A Latent Class Analysis of Personality Traits in Adults Experiencing Homelessness

This study distinguishes clinically and theoretically meaningful subgroups of people experiencing homelessness based on their endorsement of personality difficulties, using nationally representative data of the civilian, noninstitutionalized population of the United States, inclusive of those reporting past-year homelessness (<i>N</i> = 704). A bias-adjusted three-step latent class analysis was used to estimate latent class measurement models; classify cases into the optimal class solution; and, using a maximum likelihood method, test the association between demographic and behavioral health covariates with class membership. Results show that the four-class solution was optimal. The largest class (35.44%) had high probability of endorsing each personality difficulty and had high rates of behavioral health disorders. The second class (26.51%) had higher levels of antisocial traits and greater probability of endorsing substance use disorders relative to third and fourth classes. The third-largest class showed minimal personality difficulties (24.40%) and had the lowest probability of meeting criteria for each behavioral health disorder considered. The final class showed high levels of relational instability and identity diffusion (13.65%) and had higher levels of mood and anxiety disorders and suicide attempt relative to second and third classes. In conclusion, personality difficulties are commonly endorsed by adults experiencing homelessness and show differential relationships to behavioral health conditions.