Saturday, 15 January 2005 - 12:00 PM

This presentation is part of: Poster Session II

Testing a Conceptual Model of Religiosity and Psychological Distress Among African Americans

Purpose: While a number of studies have looked at the relationship of religion and psychological distress among African Americans, there remains inconsistency in study findings. Studies reveal that subjective religiosity and private devotional activities are inversely related to psychological distress. Conversely, other studies show religious affiliation and congregational participation are not associated with psychological distress. Variations in how religion is measured make it difficult to ascertain the underlying processes by which religion is related to mental health outcomes. This study will investigate three comparative measurement models of religiosity, using structural equation modeling to assess whether the underlying latent constructs of religiosity are separately more significant predictors of psychological distress, than the 2-dimensional model.

Methods: The religion subscale within the National Survey of Black Americans (NSBA) survey, Wave 4 (1992) was used to conduct this analysis. The NSBA sample, based on the 1970 United States Census, was drawn using multi-stage area probability sampling. A total of 659 respondents were interviewed for wave 4. 349 cases are available after imputation and listwise deletion. Using LISREL and Maximum Likelihood estimation, the measurement and final structural equation models were examined. Given the standardized factor loadings of each measurement model and the fit statistics, only a subset of indicators was used to measure the constructs of interest. Analysis of the measurement model revealed satisfactory fit with six of the twelve psychological distress indicators (c ²= 32.25, df = 9, GFI = .95, AGFI =.87). Equally, seven of the seventeen questions used to assess religious beliefs, values, or experiences, were necessary to specify the model (c ²= 27.68, df = 14, GFI = .93, AGFI =.86). Confirmatory factor analysis revealed two underlying constructs of religiosity: Organizational (4 items) and Non-organizational (3 items). The former model was satisfactory fit and the latter was saturated. Results: Each of the three competing structural equation models was specified. The better fit model included only the organizational dimension of religiosity as a predictor of psychological distress (c ²= 72.93, df = 34, NFI = .95, NNFI = .95, GFI = .94, AGFI =.90). Even though the results of the analysis of all equation models revealed a inverse relationship between religiosity and psychological distress, all relationships were not significant. Based on previous findings it was expected that non-organizational religiosity would be a better predictor of psychological distress than organizational or a 2-dimensional measure. Yet, the findings of this study showed inconsistent results.

Implications for Practice: It is yet unclear based on these findings which underlying constructs of religiosity are related to psychological distress. Consistent with previous literature organizational religiosity is not related to psychological distress, but contrary to the literature no relationship was found between non-organizational (private) religiosity and psychological distress. As demonstrated by existing studies, there is no consensus on how to measure the underlying construct of African American religiosity; therefore inconsistency in findings may be due, in part, to measurement variation or error. Consequently, consensus in religiosity measures might improve our understanding of how this construct contributes to psychological distress.

See more of Poster Session II
See more of Oral and Poster

See more of Celebrating a Decade of SSWR (January 13 - 16, 2005)