Moving homeless families into stable housing is an important policy priority, but little is known about how individual and household characteristics affect shelter use. This leaves homeless services providers and policy makers with few tools to make a priori decisions on which to base the targeting of services. Psychologists and economists have found that positive psychological capital leads to improvements in policy-relevant variables like academic achievement, income, and justice system recidivism; this is the first application of that framework to homelessness. This study measures three positive psychological traits – hope, resilience, and self-control – among 276 families who entered shelter in New York City, and uses survival analysis models to examine their relationships with the number of days subsequently spent in shelter and whether families who exited shelter returned. In addition, scores on these scales are compared to those of other populations. Two of the three traits are significantly associated with the duration of shelter use. On an 8-point hope scale, a one-point increase is associated with a reduction of 35 shelter days at p<.05. At the less stringent p<.10 threshold, a one-point increase on a 5-point resilience scale reduced shelter use by 32 days. Additionally, scores on the three scales were similar to those of non-homeless populations. These findings have implications for perceptions of homelessness and suggest solutions to address it. While the models did not accurately predict how long a family will stay in shelter or whether it will return after leaving, a growing body of research suggests that increasing hope and resilience through low-cost, low-burden interventions may reduce total shelter use. In addition, baseline comparisons to other populations suggest that homelessness is not associated with a deficiency of positive psychological attributes, which may be helpful in re-framing the discourse on factors associated with homelessness.