Our Executive Director Sherilyn Adams co-authored an editorial with  Dr. Colette L. Auerswald, Professor of Public Health and MS Program Director for the UC Berkeley-UCSF Joint Medical Program that was just released called “Counting All Homeless Youth Today So We May No Longer Need To Tomorrow.”

The editorial focuses on the importance of looking at youth homelessness as a national issue rather than just an urban issue, therefore emphasizing the need to conduct a more reliable national survey on youth homelessness. This following has been adapted from a forthcoming op-ed publication in the Journal of Adolescent Health, with the permission of the authors and the journal.


Somehow we have come to accept homelessness in the US, including youth homelessness, as an inevitability of modern urban life. Yet, anyone born before 1980 has lived in a world when youth homelessness at the current scale was unimaginable. In fact, today’s adult, youth and family homelessness crisis is a direct result of structural shocks less than 40 years ago, including the large-scale disinvestment in low income housing by HUD implemented in the Reagan era.

A study published this week in the Journal of Adolescent Health led by researchers at the University of Chicago calls attention to the vastness of youth homelessness, at a magnitude far exceeding prior estimates. Notably, the article also calls for solutions that will not only scale effective housing and supportive services to address all currently and imminently homeless youth, but also the implementation of new structural interventions to address causes.

The study estimates that 660,000 U.S. households had at least one 13-17 year old who experienced homelessness in the year prior to the survey. Moreover, the study estimates that 3.5 million individual 18-25 year olds experienced homelessness during the same period. Based on the conservative assumption that each of the households surveyed had a single affected minor, this adds up to 4.16 million 13 to 25 year olds experiencing homelessness in a single year. This is a number logarithmically higher than prior estimates, and for good reason, as we shall see.

The study was led by researchers at the Chapin Center at the University of Chicago, which is home to Voices of Youth Count, a public-private partnership that aims to improve national data on youth homelessness. Their work is very timely; The United States Interagency Council on Homelessness (USICH), the federal body in charge of coordinating homelessness policy, committed to a goal of ending youth homelessness by 2020 in its strategic plan. As rightly stated by the USICH, we must first be able to count and characterize the population and do so reproducibly and over time to reach this goal. However, counting homeless minors and transitional aged youth 18-24 years of age has been, to say the least, a vexing problem.

The Department of Housing and Urban Development has conducted point-in-time counts of the homeless population on a single night in January since 2005. These counts have always included a distinct count of minors (both unaccompanied minors and children in families). In response to the 2020 goal of ending youth homelessness, a separate count of 18-24 year olds was added in 2013.

Although counting transitional aged youth as part of the point-in-time counts has been an important advancement, the findings have been plagued with challenges, yielding numbers that are widely recognized as vast undercounts. Counts to date have been primarily based on visual assessments of youth in public places, though an increasing number of communities are now directly surveying youth in street venues to count them. Visual counts systematically undercount youth who go to great lengths to hide their homelessness as well as youth who shun venues where homeless youth congregate (particularly true both of youth of color and immigrant youth). Furthermore, the same underfunded county government entities entrusted with counting youth are politically disincentivized from doing so inclusively, as higher numbers suggest ineffective public investments. Unsurprisingly then, the percent of counties in California that reported no homeless minors and fewer than 100 homeless transitional aged youth (numbers that defy common sense for any county) has remained stubbornly high.

Distinct from HUD’s efforts, the Department of Education counts homeless minors in schools, dictated by the McKinney Vento Act. Unlike the point-in-time count, these numbers reflect a count over an entire year (as opposed to a single night). The definition of youth homelessness under McKinney Vento is broader than the HUD definition, and includes couch surfing. However, this count does not include out of school youth or young adults and to date has yielded a vanishing low count of unaccompanied minors.

Thus, current data do not allow us to get a population-level scope of the dimensions of the problem that accounts for the broad range of experience of homelessness from adolescence to young adulthood.

The importance of counting youth who are couch surfing cannot be overstated. The definition that the new study in the Journal of Adolescent Health employs describes a form of shelter that is neither safe, developmentally appropriate, nor conducive to accomplishing the tasks of adolescence. Furthermore, we have learned in our joint 50 years of experience in research and practice that couch surfing youth and youth experiencing street-based homelessness are highly overlapping groups, exchanging places depending on the time of the month, the circumstances of the day, or the patience and resources of the hosts. Our experience is confirmed by the findings in the current paper.

The paper’s approach to counting youth experiencing homelessness most certainly has its limitations, well highlighted by the authors. The greatest limitations stem from the reporting by adults of the homelessness status of minors and transitional aged youth under their roof. The inclusion error rate has been well accounted for through the supplemental interview that was conducted, with a large inclusion error rate for the estimate of couch surfing only youth and a smaller inclusion error rate for youth who were “explicitly” homeless. However, this correction does not correct for the inevitable exclusion error in this approach. A failure to report minor and transitional aged youth homelessness could result from social desirability and/or fear of being reported to the authorities.

There are a number of potential ways in which this approach might be improved. The first would be to interview minors directly, as was done for transitional aged youth respondents. Another improvement would be to include more data characterizing the youth population (acknowledging this would lengthen the survey, thereby decreasing feasibility). In particular, overlap with the juvenile and adult justice system would be informative to addressing the intersectionality of the experience of these youth. Finally, guidelines for samples that could be constructed to represent data at the state or local level would be worthwhile, to allow for more regional or local planning.

We should not simply throw up our hands in despair, however. The new study, complemented by existing research, provides important guidelines for moving forward. First, at a societal level, we must dispel any perception that youth homelessness is solely an urban issue, but rather a national one. Any true solution must incorporate the needs of rural youth, whose rates of homelessness are equivalent to those of urban youth. Next, we must explore ways to develop an early warning system for street homelessness for minors and youth, engaging schools and health providers and other mandated reporters, for example. Third, we must address the stigma, racism, homophobia, transphobia and xenophobia that are key drivers for youth homelessness. Fourth, education must continue to be a critical vaccine against future homelessness, as well as against negative consequences once youth are on the street. Fifth, inclusiveness, particularly among youth of color and youth who identify as LGBTQ, requires systematically incorporating youth voices into program design. Finally, we must intentionally partner with youth’s communities of origin. Researchers in particular, but providers and policy makers as well, are often guilty of visualizing youth experiencing homelessness in a vacuum, instead of in the context of their families, extended families, and communities, as we would for housed youth.

Ending youth homelessness once and for all is long overdue. Being able to count and characterize the population nationally and longitudinally may continue to be elusive. However, in a world of imperfect counts, the University of Chicago work is a welcome addition.


Learn more about Larkin Street’s work to end youth homelessness on our website.