Collaborative Research: Religious Involvement and Mortality in the U.S.

合作研究Robert Hummer Richard Rogers Christopher Ellison Jason Boardman德克萨斯大学奥斯汀分校科罗拉多大学博尔德分校宗教在美国仍然是一个充满活力的文化和制度存在。随着宗教对美国社会的强烈和持续的重要性，了解宗教参与是否与死亡风险有关就像最近的一些研究发现的那样，尤其重要的是，如果是这样，确定这种关系的大小和机制。关于该主题的重大新工作已被广泛宣传，并在社会科学和医学文献中产生了重大影响。然而，这一领域的研究仍然存在很大争议，理论和方法上都不发达，缺乏深度。通过完成对一些现有数据集的二次分析，本研究将解决一些其他当前宗教-死亡率文献特征的局限性。具体来说，它将解决的问题是，选择性是否负责死亡率和宗教之间的联系。有许多因素可能影响宗教参与和死亡率，这些因素可以解释这种统计关联。然而，这可能是科学界需要解决的关键问题，以便了解宗教与死亡率的关系是否可以被视为因果关系或虚假关系。其次，只有非常有限的研究考虑了宗教参与对人口亚群体死亡率的不同影响。这种对当代美国多样化人口格局的忽视可能会导致一些实证研究结果，尽管这些发现在整个人口中可能是正确的，但却掩盖了人口内部的实质性异质性。第三，对受宗教影响的具体死亡原因的关注很少。第四，直到最近才有几个以人口为基础的数据来源，可以在国家一级调查宗教参与与死亡率之间的关系。因此，与社会学死亡率文献中的其他社会和人口因素（例如，社会经济地位、种族/民族、性别、年龄、婚姻状况）和对了解美国的健康和长寿至关重要的因素相比，关于宗教对健康和死亡率影响的知识显然有限。在医学文献中，关于医学院是否应该教授宗教和健康的课程，以及医生是否应该或不应该使用宗教健康-死亡文献作为指导他们与病人的关系的一个主要争论。然而，迄今为止的证据还太不确定，无法得出宗教参与促进健康的结论，尽管最近的研究显然倾向于这种方式。但是，在选择性问题得到充分解决之前，有关整个关系的虚假问题将继续存在。除了总体关系之外，受影响的特定社会和人口群体，以及宗教可能影响健康和死亡率的具体机制，现在才刚刚开始得到挖掘。因此，这一项目有可能阐明引起大量公众注意以及社会科学界和医学界注意的社会学问题。此外，在社会学中，从医学社会学到人口学再到宗教社会学，都对这一主题有浓厚的兴趣，这些领域的教职员工和学生都参与了研究并对研究结果感兴趣。其次，该项目将进一步促进德克萨斯大学和科罗拉多大学之间长期而富有成效的伙伴关系。

Collborative Research SES- 0243189 SES-0243249 Robert Hummer Richard Rogers Christopher Ellison Jason Boardman University of Texas at Austin University of Colorado at BoulderReligion remains a vibrant cultural and institutional presence in the US. With the strong and continuing importance of religion to American society, it is particularly important to understand whether religious involvement is related to mortality risks as some recent research finds and, if so, to determine the magnitude and mechanisms that characterize this relationship. Major new work on the topic has been widely publicized and has made a major impact in both the social science and medical literatures. Nevertheless, this area of research remains highly controversial, theoretically and methodologically underdeveloped, and lacking in depth. Through completing secondary analyses of a number of existing datasets this research will tackle a number of other current limitations that characterize the religion-mortality literature. Specifically, it will address the issue of whether selectivity is responsible for the association between mortality and religion. There are a number of factors that may influence both religious involvement and mortality that can account for this statistical association. However, this is perhaps the key question that needs to be addressed by the scientific community in order to understand whether the religion-mortality relationship may be considered causal or spurious. Second, only very limited research has considered the differential effects of religious involvement on mortality within subgroups of the population. This inattention to the diverse demographic landscape of the contemporary US can lead to empirical findings that, while potentially true among the whole population, mask substantial heterogeneity within the population. Third, only scant attention has been given to the specific causes of death that are influenced by religious involvement. Fourth, only recently have several population-based data sources become available that allow inquiry into the association between religious involvement and mortality at the national level. Thus, compared to other social and demographic factors that are central in the sociological mortality literature (e.g., SES, race/ethnicity, gender, age, marital status) and critical to the understanding of health and longevity in the United States, knowledge regarding religious influences on health and mortality is clearly limited. There is a major debate in the medical literature about whether medical schools should teach courses on religion and health and about whether physicians should or should not use the religion health-mortality literature as one of the guides toward informing their relationships with patients. However, the evidence to date is far too tentative to conclude that religious involvement promotes health, although it is clear that the recent seems to be leaning that way. But until the issues of selectivity are fully addressed, questions regarding the spuriousness of the overall relationship will remain. Beyond the overall relationship, the specific social and demographic groups that are influenced, and the specific mechanisms by which religion may influence health and mortality, are just now beginning to be tapped. Thus, this project has the potential to shed light on sociological issues that have captured substantial public attention, as well as the attention of both the social scientific and medical communities. Further, within sociology, there is substantial interest in this subject matter ranging from medical sociology to demography to the sociology of religion, with faculty members and students from each of these areas involved in the research and interested in the findings. Second, the project will further contribute to the long-standing and very productive partnership between the Universities of Texas and Colorado.