David E. Eagle

Methodological considerations in the use of name generators and interpreters

Abstract With data from the Clergy Health Initiative Longitudinal Survey, we look for interviewer effects, differences between web and telephone delivery, and panel conditioning bias in an “important matters” name generator and interpreter, replicated from the U.S. General Social Survey. We find evidence of phone interviewers systematically influencing the number of confidants named, we observe that respondents assigned to the web survey reported a larger number of confidants, and we uncover strong support for panel conditioning. We discuss the possible mechanisms behind these observations and conclude with a brief discussion of the implications of our findings for similar studies.

Religion and Inequality: The Role of Status Attainment and Social Balance Processes

Religion is an important determinant of social and economic inequality, but the mechanisms that underlie this relationship are not well-understood. Early scholars recognized this connection, but their ideas do not adequately explain contemporary stratification patterns. Recent research documents robust empirical relationships between religion and material outcomes but has not yet begun to identify causes of these patterns. We fill this gap by providing a theoretical explanation of the religion-inequality link that synthesizes ideas from early and recent sociology. We propose that the process is inherently multilevel. We draw on ideas from status attainment theory to develop a micro-model and ideas from social balance theory to aggregate the Social Thought & Research 142 model’s outcomes. The synthesis of ideas from these theoretical traditions provides a unique, and potentially useful way to understand the relationship between cultural orientation and material resources. Religion plays a central role in creating and maintaining social and economic inequality, but the mechanisms driving this relationship are not well-understood. Webers The Protestant Ethic and the Spirit of Capitalism (2002) contains the most famous early argument for a link between religion and inequality, but the ideas that he and other early theorists proposed have garnered significant critique. At the heart of Weber’s thesis is the contention that changes in the religious orientation of Europeans through the influence of Protestantism, especially Calvinism, led to fundamental changes in the economic system. Scholars have challenged Weber’s thesis on several grounds. Economic historians have argued that in many parts of Europe, the transition to capitalism preceded the emergence of reformed Protestantism (Bainton 1985; Samuelsson 1964). Therefore, it is equally plausible that the emergence of new forms of economic behavior produced changes in religious beliefs, or that religious and economic change are interdependent and often difficult to disentangle. Coleman (1990) criticizes Weber for positing a theory that claims individual beliefs straightforwardly aggregate into societal-level values. Such a simple process fails to account for how stratification develops, and how individual-level beliefs aggregate into different cultural orientations depending on one’s social position. Even if we disregard these internal problems, the religious landscape and the processes that account for stratification have changed considerably since Weber and his contemporaries developed their ideas. Looking only at patterns in North America, these theories cannot accommodate the decline of so-called mainline Protestant denominations, the proliferation of conservative Protestant groups, the impact of immigration on the composition of the Catholic Church, and the increased presence of other religious traditions. It has also become increasingly clear that the relationship between religion and stratification is not a function of large-scale shifts in the mode of production but rather reflects Religion and Inequality 143 changing individual and group approaches to human capital, work, entrepreneurship, saving, and investing. In the 1960s, researchers revived questions about religion and inequality. However, these debates lost momentum when the convergence of the socioeconomic status (SES) of mainline Protestants and Catholics came to dominate the literature (Glenn and Hyland 1967; Lenski 1961; Roof and McKinney 1987). The more recent years have witnessed something of a renaissance in the study of religion and inequality. However, although contemporary research is empirically rich, it has paid relatively little attention to the causal processes linking religion and material outcomes. Modern data provide powerful evidence that religion affects education for adults (Chiswick 1988; Darnell and Sherkat 1997; Lehrer 1999b, 2004) and adolescents (Muller and Ellison 2001; Sherkat and Darnell 1999), gender roles in the home (Ellison and Bartkowski 2002; Read 2004) and in the labor market (Lehrer 1999a, 2000; Sherkat and Darnell 1999), fertility and family formation (Glass and Jacobs 2005; Lehrer 1996b, 1996c; McQuillan 2004), wages (Keister 2010; Smith and Faris 2005; Steen 1996; Wilder and Walters 1998), work and occupational outcomes (Smith and Faris 2005), and saving behavior and wealth (Crowe 2008; Keister 2003, 2007, 2008). Researchers have understandably taken advantage of modern data and methods to provide careful empirical support for key relationships that eluded prior generations of scholars. However, the resulting body of research contains a large amount of evidence for bivariate relationships but minimal discussion of how the interactions among variables produce social stratification. The theoretical discussion accompanying most of the empirical evidence contains an implicit status attainment model but rarely articulates or demonstrates causal processes linking religion and inequality. Moreover, contemporary researchers do not always adequately attend to the problems inherent in using individual-level data to draw inferences about group-level outcomes. Our objective is to provide a contemporary theoretical explanation of how religion affects inequality. The first step in understanding this relationship is clarifying that there are separate but related processes operating at the individual (micro) and group (macro) levels. This starting point allows us to integrate ideas from Social Thought & Research 144 early and contemporary research while moving beyond the challenges faced in both. We propose that micro-level processes connect religion to individual and family material well-being following a status attainment logic. We articulate the details of this model, discuss how important behaviors and processes are interrelated, and address how various processes interact to affect inequality. We pay particular attention to integrating the large amount of empirical evidence available from contemporary scholarship. Next, we show that ideas from social balance theory can explain how individual and family processes combine to produce group-level patterns. Finally, we specify the details of the balance model and discuss the implications for aggregate indicators of social inequality. Religion and Social Stratification: Two Levels of Aggregation Most efforts to explain how religion affects social stratification attempt to isolate either individual or group outcomes. Effective isolation of this sort is challenging because, under most circumstances, it requires focusing on one level of aggregation without invoking behaviors or outcomes at the other level. Early theorists attempted to isolate macro-level outcomes by addressing how national patterns of religious affiliation affected national patterns of economic organization (Sombart 1911; Weber 2002). Only a limited number of cases exist at the macro level, making comparison difficult. Additionally, it is difficult to collect data at the macro-level. Thus, in most cases, early theorists resorted to discussing how individual behaviors lead to aggregate patterns (Coleman 1990). More recent research focuses more narrowly on micro-level outcomes (e.g., individual income, family wealth) because these processes are essential to understanding contemporary stratification processes. Moreover, data are often collected at the level of individuals or other levels below that of the social system. Although contemporary explanations focus on micro-level explanations of micro-level outcomes, it is common to invoke macro-level causes (e.g., neighborhood, generational effects) or to draw conclusions about macro-level issues (e.g., group differences in income, work, wealth) from micro-level data without carefully specifying how the levels interact (Coleman 1990). Social Thought & Research 146 ideology to micro-level material outcomes (Figure 2). Status attainment refers to the process by which individuals arrive at socioeconomic standing over their lives, and the status attainment approach has become one of the most widely used theoretical perspectives in sociological research on social and economic wellbeing. Work in this tradition specifies individual traits or attributes, usually family background, education, and work behaviors that contribute to adult attainment. Although other theoretical models may potentially relate religious ideology and micro-level material outcomes, a growing body of related empirical evidence supports the status attainment model. Equally important, the status attainment model includes most of the important behaviors and processes that affect attainment, allows for interactions among these processes, effectively describes how family characteristics affect adult outcomes, and retains the appropriate time-ordering (e.g., family background affects education, which affects adult occupation). Because the status attainment model incorporates both individual/family and contextual influences on micro-outcomes, it also efficiently and accurately represents the macro-micro and micro-micro links pictured in Figure 1. In the status attainment model, childhood religion refers to affiliation, belief, practice, congregation traits, and the broader religious environment. Affiliation is the religious denomination or tradition with which an individual is associated. In the United States, persuasive evidence shows that members of religious groups behave similarly in consequential ways (Burstein 2007; Glass and Jacobs 2005; Keister 2008; Lehrer 2009; Smith and Faris 2005). A long history of research shows that affiliation with mainline Protestant, conservative Protestant, black Protestant, Roman Catholic, Jewish, or other religious groups affects wellbeing (Herberg 1983). The proliferation of Protestant denominations in recent decades and fluctuations in the sizes of some denominations and traditions suggests that it is becoming increasingly im

A 2-Year Holistic Health and Stress Intervention: Results of an RCT in Clergy

INTRODUCTION This study sought to determine the effect of a 2-year, multicomponent health intervention (Spirited Life) targeting metabolic syndrome and stress simultaneously. DESIGN An RCT using a three-cohort multiple baseline design was conducted in 2010-2014. SETTING/PARTICIPANTS Participants were United Methodist clergy in North Carolina, U.S., in 2010, invited based on occupational status. Of invited 1,745 clergy, 1,114 consented, provided baseline data, and were randomly assigned to immediate intervention (n=395), 1-year waitlist (n=283), or 2-year waitlist (n=436) cohorts for a 48-month trial duration. INTERVENTION The 2-year intervention consisted of personal goal setting and encouragement to engage in monthly health coaching, an online weight loss intervention, a small grant, and three workshops delivering stress management and theological content supporting healthy behaviors. Participants were not blinded to intervention. MAIN OUTCOME MEASURES Trial outcomes were metabolic syndrome (primary) and self-reported stress and depressive symptoms (secondary). Intervention effects were estimated in 2016 in an intention-to-treat framework using generalized estimating equations with adjustment for baseline level of the outcome and follow-up time points. Log-link Poisson generalized estimating equations with robust SEs was used to estimate prevalence ratios (PRs) for binary outcomes; mean differences were used for continuous/score outcomes. RESULTS Baseline prevalence of metabolic syndrome was 50.9% and depression was 11.4%. The 12-month intervention effect showed a benefit for metabolic syndrome (PR=0.86, 95% CI=0.79, 0.94, p<0.001). This benefit was sustained at 24 months of intervention (PR=0.88; 95% CI=0.78, 1.00, p=0.04). There was no significant effect on depression or stress scores. CONCLUSIONS The Spirited Life intervention improved metabolic syndrome prevalence in a population of U.S. Christian clergy and sustained improvements during 24 months of intervention. These findings offer support for long-duration behavior change interventions and population-level interventions that allow participants to set their own health goals. TRIAL REGISTRATION This study is registered at www.clinicaltrials.gov NCT01564719.

Household Charitable Giving at the Intersection of Gender, Marital Status, and Religion:

Past research reveals mixed results regarding the relationship between gender and charitable giving. We show gender plays a significant role in giving but only when considered alongside marital status and religion. Using the 2006 Portraits of American Life Study, we model a household’s propensity to give and the amount given. We extend past research by disaggregating unmarried households to look at divorced, widowed, and never-married households, and by including multiple religion measures. Results indicate households headed by never-married females have lower giving levels compared with those headed by divorced and widowed women. In households headed by single males, these differences are largely absent. Religious attendance has a stronger association with giving in male-headed households. The respondent’s gender is also related to the amount married households report giving to charity. Future research on giving should consider both gender and marital status to more fully capture increasing diversity in American families.

Disentangling Race and Socioeconomic Status in Health Disparities Research: an Examination of Black and White Clergy

AimSophisticated adjustments for socioeconomic status (SES) in health disparities research may help illuminate the independent role of race in health differences between Blacks and Whites. In this study of people who share the same occupation (United Methodist Church clergy) and state of residence (North Carolina), we employed naturalistic and statistical matching to estimate the association between race—above and beyond present SES and other potential confounds—and health disparities.MethodsWe compared the health of 1414 White and 93 Black clergy. Then, we used propensity scores to match Black and White participants on key socioeconomic, demographic, occupational, and physical activity characteristics and re-examined differences in health.ResultsPrior to propensity score matching, Black clergy reported worse physical health than their White counterparts. They had greater body mass index, higher prevalence of diabetes and hypertension, and lower physical health functioning. White clergy reported less favorable mental health. They had higher severity of depression and anxiety symptoms as well as lower quality of life and mental health functioning. Propensity score analysis revealed that matching on SES and other key variables accounted for most, but not all, of the observed racial differences. Racial disparities in hypertension, depression severity, and mental health functioning persisted despite adjustments.ConclusionsRace contributed to health disparities in some outcomes in our study population, above and beyond our measures of participants’ present SES and key demographic, occupational, and physical activity variables. This study provides evidence supporting the position that race contributes to health disparities through pathways other than SES.

Perceived social support, received social support, and depression among clergy

We argue that perceived support is best conceptualized more as a measure of how an individual appraises his/her situation rather than a true reflection of how much support he/she receives. To test this theory, we used survey data from the Clergy Health Initiative Panel Survey to examine the relationship between perceived and received social support and their association with depressive symptoms in clergy (N = 1,288). Overall, analyses revealed perceived support had a weak association with received support. Greater perceived support had a significant relationship with lower depressive symptoms. In contrast, greater received support had only a small relationship with lower depressive symptoms, which was fully mediated by perceived support. Our results raise questions about the effectiveness of many clergy social support interventions, which often aim to boost the quality and/or quantity of received social support. We suggest it may be more advantageous to boost perceptions of social support, possibly through cognitive reframing or positive mental health interventions.

Prospective Associations Between Depressive Symptoms and the Metabolic Syndrome: the Spirited Life Study of Methodist Pastors in North Carolina

BackgroundMetabolic syndrome (Met-S) has a robust concurrent association with depression. A small, methodologically limited literature suggests that Met-S and depression are reciprocally related over time, an association that could contribute to their overlapping influences on morbidity and mortality in cardiovascular disease, diabetes, and cancer.PurposeUsing a refined approach to the measurement of Met-S as a continuous latent variable comprising continuous components, this study tested the prospective associations between Met-S and depression.MethodsThis study of 1114 clergy included four annual assessments of depressive symptoms and Met-S components. Standard methods were used to measure Met-S risk factors, and the Patient Health Questionnaire-8 was used to assess depressive symptoms. We used confirmatory factor analysis to verify the structure of Met-S and depression and structural equation modeling to quantify the prospective relationships.ResultsThe statistical models confirmed the validity of quantifying Met-S as a continuous latent variable, replicated previous evidence of a concurrent association, and indicated a significant prospective association of initial depressive symptoms with subsequent Met-S. Initial Met-S was at most only weakly associated with subsequent depressive symptoms, and the former prospective effect was significantly larger. Associations of depressive symptoms and Met-S were significant for both men and women, but somewhat stronger among men.ConclusionsResults support representation of Met-S as a continuous latent variable. The association of initial depressive symptoms with later Met-S suggests that interventions addressing these correlated risk factors may prove useful in preventive efforts.

The Negative Relationship between Size and the Probability of Weekly Attendance in Churches in the United States

Are people less likely to attend large churches? Using nationally representative data, I find a negative relationship between size and the probability of attendance for Conservative, Mainline, and black Protestants and for Catholics in parishes larger than 500 attenders. The addition of control variables does not change these patterns. These results support the theory that group cohesion lies at the heart of the size-participation relationship in churches. With the rapid growth of megachurches in the United States, a negative relationship between size and frequency of attendance could serve to accelerate aggregate declines in attendance. Methodologically, this article demonstrates how Bayesian hierarchical models can handle complex data structures and imputation procedures in a unified model.