What Is a Religious NGO?

This chapter addresses definitional and classificatory problems that emerge in collecting data on transnational religious NGOs (nongovernmental organizations). It shows how the parameters used to define “religious NGOs” are not only of practical, methodological importance but also speak to underlying theoretical concerns about the boundaries between the religious and the secular, and between government and civil society. Of particular significance are, first, organizational variation in the relationships to religious institutions proper and, second, international variation in religion-state relations. Taken together, these variations make it difficult to determine whether particular religious actors are best conceived as NGOs, religious organizations proper, or extensions of government. These variations can also compromise the reliability of data used for international and interreligious comparisons of “religious NGOs.”

Using the Total Survey Error Paradigm to Improve Cross-National Research on Religion

Cross-national research is an absolute necessity if we are to understand contemporary human societies in general and the role of religion in particular. To be useful, comparative survey research needs to meet high scientific standards of reliability and validity and achieve functional equivalence across surveys. This chapter examines how well-designed and well-executed cross-national/cultural survey research can best minimize error and maximize equivalence. The chapter first introduces the concept of total survey error (TSE), including interactions among the error components, its application when multiple surveys are involved, and comparison error across cross-national surveys. Second, obtaining functional equivalence and similarity in cross-national surveys is discussed. Third, the challenges of doing cross-national surveys are considered, as well as how combining traditional approaches for maximizing functional equivalence with the utilization of TSE can minimize comparison error and maximize comparative reliability and validity. Fourth, attention is given to minimizing comparison error in question wordings in general. Special attention is given to dealing with differences in language, structure, and culture. Lastly, specific issues relating to studying religion cross-nationally are examined.

Evaluating Survey Measures Using the ARDA’s Measurement Wizard

Developing survey measures of religion is difficult. As religion in America becomes more diverse and research projects become more global, finding a shared vocabulary that taps into concepts of interest and selecting response categories that capture the full range of choices become ever more challenging. This chapter introduces new tools and techniques for exploring and evaluating past survey measures of religion, including a new software tool (the Measurement Wizard) that draws on the Association of Religion Data Archives’ (www.theARDA.com) massive collection of survey measures. The chapter offers examples to demonstrate how these resources can be used to evaluate past survey measures of religion and improve future measures.

Conclusion

This chapter reviews the major lessons learned from the new approaches and technologies utilized by the authors of each chapter. First, the authors discuss the urgent need for new measures in the study of religion. Second, they note the need for using multiple methodologies for researching religion. Third, they discuss how rapid advances in methodology and technology allow for new approaches to research. Fourth, they note that, regardless of approach, the criteria for quality measures remain the same. Fifth, they discuss that many of the measures proposed in this volume are less intrusive than typical survey approaches. Finally, they conclude by discussing some of the challenges that await religion researchers.

Lessons Learned from SoulPulse, a Smartphone-Based Experience Sampling Method (S-ESM) Study of Spirituality

SoulPulse is an ongoing experience sampling method that uses participants’ smartphones to collect data over a two-week period. Launched in late 2013, it is a new approach for studying spiritual and religious experiences. In creating it, the authors had few resources to guide them, and so they went through a lot of trial and error. This chapter reviews some of the lessons learned in an effort to guide researchers wanting to use or evaluate this methodology. These lessons relate to how often to administer surveys and questions within surveys, creating suitable measures, designing an appealing software interface, recruiting participants, and analyzing the uniquely structured, multilevel data generated by this approach.

Introduction

Despite an avalanche of innovations in conducting data analysis on religion, the measures used to generate the data have undergone remarkably modest change. This book introduces innovative measures and methods that revisit old research findings and reveal new research frontiers. The introduction provides an overview of some of the challenges facing the measurement of religion and a summary of the book’s chapters.

Pathways to Discovery and Enlightenment

Online shopping has produced a wealth of new information in the form of recommendations based on items that are frequently purchased together. This data can be collected from shopping websites like Amazon.com and used to construct networks of books or other products and content. This chapter demonstrates how such data can be collected and visualized by anyone with Internet access and basic spreadsheet software, and how such analysis can inform other research. Using both individual and large network approaches, the authors explore patterns in the consumption of New Age/paranormal materials. Using these networks of purchases and recommendations, they are able to produce a picture of paranormal consumers that is both more comprehensive and more nuanced than previous efforts.

Historical Research

This chapter uses a well-known religious group of the nineteenth century to illustrate many of the ways historical data can be assembled, and many of the problems faced in using online sources to develop a coherent and theory-relevant picture of the past. Originally emerging in New England, the Oneida community was established in upstate New York in 1848, under the leadership of John Humphrey Noyes, a religious leader who believed he had achieved perfection and knew how to lead others to that goal. The example of Oneida allows us to explore with efficiency and clarity the possibilities for studying historical religious phenomena via online resources. Of necessity, this chapter often emphasizes details of research methodology, to alert the reader to problems and their possible solutions, but it also will consider how the data relate to larger theoretical issues.

From Documents to Data

There was a time when conducting research using documents meant spending months in a remote location sorting through filing cabinets in a basement. Today, though, websites, electronic databases, search tools, and digitization efforts have made organizational and governmental documents increasingly accessible. This chapter examines how such documents can be used in social science research examining religion using three illustrations from the author’s own work. One case looks at mission statements on congregational websites to assess variations in addressing openness to gay and lesbian individuals. Another example examines using newspaper archives to measure crimes against religious congregations. The third example illustrates the use of religious nonprofits’ tax returns to generate measures of finances, identity, and mission.

Assessing Measures of Religion and Secularity with Crowdsourced Data from Amazon’s Mechanical Turk

This chapter examines Amazon’s Mechanical Turk (MTurk), a crowdsourcing pool of potential participants for completing online tasks, as a resource for gathering novel data on religiosity and secularity (or other topics). MTurk provides an easily accessible, cost-efficient option for piloting new measures and conducting split-ballot experiments to assess measurement effects. The authors use MTurk data to evaluate measures of religious identity, demonstrating how question format can influence the percentage of respondents classified as religiously affiliated. They also use new measures to provide descriptive and analytical information on the rationales individuals give for being either religious or secular across different religious traditions and types of secularity. They conclude by outlining the opportunities and limitations of crowdsourcing data for exploring issues of measurement—as well as substantive areas of inquiry—in religion and beyond.