Mediation of Psychological Capital in Youth Experiencing Homelessness.

The reported analysis focused on mediation of risk factors through psychological capital, not on whether the intervention directly changed behavior. The model fit was poor for both the intervention group and the combined sample. The combined sample included both intervention and services-as-usual participants, which may have obscured intervention-specific mediation; the authors then analysed the intervention group only. Psychological capital was measured with added social connectedness and gratitude, and with domain-specific self-efficacy, not established for this population. Important external factors of homelessness, such as victimization, hunger, fear and pain, were not measured in the model. The sites differed in service availability and participant composition, and posttest assessments were higher in Ohio than Texas. The intervention was delivered by graduate research assistants, not routine clinical staff.

The supplied text does not include an author conflicts-of-interest declaration. The publication types list NIH extramural research support.

Do not read the null mediation as proof that the intervention does not work. This analysis tested mechanisms, not direct intervention efficacy, and the model fit poorly.