Predictors of past and current homelessness

Multiple regression analysis was used to predict both the length of time since first homeless and the length of the current episode of homelessness. Predictor variables measuring lack of human capital were education and income. Social alienation predictors included prior homelessness, marital status, and childhood placement. Psychiatric predictors included prior hospitalization and level of functioning. Age, race, gender, and military veteran status were also included in the analyses. The hypothesized causes of homelessness (lack of human capital, social alienation, and psychiatric pathology) explained only a small proportion of the variance of both dependent variables. Age was the most powerful predictor.