- Introduction
- Chapter 1 Frameworks for Studying Religion with Data
- Chapter 2 Operationalizing Belief, Practice, and Affiliation
- Chapter 3 Ethics, IRB, and Sacred Data Stewardship
- Chapter 4 Survey Design: Modes, Frames, and Logistics
- Chapter 5 Sampling, Nonresponse, and Post-Stratification
- Chapter 6 Questionnaire Design and Scale Construction
- Chapter 7 Data Management and Reproducible Workflows
- Chapter 8 Exploratory Data Analysis for Religious Data
- Chapter 9 Descriptive Statistics and Visualization
- Chapter 10 Statistical Inference: Estimation and Uncertainty
- Chapter 11 Generalized Linear Models for Religious Outcomes
- Chapter 12 Multilevel Models for Congregations and Contexts
- Chapter 13 Causal Inference: Experiments and Quasi-Experiments
- Chapter 14 Time Series and Seasonal Cycles in Religious Life
- Chapter 15 Text as Data: Sermons, Scriptures, and Social Media
- Chapter 16 Computational Approaches to Digital Trace Data
- Chapter 17 Network Analysis of Communities and Congregations
- Chapter 18 GIS Basics: Projections, Layers, and Spatial Joins
- Chapter 19 Mapping Religious Landscapes and Accessibility
- Chapter 20 Spatial Statistics: Autocorrelation, Hotspots, and Kriging
- Chapter 21 Spatial Econometrics and Hierarchical Spatial Models
- Chapter 22 Remote Sensing and Nighttime Lights for Religious Infrastructure
- Chapter 23 Small Area Estimation and Data Fusion
- Chapter 24 Machine Learning for Prediction and Classification
- Chapter 25 Communicating Results: Dashboards, Reports, and Decision Support
Data and Devotion: Quantitative Methods for the Study of Religion
Table of Contents
Introduction
Religion is both intensely personal and profoundly social. It shapes identity, community, migration, health, and politics, while also guiding daily rituals, charitable work, and long-horizon visions of meaning. This book starts from a simple conviction: careful measurement and analysis can illuminate these dynamics without reducing their complexity or diminishing their significance. By pairing devotion with data—reverence with rigor—we can describe patterns, test ideas, and support decisions that affect congregations and the broader public sphere.
Data and Devotion is a practical methods book. It introduces researchers to survey design, spatial analysis, and computational approaches for studying religion at scale. Each chapter balances conceptual foundations with hands-on guidance, offering reproducible examples and short code snippets to help you move from question to insight. We emphasize workflows that a solo analyst, a research team, or a faith-based organization can sustain: from gathering and cleaning data to modeling, mapping, and communicating results.
A central theme of the book is measurement. Concepts like belief, practice, affiliation, and belonging do not spring ready-made from a dataset; they must be operationalized thoughtfully. We discuss how to craft items and scales, diagnose reliability and validity, and interpret indicators alongside qualitative knowledge and community expertise. You will learn to translate theories of religious change into variables, instruments, and designs capable of testing hypotheses and informing practice.
Ethics is not an afterthought here; it is an organizing principle. We treat religious information as “sacred data” that deserves special care. The book addresses consent, privacy, and de-identification; risks and benefits for participants and communities; bias and fairness in sampling and modeling; and the responsibilities that come with publishing maps and dashboards. Throughout, we encourage collaborative approaches—partnering with congregations, denominational offices, NGOs, and local leaders—to ensure that research questions are relevant and results are actionable.
The analytic toolkit unfolds progressively. We begin with exploratory data analysis and visualization, then move through estimation, generalized linear models, multilevel modeling for individuals nested within congregations and places, and designs for causal inference. Because religious life ebbs and flows with calendars and seasons, we also introduce time-series tools for identifying cycles and structural change. At each step, examples are tailored to common questions in the study of religion, such as attendance dynamics, conversion and disaffiliation, clergy supply, and program evaluation.
Place matters for religion, so a significant portion of the book is geographic. We cover GIS basics—projections, layers, and spatial joins—before turning to thematic mapping, measures of accessibility to worship spaces and services, spatial autocorrelation, hotspot detection, and spatial econometrics. We also explore remote sensing and related proxies to understand built environments and infrastructural patterns that correlate with religious presence and activity, always with attention to the limits of inference and the ethics of visibility.
Modern research also means working with text and traces. You will learn methods for analyzing sermons, mission statements, and social media; approaches to digital trace data and event logs; and network techniques for mapping relationships among people, congregations, and organizations. We complement these with machine learning for prediction and small area estimation to generate local insights where direct measurement is sparse—while foregrounding interpretability, uncertainty, and the dangers of overreach.
Reproducibility ties the whole enterprise together. We present a pragmatic workflow that uses literate programming, version control, and environment management so that analyses can be re-run, audited, and extended. Code snippets in popular open-source tools are kept concise and annotated, with pointers on adapting them to your own context. Templates for surveys, data dictionaries, metadata, and reporting help you standardize practice and scale your efforts.
This book is written for social scientists, students, journalists, and data professionals; for denominational researchers and staff at religious nonprofits; and for civic planners and public officials who must understand the religious dimensions of local life. Whether your goal is to test a theory, evaluate a program, allocate resources, or communicate with stakeholders, the chapters ahead equip you to make evidence-based decisions with humility, transparency, and care. Our hope is that you will finish not only with new tools, but with a renewed sense of responsibility for how those tools are used.
CHAPTER ONE: Frameworks for Studying Religion with Data
Understanding religion through a quantitative lens might, at first glance, seem like a stark departure from traditional humanistic approaches. Historically, the study of religion has often emphasized textual analysis, theological interpretation, ethnographic immersion, and philosophical inquiry. These methods excel at capturing the nuanced, experiential, and culturally specific dimensions of faith. However, they sometimes struggle to identify broad patterns, track changes over time at scale, or rigorously test hypotheses about the factors influencing religious life across diverse populations. This is where data-driven approaches offer a powerful complement, not a replacement.
The aim of this chapter is to lay the theoretical and methodological groundwork for integrating quantitative methods into the study of religion. We will explore various frameworks that bridge the qualitative-quantitative divide, demonstrating how empirical data can enrich our understanding of religious phenomena without sacrificing their inherent complexity. Far from reducing religion to mere numbers, these frameworks help us see both the forest and the trees—the sweeping trends alongside the deeply personal narratives. We begin by considering some core challenges in defining and measuring religion, then move to established social scientific theories that lend themselves to quantitative investigation, and finally discuss how data can illuminate the dynamic interplay between individual belief, communal practice, and societal influence.
One of the initial hurdles in any empirical study of religion is the very definition of the subject itself. Is religion a set of beliefs, a collection of practices, a form of social organization, a cultural system, or a personal experience of the divine? Scholars have grappled with this question for centuries, offering substantive definitions that focus on specific content (e.g., belief in a god or gods) or functional definitions that emphasize the role religion plays in individual and social life (e.g., providing meaning, fostering community, regulating behavior). Each definition carries implications for how we choose to measure religion. A substantive definition might lead us to survey specific theological affirmations, while a functional definition might prompt questions about how individuals find purpose or social connection. The choice of definition, therefore, is not a trivial academic exercise; it directly shapes the data we collect and the insights we can glean.
Consider, for instance, the concept of "religious affiliation." On the surface, it seems straightforward: you either belong to a religious group or you don't. However, delving deeper reveals a spectrum of commitment and identification. Some individuals may tick a box on a survey indicating a denominational affiliation out of cultural heritage, despite rarely attending services or holding strong doctrinal beliefs. Others may be deeply spiritual but resist formal affiliation. These nuances challenge simplistic measurement and necessitate frameworks that can accommodate the fluidity and multidimensionality of religious identity. Our goal is not to impose a single, rigid definition, but rather to acknowledge the various ways religion manifests and to select the most appropriate operationalization for the specific research question at hand.
Another critical framework involves understanding the levels of analysis at which religion operates. We can study religion at the individual level, examining personal beliefs, practices, and experiences. We can also investigate religion at the congregational level, analyzing the dynamics of specific faith communities, their demographics, programs, and social impact. Furthermore, religion can be studied at the societal or national level, exploring trends in religious adherence, secularization, and the interplay between religion and politics, economics, and culture. Each level requires different data sources, methodologies, and theoretical lenses. For example, understanding individual prayer habits might involve self-report surveys, while analyzing the growth or decline of religious denominations might draw on demographic census data or organizational records. Recognizing these distinct levels helps us design studies that are appropriately scoped and avoid making inferences about one level based solely on data from another.
Classical sociological theories of religion offer rich frameworks for quantitative investigation. Émile Durkheim, for example, emphasized the social functions of religion, arguing that it binds communities together through shared rituals and beliefs, fostering social cohesion and collective effervescence. From a quantitative perspective, Durkheim's ideas can inspire research into the correlation between religious participation and civic engagement, social capital, or even health outcomes. We might examine whether individuals with higher levels of religious involvement report stronger community ties or greater volunteerism. Such studies can use survey data on religious attendance, social network measures, and various indicators of civic participation to test Durkheimian hypotheses about the integrative power of religion.
Max Weber, on the other hand, explored the relationship between religion and economic behavior, most famously in his work on the "Protestant Ethic" and the spirit of capitalism. Weber posited that certain religious doctrines, particularly those emphasizing asceticism, hard work, and worldly success as signs of divine favor, fostered a rational and industrious approach conducive to capitalist development. While highly debated, Weber's ideas provide a compelling framework for quantitative research. We could investigate, for instance, whether regions with a historical prevalence of certain religious traditions exhibit different patterns of economic development, innovation, or entrepreneurial activity, controlling for other relevant factors. This might involve historical economic data, demographic religious composition, and indicators of economic growth.
More contemporary theoretical frameworks also lend themselves readily to quantitative analysis. Rational choice theory, for example, views religious behavior as a series of decisions individuals make to maximize their perceived benefits (e.g., spiritual rewards, social support) and minimize costs (e.g., financial contributions, time commitment). This framework often inspires studies that examine the "religious economy"—how religious organizations compete for adherents and how individuals choose among various religious "firms." Researchers might use survey data on religious preferences, attendance, and perceived benefits to model individual religious choices, or analyze congregational data to understand market dynamics among religious groups.
Another influential framework is secularization theory, which posits a long-term decline in the social significance of religion in modern societies. While the strong version of secularization theory has been widely challenged, it continues to provide a backdrop for quantitative studies tracking trends in religious affiliation, belief, and practice over time. Researchers use longitudinal surveys, census data, and other large-scale datasets to observe patterns of disaffiliation, changes in religious observance, and shifts in the public role of religious institutions. Data can help us understand whether secularization is a global phenomenon, whether it manifests differently in various cultural contexts, and what factors might contribute to or counteract these trends.
The rise of new forms of religiosity and spirituality also demands robust quantitative investigation. Many individuals today identify as "spiritual but not religious," or engage in practices outside of traditional institutional frameworks. Quantitative methods can help us map these emerging spiritual landscapes, identifying the prevalence of such identifications, their demographic correlates, and their impact on traditional religious institutions. Survey questions designed to capture diverse forms of spirituality, alongside measures of well-being and social engagement, can shed light on these evolving patterns. This pushes us beyond simple denominational categories and into a more nuanced understanding of how individuals construct meaning and connect with the sacred in contemporary society.
Finally, a crucial framework for studying religion with data involves recognizing the dynamic interplay between macro-level societal forces and micro-level individual experiences. For example, social scientists might hypothesize that economic inequality (a macro-level factor) influences individual religious participation (a micro-level outcome), perhaps by driving some individuals towards religious communities for social support or others away due to perceived irrelevance. Conversely, individual religious conversions or increased participation can, in aggregate, lead to changes in societal religious landscapes. Understanding these reciprocal relationships often requires sophisticated statistical models, such as multilevel modeling, which can simultaneously analyze data at different hierarchical levels, accounting for both individual characteristics and contextual factors.
In embracing these diverse frameworks, we also commit to methodological pluralism. Quantitative methods are not designed to answer every question about religion, nor should they replace the invaluable insights offered by qualitative approaches. Instead, they provide a distinct and powerful lens, enabling us to systematically observe, measure, and analyze religious phenomena at scales and with levels of precision that other methods might not achieve. The journey ahead in this book will equip you with the tools to apply these frameworks, transforming abstract theoretical questions into concrete, data-driven investigations that illuminate the complex and multifaceted world of devotion.
This is a sample preview. The complete book contains 27 sections.