, and Hispanic/Latino youth (Corliss et al., 2011; Evangelist & Shaefer, 2020; Hatchimonji et al., 2021; Morton et al., 2018). Thus, services for students experiencing homelessness should be responsive to the needs specific to youth with historically marginalized race, ethnic, sexual and/or gender identities (McCann & Brown, 2019; Reisner et al., 2015). Students from Native and multiple race groups were overrepresented in one profile marked by homelessness and other risks, while a second profile exclusively contained students who identified as Hispanic/Latino and showed high rates of homelessness and risk. The differentiation of a very high-risk group that was also exclusively of Hispanic/Latino ethnicity strongly warrants additional research to understand unmeasured features of this profile, a finding that would be less apparent in most variable-focused approaches. Future studies and administrations of the YRBSS should attempt to discern important and yet unmeasured factors that might ultimately help define this profile. For example, the standard YRBSS does not include questions about experiences of racism or discrimination, immigration status, family socioeconomic status, or sense of ethnic identity and pride, to name just a few additional variables that might better account for this distinct profile.

Limitations

Results should be interpreted in light of several limitations. First, the cross-sectional design of the YRBSS prohibits us from making causal inferences about the relations among the variables analyzed here. Second, the purpose of this surveillance system is to track many population-level risks over time, but this also means that most variables are informed by only one or a few items. Methods that deploy richer, psychometrically established assessment tools would be better measures of key constructs. Relatedly, with an existing dataset, any analysis is limited by the particular questions on the 2019 YRBSS. For example, we were unable to include additional forms of violence, victimization, or abuse beyond the items included here. We were also unable to account for involvement in social service systems or foster care in the current study. Specific experiences of adversity that may be related to marginalization and oppression are also not tracked by this survey, such as discrimination experiences based on racial, gender, or sexual identities. In addition, YRBSS surveys track very few developmental assets or strengths that may buffer against risks and contribute to resilience. Overall, our analyses suggest that homelessness is a complex issue given patterns of co-occurring risk.

Implications for Practice

Social workers have opportunities to support youth experiencing homelessness, marginalization, and victimization at micro (e.g., direct service), mezzo (e.g., supporting schools or health care systems), and macro (e.g., advocacy, policy change) levels (Rine & LaBarre, 2020). Adequately addressing high school homelessness will require investment in and coordination among youth-serving systems, a mezzo-level process social workers are uniquely poised to undertake. Universal youth-serving systems like schools and primary care offices are contexts where experiences of homelessness and related adversities may be known. Coordinating within and across these systems could improve the identification of high school students experiencing or at risk for homelessness (Ball et al., 2021). Increasingly, integrated social services and behavioral health care within these universal systems are equipped to provide services at the selected level of intervention (Mancini et al., 2020). Micro-level interventions in these contexts might include trauma-informed care and interventions that directly address substance use, sexual health, education needs, and mental health concerns (Bassuk et al., 2014; Crawford, 2022; Hopper et al., 2010) or peer support processes (Kidd et al., 2019).

At the macro level, current findings call attention to the long-standing oppressive social structures that have created and perpetuated inequity across education, housing, and health domains. A robust response to homelessness must recognize and work to dismantle the structural racism that manifests as discriminatory housing practices, high rates of parental incarceration (impacting social and financial resources), poor access to health care (exacerbating the financial and social impact of maternal and child illnesses), and disproportionately high rates of poverty (Evangelist & Shaefer, 2020; Giano et al., 2020). In addition, a comprehensive response to youth homelessness must address the unique discriminatory experiences faced by LGBT youth, such as bias in education and health systems, peer victimization, and family conflict and rejection (McCann & Brown, 2019). Social workers have an ethical responsibility to provide—to the best of their ability—necessary resources, services, and opportunities to meet the client’s basic needs; thus, it is necessary for social workers to get the most accurate depiction of the number of students missing this crucial resource to better service individuals and families (NASW, 2021). Social workers also have an ethical obligation to advocate for policies and practices that promote equitable distribution of social and financial resources and equitable access to health care and quality education.

Homelessness represents a context of risk for high school students. Student homelessness is far more prevalent than reported in official counts from either education or housing systems. Variable-focused analyses confirm relations between homelessness and various forms of poor functioning, beyond experiences of victimization and differences between measured demographic and identity factors. Findings from latent class analysis emphasize that homelessness is overrepresented in profiles marked by multiple high rates of varied victimization experiences as well as problems. Furthermore, rates of homelessness intersect with different marginalized identities, especially LGBT, Native and multiracial race, and Hispanic/Latino ethnicity. These findings not only inform calls for interventions and policy reforms that are sensitive to trauma, marginalization, and the multidomain needs of students experiencing homelessness but also amplify the urgency of establishing sustainable direct service programs, supporting the systems that provide those services, and enacting policies that promote the prevention of homelessness and provide a comprehensive response when it occurs.

Supplemental Material

sj-docx-1-fis-10.1177_10443894231215522 – Supplemental material for Student Homelessness in High School: Prevalence, Individual Characteristics, and Profiles of Risk and Multidomain Functioning

Supplemental material, sj-docx-1-fis-10.1177_10443894231215522 for Student Homelessness in High School: Prevalence, Individual Characteristics, and Profiles of Risk and Multidomain Functioning by Danielle R. Hatchimonji, Janette E. Herbers, Claire Flatley, Dan Treglia and J. J. Cutuli in Families in Society

Footnotes

Acknowledgements

We are grateful for help, guidance, and encouragement from many in preparing this article, especially Lisa Whittle and colleagues at the Centers for Disease Control and Prevention, Amanda Lewis at the Nemours Center for Health Care Delivery Science, John McLaughlin of the U. S. Department of Education, Joe Willard of HopePHL, and Patricia Julianelle of SchoolHouse Connection. We are also thankful for the assistance of the state and local district education liaisons who identify, serve, and advocate on behalf of students experiencing homelessness every day.

Disposition editor: Cristina Mogro-Wilson

Declaration of Conflicting Interests

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

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: J. J. Cutuli’s work on this project was supported, in part, by the Research Expanding Access to Child Health (REACH) Center, which is supported by the National Institute of General Medical Sciences of the NIH under grant number P20GM144270.

ORCID iD

J. J. Cutuli

Supplemental Material

Supplemental material for this article is available online.

References

Adlaf

E. M.

Zdanowicz

Y. M.

(1999). A cluster-analytic study of substance problems and mental health among street youths. American Journal of Drug and Alcohol Abuse, 25(4), 639–660. https://doi.org/10.1081/ADA-100101884

Armstrong

J. M.

Owens

C. R.

Haskett

M. E.

(2018). Mental health of homeless youth: Moderation by peer victimization and teacher support. Child Psychiatry & Human Development, 49(5), 681–687. https://doi.org/10.1007/s10578-018-0790-z

Ball

Bowen

E. A.

Jones

A. S.

(2021). Integrating trauma-informed care and collective impact: Perspectives of service providers working with cross-system youth. Journal of the Society for Social Work and Research, 12(1), 59–81.

Bassuk

E. L.

DeCandia

C. J.

Tsertsvadze

Richard

M. K.

(2014). The effectiveness of housing interventions and housing and service interventions on ending family homelessness: A systematic review. American Journal of Orthopsychiatry, 84(5), 457–474. https://doi.org/10.1037/ort0000020

Bassuk

E. L.

Richard

M. K.

Tsertsvadze

(2015). The prevalence of mental illness in homeless children: Systematic review and meta-analysis. Journal of the American Academy of Child & Adolescent Psychiatry, 54(2), 86–96.e2. https://doi.org/10.1016/j.jaac.2014.11.008

Bucher

C. E.

(2008). Toward a needs-based typology of homeless youth. Journal of Adolescent Health, 42(6), 549–554. https://doi.org/10.1016/j.jadohealth.2007.11.150

Collins

L. M.

Lanza

S. T.

(2009). Latent class and latent transition analysis: With applications in the social, behavioral, and health sciences (Vol. 718). Wiley.

Corliss

H. L.

Goodenow

C. S.

Nichols

Austin

S. B.

(2011). High burden of homelessness among sexual-minority adolescents: Findings from a representative Massachusetts high school sample. American Journal of Public Health, 101(9), 1683–1689. https://doi.org/10.2105/AJPH.2011.300155

Crawford

(2022). Evaluating trauma-informed care practices in an interdisciplinary homeless service collaboration. Journal of Evidence-Based Social Work, 19(2), 212–227. https://doi.org/10.1080/26408066.2022.2026267

10.

Cutuli

J. J.

(2018). Homelessness in high school: Population-representative rates of self-reported homelessness, resilience, and risk in Philadelphia. Social Work Research, 42(3), 159–168. https://doi.org/10.1093/swr/svy013

11.

Cutuli

J. J.

Desjardins

C. D.

Herbers

J. E.

Long

J. D.

Heistad

Chan

C.-K.

Hinz

Masten

A. S.

(2013). Academic achievement trajectories of homeless and highly mobile students: Resilience in the context of chronic and acute risk. Child Development, 84(3), 841–857. https://doi.org/10.1111/cdev.12013

12.

Cutuli

J. J.

Montgomery

A. E.

Evans-Chase

Culhane

D. P.

(2017). Childhood adversity, adult homelessness and the intergenerational transmission of risk: A population-representative study of individuals in households with children. Child & Family Social Work, 22(1), 116–125.

13.

Cutuli

J. J.

Steinway

Perlman

Herbers

J. E.

Eyrich-Garg

K. M.

Willard

(2015). Youth homelessness: Prevalence and associations with weight in three regions. Health & Social Work, 40(4), 316–324.

14.

Cutuli

J. J.

Treglia

Herbers

J. E.

(2020). Adolescent homelessness and associated features: Prevalence and risk across eight states. Child Psychiatry & Human Development, 51, 48–58. https://doi.org/10.1007/s10578-019-00909-1

15.

Dawson-Rose

Shehadeh

Hao

Barnard

Khoddam-Khorasani

Leonard

Clark

Kersey

Mousseau

Frank

Miller

Carrico

Schustack

Cuca

Y. P.

(2020). Trauma, substance use, and mental health symptoms in transitional age youth experiencing homelessness. Public Health Nursing, 37(3), 363–370. https://doi.org/10.1111/phn.12727

16.

DiPerro

Mitchell

(2022). Hidden toll: Thousands of schools fail to count homeless students. https://publicintegrity.org/education/unhoused-and-undercounted/schools-fail-to-count-homeless-students/

17.

Edidin

J. P.

Ganim

Hunter

S. J.

Karnik

N. S.

(2012). The mental and physical health of homeless youth: A literature review. Child Psychiatry & Human Development, 43(3), 354–375. https://doi.org/10.1007/s10578-011-0270-1

18.

Evangelist

Shaefer

H. L.

(2020). No place called home: Student homelessness and structural correlates. Social Service Review, 94(1), 4–35. https://doi.org/10.1086/707569

19.

Giano

Williams

Hankey

Merrill

Lisnic

Herring

(2020). Forty years of research on predictors of homelessness. Community Mental Health Journal, 56(4), 692–709. https://doi.org/10.1007/s10597-019-00530-5

20.

Halverson

Hatchimonji

D. R.

Treglia

Flatley

C. A.

Herbers

J. E.

Cutuli

J. J.

(2022). Risky sexual behavior and STI testing among teens experiencing homelessness. Children and Youth Services Review, 139, Article 106538. https://doi.org/10.1016/j.childyouth.2022.106538

21.

Hatchimonji

D. R.

Flatley

C. A.

Treglia

Cutuli

J. J.

(2021). High school students experiencing homelessness: Findings from the 2019 youth risk behavior surveillance system (YRBSS). https://www.nemours.org/content/dam/nemours/shared/collateral/community-reports/yrbss-2019-high-school-homelessness-report.pdf

22.

Heerde

J. A.

Scholes-Balog

K. E.

Hemphill

S. A.

(2015). Associations between youth homelessness, sexual offenses, sexual victimization, and sexual risk behaviors: A systematic literature review. Archives of Sexual Behavior, 44(1), 181–212. https://doi.org/10.1007/s10508-014-0375-2

23.

Hodgson

K. J.

Shelton

K. H.

van den Bree

M. B.

(2015). Psychopathology among young homeless people: Longitudinal mental health outcomes for different subgroups. British Journal of Clinical Psychology, 54(3), 307–325. https://doi.org/10.1111/bcj.12075

24.

Hodgson

K. J.

Shelton

K. H.

van den Bree

M. B. M.

Los

F. J.

(2013). Psychopathology in young people experiencing homelessness: a systematic review. American Journal of Public Health, 103(6), Article e2437. https://doi.org/10.2105/AJPH.2013.301318

25.

Hopper

E. K.

Bassuk

E. L.

Olivet

(2010). Shelter from the storm: Trauma-informed care in homelessness services settings. Open Health Services and Policy Journal, 3(1), 80–100. https://doi.org/10.2174/1874924001003010080

26.

Kidd

S. A.

(2007). Youth homelessness and social stigma. Journal of Youth and Adolescence, 36(3), 291–299.

27.

Kidd

S. A.

Vitopoulos

Frederick

Daley

Peters

Cohen

Gutierrez

Leon

McKenzie

(2019). Peer support in the homeless youth context: Requirements, design, and outcomes. Child and Adolescent Social Work Journal, 36, 641–654.

28.

Mancini

M. A.

Maynard

Cooper-Sadlo

(2020). Implementation of an integrated behavioral health specialization serving children and youth: Processes and outcomes. Journal of Social Work Education, 56(4), 793–808.

29.

Martijn

Sharpe

(2006). Pathways to youth homelessness [Case Reports]. Social Science & Medicine, 62(1), 1–12. https://doi.org/10.1016/j.socscimed.2005.05.007

30.

Masyn

K. E.

(2013). Latent class analysis and finite mixture modeling. In Little

T. D.

(Ed.), The Oxford handbook of quantitative methods (pp. 551–611). Oxford University Press.

31.

McCann

Brown

(2019). Homelessness among youth who identify as LGBTQ+: A systematic review. Journal of Clinical Nursing, 28(11–12), 2061–2072. https://doi.org/10.1111/jocn.14818

32.

McKinney-Vento Homeless Assistance Act of 1987, § 11431-11434a. (2015). https://uscode.house.gov/view.xhtml?path=/prelim@title42/chapter119/subchapter6/partB&edition=prelim

33.

Medlow

Klineberg

Steinbeck

(2014). The health diagnoses of homeless adolescents: A systematic review of the literature. Journal of Adolescence, 37(5), 531–542. https://doi.org/10.1016/j.adolescence.2014.04.003

34.

Meltzer

Ford

Bebbington

Vostanis

(2012). Children who run away from home: Risks for suicidal behavior and substance misuse. Journal of Adolescent Health, 51(5), 415–421. https://doi.org/10.1016/j.jadohealth.2012.04.002

35.

Moore

Astor

R. A.

Benbenishty

(2019). A statewide study of school-based victimization, discriminatory bullying, and weapon victimization by student homelessness status. Social Work Research, 43(3), 181–194. https://doi.org/10.1093/swr/svz010

36.

Morton

M. H.

Dworsky

Matjasko

J. L.

Curry

S. R.

Schlueter

Chavez

Farrell

A. F.

(2018). Prevalence and correlates of youth homelessness in the United States. Journal of Adolescent Health, 62(1), 14–21. https://doi.org/10.1016/j.jadohealth.2017.10.006

37.

Muthén

L. K.

Muthén

(2017). Mplus user’s guide: Statistical analysis with latent variables, user’s guide.

38.

National Assosciation of Social Workers. (2021). Code of ethics of the national association of social workers.

39.

National Center for Homeless Education. (2021). Federal data sumary school years 2016-17 through 2018-2019: Education for homeless children and youth. http://profiles.nche.seiservices.com/ConsolidatedStateProfile.aspx#

40.

Nylund-Gibson

Masyn

K. E.

(2016). Covariates and mixture modeling: Results of a simulation study exploring the impact of misspecified effects on class enumeration. Structural Equation Modeling: A Multidisciplinary Journal, 23(6), 782–797.

41.

Obradović

Long

J. D.

Cutuli

J. J.

Chan

C. K.

Hinz

Heistad

Masten

A. S.

(2009). Academic achievement of homeless and highly mobile children in an urban school district: Longitudinal evidence on risk, growth, and resilience. Development and Psychopathology, 21(2), 493–518. https://doi.org/10.1017/S0954579409000273

42.

Perlman

Willard

Herbers

J. E.

Cutuli

J. J.

Eyrich Garg

K. M.

(2014). Youth homelessness: Prevalence and mental health correlates. Journal of the Society for Social Work and Research, 5(3), 361–377. https://doi.org/10.1086/677757

43.

Reisner

S. L.

Bradford

Hopwood

Gonzalez

Makadon

Todisco

Cavanaugh

Vanderwarker

Grasso

Zaslow

Boswell

S. L.

Mayer

(2015). Comprehensive transgender healthcare: The gender affirming clinical and public health model of fenway health. Journal of Urban Health, 92(3), 584–592. https://doi.org/10.1007/s11524-015-9947-2

44.

Rine

C. M.

LaBarre

(2020). Research, practice, and policy strategies to end homelessness. Health & Social Work, 45(1), 5–8.

45.

Shaw

S. H.

Herbers

J. E.

Cutuli

(2019). Medical and psychosocial risk profiles for low birthweight and preterm birth. Women’s Health Issues, 29(5), 400–406. https://doi.org/10.1016/j.whi.2019.06.005

46.

Silverman

Raj

Mucci

Hathaway

(2001). Dating violence against adolescnt girls and associated substance use, unhealthy weight control, sexual risk behavior, pregnancy, and suicidality. JAMA, 286, 572–579. https://doi.org/10.1001/jama.286.5.572

47.

Smith-Grant

Kilmer

Brener

Robin

Underwood

J. M.

(2022). Risk behaviors and experiences among youth experiencing homelessness-youth risk behavior survey, 23 U.S. states and 11 local school districts, 2019. Journal of Community Health, 47(2), 324–333. https://doi.org/10.1007/s10900-021-01056-2

48.

Solorio

M. R.

Rosenthal

Milburn

N. G.

Weiss

R. E.

Batterham

P. J.

Gandara

Rotheram-Borus

M. J.

(2008). Predictors of sexual risk behaviors among newly homeless youth: A longitudinal study. Journal of Adolescent Health, 42(4), 401–409. https://doi.org/10.1016/j.jadohealth.2007.09.023

49.

Tyler

K. A.

(2006). A qualitative study of early family histories and transitions of homeless youth. Journal of Interpersonal Violence, 21(10), 1385–1393. https://doi.org/10.1177/0886260506291650

50.

Underwood

J. M.

Brener

Thornton

Harris

W. A.

Bryan

L. N.

Shanklin

S. L.

Deputy

Roberts

A. M.

Queen

Chyen

Whittle

Lim

Yamakawa

Leon-Nguyen

Kilmer

Smith-Grant

Demissie

Jones

S. E.

Clayton

Dittus

(2020). Overview and methods for the youth risk behavior surveillance system-United States 2019. MMWR Supplements, 69(1), 1–10. https://www.cdc.gov/mmwr/volumes/69/su/pdfs/su6901a1-H.pdf

51.

United States Government Accountability Office. (2014). Education of homeless students: Improved program oversight needed.

52.

United States Government Accountability Office. (2020). Homelessness: Better HUD oversight of data collection could improve estimates of homeless population.

53.

US Department of Housing and Urban Development. (2020). The 2019 Annual Homeless Assessment Report (AHAR) to congress, Part 1: Point-in-time estimates of homelessness. https://www.hudexchange.info/resource/5948/2019-ahar-part-1-pit-estimates-of-homelessness-in-the-us/

54.

Whitbeck

L. B.

Hoyt

D. R.

Yoder

K. A.

Cauce

A. M.

Paradise

(2001). Deviant behavior and victimization among homeless and runaway adolescents. Journal of Interpersonal Violence, 16(11), 1175–1204. https://doi.org/10.1177/088626001016011005

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB