Study Demonstrates Need for Greater Resources to Support Students Experiencing Homelessness

A new study published in Social Service Review, “Who Counts? Educational Disadvantage among Children Identified as Homeless and Implications for the Systems That Serve Them,” examines the educational risks faced by children who are identified as homeless by the Department of Education (ED) and HUD. HUD’s definition of homelessness is narrower than the definition used by ED and is meant to reflect “literal homelessness,” while ED’s definition is broader and includes “doubled-up” families. The authors find that ED- and HUD-identified homeless students both experienced increases in chronic absenteeism (attendance rates <90%) and school mobility (moving to a new school during the school year) compared to peers from low-income but stably housed families. Yet ED-identified homeless students were significantly more likely than HUD-identified homeless students to experience school mobility. The authors conclude by suggesting that decisionmakers should reconsider how to prioritize resources for homeless students. 

To analyze differences among HUD- and ED-identified homeless students, the researchers started with administrative data from the 2013-2017 period for over 195,000 Minnesota public school students residing in the Hennepin County, Ramsey County, and West Central Continuums of Care (CoCs). Nearly 13% of students were identified as homeless under the ED definition and 3.5% qualified as homeless under the HUD definition during at least one year in the sample period. American Indian and Black students were much more likely to experience either form of homelessness than White, Hispanic, or Asian students. Twenty percent of American Indian and 17% of Black students experienced homelessness according to either definition over the study period compared to just 2% of White students.

The authors assessed educational outcomes in terms of school mobility, school attendance, reading proficiency, and math proficiency. Overall, students identified as homeless under either agency definition were at a disadvantage across all four educational outcomes compared to their peers whose families were low-income but not experiencing housing instability. After controlling for other factors, ED-identified homelessness was associated with a 28% increase in chronic absenteeism (attendance rate < 90%) and a 52% increase in school mobility relative to the population of students from low-income but stably housed families. ED-identified homeless students also experienced a 4% decline in math proficiency, but there was not a significant relationship between ED-identified homelessness and reading proficiency. HUD-identified homelessness was associated with 22% increases in both chronic absenteeism and school mobility relative to the population of students from low-income but stably housed families. However, after controlling for other factors, HUD-identified homelessness was not associated with changes in math or reading proficiency.

The authors also directly compared educational outcomes between HUD- and ED-identified homeless students. After controlling for other factors, ED-identified homeless students had a significantly higher likelihood of school mobility compared to HUD-identified homeless students relative to the population of students from low-income but stably housed families. Significant differences were not observed between ED-identified and HUD-identified homeless students regarding the other educational outcomes after controlling for a full range of factors. 

These findings challenge the common assumption that students experiencing HUD’s definition of “literal homelessness” have greater needs than those students who fall under ED’s more expansive definition. The authors argue that restricting homeless resources and services to HUD-identified homeless families should be reconsidered. They acknowledge, however, that expanding HUD’s definition of homelessness to include “doubled-up” families would likely overwhelm the already inadequate resources and capacities of CoCs.

Read the article at: