November 20, 2015 – Depression and Mood Disorder Among African American and White Women-Reply
In Reply We agree that our results are based on a small number of respondents who met criteria for major depressive disorder and mood disorders. We were transparent about this and noted sample size as the study’s primary limitation. As stated in our article,1 the National Survey of American Life (NSAL) remains one of the best data sets for this study. We considered using National Comorbidity Survey Replication data; however, it contained a limited number of rural cases. Given the sample size limitation, our article1 emphasized the importance of future research examining major depressive disorder and mood disorders among rural African American individuals with other samples.
Second, Keyes and colleagues’ statement that “the national sample weights cannot be assumed to be appropriate for approximating regional-specific estimates…” is not a valid statement for the NSAL. Because the nonresponse adjustment factors in the NSAL weights are computed separately for NSAL respondents from the South region and primary stage sampling units with the design strata representing the South, the weights (including the implicit nonresponse) are uniquely determined for the South and the weight computation process is not influenced by the response experience or the patterns of depression in other regions of the country. We are confident that the NSAL weights are computed “appropriately” for both national estimates and analyses for subpopulations.
Third, Keyes et al criticize us for speculating on our findings. However, that is the purpose of a Discussion section. They minimize the importance of religiosity as a potential mechanism to explain lower rates of major depressive disorder among African American women. However, research consistently reports higher rates of religiosity among African American individuals,2 and that religiosity is protective for depression.3Krause’s4 extensive body of work suggests religion and church-based social support are protective of mental and physical health, including mortality. Most of his studies demonstrate that older African American individuals are more likely than older white individuals to reap mental and physical health–related benefits of religion owing to higher levels of religious participation and church-based support networks. Additionally, Reese et al5 found that when controlling for religious service attendance, the black-white difference in depression was no longer significant. Clearly, our speculation about the importance of religious participation is not “fraught with problematic assumptions lacking empirical support” as stated by Keyes et al.
Last, Keyes et al argue that our focus on religiosity and social ties risks minimizing “the role of structural, economic, and sex discrimination experienced by African American women, while potentially reifying stereotypes.” This is by no means the case. In fact, much of the contemporaneous and historical research on black-white differences in the social sciences has taken a deficit approach and has not adequately considered African American women’s strengths. That is, despite a history of slavery and Jim Crow laws, high levels of poverty, segregated and low-quality education, and substandard housing, African American women in the rural South continue to cope. We take a strengths-based approach and argue that low rates of depression among rural African American women are potentially owing to their high levels of religiosity and their family ties.