1) Based on large samples (e.g., thousands of participants) thus, there is sufficient statistical power to conduct high confidence comparisons between groups.
2) Nationally representative samples thus can be generalized to the broader U.S. population.
3) Examines (and/or control for) many different factors/variables (often 10+) thus provides a broad picture of “what is going on.”
4) Longitudinal method allowing comparison of measures across time to establish causality.

Two key limitations of this research include the realities that it:
1) Is often limited to reports from one person per family (Handel, 1996).
2) Is often limited to survey responses without fielding participants’ in-depth verbal responses, contextualizing comments, narratives, or explanations.

1) Based on small samples of fewer than 30 participants.
2) Based on racially and culturally homogenous samples without comparison groups (e.g., white Jewish women; see Davidman, 1991; Kaufman, 1993).
3) Conducted within a limited cultural/geographic area, based on researcher convenience.
4) Involves only one participant per family (Handel, 1996).
5) Conducted primarily or solely by one or two researchers, whose biases go unchecked [see Marks (2015) for criticism of the “sole scholar” or “monk/nun in a cell” approach to qualitative research].

Two key limitations of this research include the realities that it:
1) Often provides unique depth—but is limited by constrained sample size, limited scope, and questionable applicability.
2) Further, the above limitations leave most qualitative studies vulnerable to criticisms regarding validity, reliability, and replicability.

1) Based on a sample of 476 participants from 200 families collected through a replicable sampling strategy (recommendations from religious gatekeepers).
2) Heterogeneous sample characterized by: 2A) Rich racial/ethnic diversity (50%+ of the sample); including African American, Asian American, Latino, Native American, Middle Eastern, East Indian, and Pacific Islander families. 2B) Religious diversity (more than 20 Christian, Jewish, and Muslim denominations) 2C) Socio-economic diversity (educational experience ranged from GED-PhD/MD)
3) Based on a national (although not nationally representative) sample that includes participants from 17 states in all eight religio-cultural regions of the United States (see Silk & Walsh, 2009).
4) Includes in-depth, interview-based questioning of multiple participants per family [mother, father, child(ren)], consistent with Handel’s (1996) call for true “family” research.
5) Involved 100+ diverse researchers/coders to provide data audit trails, checks, balances, and inter-rater reliability for reported themes (see Marks, 2015 for discussion).
6) Face-to-face, in-depth interviews (2-4 hours) in the homes and worship places of families (as opposed to phone or online interviews).
7) The sample consists of “exemplars” of both religious involvement and family strength achieved by inviting religious gate-keepers to recommend known exemplars.
8) “Trans-Historical” in that data collection occurred from 2001-2015 Sample not limited by historical events such as 9-11, the housing collapse of 2008, natural disasters (e.g., hurricanes Katrina and Rita), and the presidential elections of 2004, 2008, and 2012.)