The Structure of Religion and Spirituality in a Diverse Sample of Adults in the U.S.: A Report on Exploratory and Confirmatory Factor Analyses in the Study on Stress, Spirituality, and Health.

Résumé

OBJECTIVE: This paper details psychometric analyses of religion and spirituality (R/S) measures in a diverse sample of U.S. adults in the Study on Stress, Spirituality, and Health (SSSH).METHODS: Participants (n = 16,372) completed the SSSH Baseline Survey which assessed a broad range of R/S practices, beliefs, and experiences. The survey was administered in six NIH-funded cohort studies. Four cohorts were homogeneous regarding participant race/ethnicity: , or . Two cohorts included predominantly participants. Given our goal of identifying a robust, parsimonious set of R/S measures for common use in U.S. cohort studies, we analyzed the structure of individual R/S items in a combined sample and disregarded whether items were part of a pre-existing scale.RESULTS: An Exploratory Factor Analysis (EFA) of 60 R/S items in a randomly selected half of the sample (1=8,186) revealed nine R/S dimensions; four were robust to retention criteria. The four remaining R/S dimensions were Positive Religion/Spiritualty and Coping, Negative Religious Coping/Spiritual Struggle, Spirituality as Meaning-Purpose-Connection, and Non-Theistic Daily Spiritual Experiences. A Confirmatory Factor Analysis of 41 retained items conducted in the second half of the sample (2=8,186) demonstrated that a correlated 4-factor model fit better than a unidimensional model or orthogonal (uncorrelated) 4-factor model.CONCLUSIONS: Future work is needed to confirm this model in other datasets, to demonstrate the external validity of these core R/S constructs in relation to a range of outcomes, and to finalize a robust but efficient set of R/S measures for use in U.S. cohort studies.