Oxford House Recovery Home Data Over Time

Methods
Data are provided from 627 residents of Oxford Houses, which are a network of self-run recovery homes. Data were collected in a longitudinal study that occurred over a two-year period. When participants entered the recovery homes, they completed the baseline interview. Every four months, participants were interviewed for as long as they lived in these recovery homes.

Baseline Measures
We provide the following surveys that were taken at the first baseline session participants completed: the demographic Survey; the Addiction Severity Index (Mclellan et al. 1992); the Form 90 (Miller, 1996); the Drug Taking Confidence Questionnaire (Annis & Graham, 1988); the Psychological Sense of Community Scale (Jason, Stevens, & Ram, 2015); the Perceived Stress Scale-Brief Version (Cohen, Kamarck, & Mermelstein, 1983); the AA Affiliation instrument (Humphreys, Kaskutas, & Weisner, 1998); the Rosenberg Self-Esteem Scale (Rosenberg, 1965); the Important People Inventory; the World Health Organization Quality of Life Assessment Brief Version (WHOQOL Group, 1998); the Snyder State Hope Scale (Snyder et al., 1996; Stevens, Buchannan, Ferrari, Jason, & Ram, 2014); the BFl-10 Personality Scale (Rammstedt, & John, 2007); and the Recovery Home Processes Questionnaire (included in interviews with the house presidents only).

Repeated Measures
Following the baseline wave, similar measures were administered. At these waves, we used the Addiction Severity Index-Lite (Only the Employment/Support Status, Alcohol/Drugs, and Legal Status sections).

Outcome Variable
We found a single latent outcome factor fitting the data, or a recovery factor behavior variable. Recovery factors across waves are located in the dataset (“recoveryfactor.1”; “recoveryfactor.2”; etc.). For details related to outcome variables and the original authors’ finding of a single latent recovery factor, see: Jason, L.A., Guerrero, M., Salomon-Amend, M., Stevens, E., Light, J.M. and Stoolmiller, M. (2021), Context Matters: Home-level But Not Individual-Level Recovery Social Capital Predicts Residents’ Relapse. Am J Community Psychol, 67: 392-404. https://doi.org/10.1002/ajcp.12481

Datasets
Some subscales or specific questions have been removed from the dataset for deidentification purposes. Reference “Data Dictionary Codebook” for names, labels, and attributes of variables, and “Baseline Interview” (in “Interviews” folder) for a copy of the survey instrument distributed to participants at baseline.