Re-evaluating the notion that resilience is commonplace: A review and distillation of directions for future research, practice, and policy.

The majority of multi-wave studies examining resilience in adulthood have involved growth mixture modeling (GMM). We critically evaluate the central conclusion from this body of work that "resilience is commonplace". Our emphasis is on two questionable methodological assumptions underlying this conclusion: (1) the variances are the same across trajectories (i.e., homogeneity of variance) and (2) the amount of change does not differ across individuals (i.e., slope variances are zero). Seventy-seven empirical studies were included that used GMM to examine resilience to diverse adversities in adulthood. Of these 77 relevant studies, 66 (86%) assumed homogeneity of variances across trajectories and 52 (68%) set slope variances to zero; in the minority of studies where these assumptions were not applied (particularly the homogeneity of variance assumption), the resilient trajectory was among the smallest. Furthermore, 63 (82%) of the 77 studies conferred labels of resilience based on a single outcome, which is problematic as resilience is never an "across-the-board" phenomenon. Based on our conclusions, we discuss three important directions for future research: (1) replication across samples and measures, (2) illumination of processes leading to resilience, and (3) incorporation of a multidimensional approach. We conclude by outlining a resilience framework for research, practice, and policy.