PROJECT SUMMARY There is growing consensus that recovery extends beyond remission to broader biopsychosocial improvements in domains such as emotional health and community integration (Kelly & Hoeppner, 2014; Richardson et al., 2019; White, 2012). Qualitative (e.g., Borkman et al., 2016) and quantitative studies of between-persons differences (Garner et al., 2014; Lusczakoski et al., 2014) appear consistent with broader recovery definitions. These definitions describe a process, however, and research has not yet addressed the validity of the broader recovery construct at the within-persons level. This is crucial because evidence that domains included in recovery definitions converge at the between-persons level does not guarantee that within- persons change is similarly patterned. Moreover, it is not clear if adult-based conceptualizations of recovery can be applied to youth (Finch et al., 2020) or if recovery has the same meaning across race/ethnicity, sex, recovery pathways, and disordered vs. normative use. Our long-term goal is to facilitate efforts to improve recovery- oriented systems of care. The overall objective for this R21 application is to evaluate the validity of the recovery construct across nearly 25 years from adolescence to mid-adulthood using an exploratory, data-driven approach. Our central hypothesis is that domains included in recovery definitions are moderately correlated at the between- persons differences level, exhibit some important differences in empirical meaning across subgroups, and are significantly correlated within persons. The rationale for the current project is that it informs operational definition of recovery from adolescence to mid-adulthood and provides a point of comparison for future R01 level efforts to validate recovery measures in late adulthood. This study uses data from the National Longitudinal Study of Adolescent to Young Adult Health, longitudinal measurement invariance analysis, and parallel process latent growth modeling to achieve the following specific aims: 1) Determine the importance of time and subgroup membership in moderating the meaning of recovery scores. 2) Determine whether domains included in recovery definitions are linked within persons. It is our expectation that this study will determine the validity of the alcohol use disorder (AUD) recovery construct from adolescence to mid-adulthood, identifying whether recovery indicators function in the same way across groups and intercorrelate within persons. This contribution is significant because it informs development of measures that capture the multidimensional and dynamic nature of the AUD recovery process. This project is innovative because it uses a longer-term within-persons design, includes data from adolescence, conducts longitudinal measurement invariance analyses of recovery moderators, evaluates criterion validity via alcohol use within persons, and analyzes national data spanning adolescence to mid-adulthood. These results ...