An Examination of Race and Gender in Predicting Housing and Time to Housing for Youth Experiencing Homelessness

Background and Purpose: The number of unaccompanied youth experiencing homelessness increased by 15 percent between 2022 and 2023 (de Sousa, 2023). Furthermore, youth experiencing homelessness are more likely to be non-white and women or girls (de Sousa, 2023). This study analyzed Homeless Management Information System (HMIS) data for evidence of racial and gender parity in the match and referral process of the coordinated entry system for youth. The purpose of this study was threefold: (1) to determine predictors of housing for youth experiencing homelessness, (2) to determine predictors of time to housing, and (3) to examine the effect of possible racial and gender disparities on housing outcomes.

Methods: This study utilized 15 months of community data from one HMIS (n=215) that consisted of youth ranging in age from 18 to 24 (M=21.05, S.D. 1.948). A chi-square goodness-of-fit examined the relationship between race and housed status. A two-way (factorial) analysis of variance (ANOVA) was conducted to determine whether there was an interaction effect between race and gender on the Vulnerability Index - Service Prioritization Decision Assistance Tool (VI-SPDAT) scores. A binary logistic regression was performed to ascertain the impact of other factors that may facilitate or attenuate housing status. A one-way between-subjects ANOVA was conducted to determine if there was a statistically significant difference in the average time (days) to housing by race. A Kaplan Meier survival analysis was performed to assess the relative disparate influence of several variables on time to housing.

Results: Results show a statistically significant association between race and housing status, χ²(1) = 4.274, p = .039.Results from a two-way ANOVA show no statistically significant interaction between gender and race on VI-SPDAT scores (F(3, 193) = 5.611, p = .084). However, results indicate a statistically significant difference in VI-SPDAT scores by race F (3, 193) = 11.318, p<.001. The logistic regression model was statistically significant, χ²(7) = 35.106, p < .001);with age (p = .022), gender (p <.001), and prioritization score (p=.009) all predictors of housing in youth. For every one-unit increase in age, the odds of housing increase. Being a female and assignment of a prioritization score significantly predicted being housed. Results from Kaplan Meier indicated that gender (p=.016) was the sole significant predictor of time to housing.

Conclusions and Implications: Analysis results suggest that gender – not race - is a significant factor in time to housing within the youth population. Males were 2.1 times less likely to be housed than their female counterparts. Despite evidence of racial bias in VI-SPADAT scores (Brown et al., 2018; Cronley, 2020), race did not predict housing placement or time to housing. Given consistent evidence of Black/African American youth’s over-representation within homelessness, future research needs to model other pathways out of homelessness and consider additional variables not captured in HUD-designated HMIS data, e.g., retention in housing, return to homelessness, etc. In addition, future research could test whether behavioral health needs are greater among male youth experiencing homelessness compared to females and if this contributes to lower housing placement.