Welcome to the International Society for Quantitative Ethnography (ISQE) Resource Hub.
Whether you are a student, researcher, educator, or practitioner, this page is designed to help you begin your journey into Quantitative Ethnography (QE). Here you will find foundational concepts, introductory readings, videos, and practical resources to help you understand what QE is, why it matters, and how it is used to study complex social phenomena and processes.
Quantitative Ethnography is an interdisciplinary field that combines qualitative, quantitative, and computational approaches to study culture, meaning-making, discourse, collaboration, and social processes. QE seeks to preserve the richness and contextual sensitivity of qualitative data while leveraging computational, statistical, and network-based methods to identify patterns and generate new insights. Rather than treating qualitative and quantitative approaches as separate traditions, QE integrates them to address how people think, interact, learn, communicate, and construct meaning across diverse settings.
QE is guided by a set of principles that reflect the field’s commitment to studying meaning-making while leveraging quantitative tools and computational visualizations to support rigorous interpretation.
At its core, QE seeks to understand how people make meaning through discourse, actions, interactions, and artifacts. Rather than focusing solely on individual variables or behaviors, QE examines the patterns of thinking, participation, decision-making, and communication that emerge within social and cultural contexts.
In QE, quantitative methods and computational modeling are used to support, not replace, qualitative interpretation. Statistical analyses and computational visualizations provide representations of relationships among concepts, ideas, and practices. These representations serve as tools for inquiry, helping researchers deepen and refine their interpretations of complex phenomena.
QE is fundamentally contextual. Meaning cannot be separated from the cultural, historical, temporal, and situational conditions in which it is produced. As a result, quantitative analyses and models are only meaningful insofar as they remain grounded in the lived experiences, practices, and perspectives of the individuals and communities being studied.
Coding serves as the bridge between raw data and interpretation. Through systematic coding, researchers identify concepts, practices, beliefs, and experiences that are meaningful within a particular context. These codes form the foundation for both qualitative analyses and quantitative models, ensuring that computational representations remain anchored in emic, or locally meaningful, understandings.
QE emphasizes the importance of “closing the interpretive loop”. Researchers move iteratively between qualitative interpretation and quantitative modeling, using each to inform and refine the other. Quantitative results generate new questions and insights, which are examined through qualitative analysis; those interpretations, in turn, inform revisions to codes, models, and theoretical understandings. This recursive process leads to a richer and more integrated account of meaning-making.
Because coding plays a central role in QE, researchers strive to make coding decisions transparent, systematic, and well-justified. Traditional approaches often emphasize interrater reliability or consensus coding as evidence that coding decisions are applied consistently. More broadly, QE recognizes that all representations reflect particular perspectives and interpretive choices. Researchers therefore seek to develop analytic processes that are fair, rigorous, and reflective of the communities, phenomena, data, and theory under study.
QE emerged from efforts to bridge the qualitative and quantitative traditions, which are often treated as fundamentally separate. The field recognizes that both traditions seek to make warranted claims from evidence and that each offers valuable tools for understanding complex social phenomena. By integrating these perspectives, QE creates new methods and analytic approaches that preserve the richness of qualitative inquiry while providing systematic quantitative evidence to support interpretation.
A central goal of QE is to develop interpretations that are both meaningful and well-supported by evidence. Drawing on traditions such as theoretical saturation in qualitative research and statistical inference in quantitative research, QE provides systematic methods for assessing whether researchers have gathered and analyzed sufficient evidence to support their claims. In this sense, quantitative models serve as warrants that strengthen, rather than replace, qualitative understanding.
If you are new to QE, we recommend the following sequence:
Although projects vary, many QE studies follow a common process:
Becausality: A meaning-based relationship in which one emic symbol helps explain why another idea, action, or outcome occurs within a particular cultural context. In QE, becausal relationships are grounded in signifiers and co-occurrences in the original data and are used to explain why people acted as they did—not necessarily to establish causation in the experimental sense.
Closing the Interpretive Loop: A hallmark of QE in which researchers move iteratively between qualitative interpretation and quantitative modeling. Quantitative results generate new insights and questions that are examined through qualitative analysis; these interpretations then inform revisions to coding schemes, models, and theoretical understandings. This recursive process produces a deeper and more integrated understanding of meaning-making.
Codes: Representations of meaningful concepts, practices, ideas, identities, beliefs, or behaviors identified within data. Codes serve as a bridge between raw data and interpretation, allowing researchers to systematically represent meaning within a dataset.
Coding: The process of categorizing and labeling data to represent concepts, actions, beliefs, experiences, or practices that are meaningful within a particular context. In QE, coding is central because it links qualitative interpretation to quantitative analysis.
Connections: Relationships among codes. Connections represent individual associations among concepts, practices, or ways of thinking and are the foundation of many QE analyses.
Consensus Coding: A collaborative coding approach in which coders discuss and reconcile differences until they reach agreement on the codes assigned to a segment of data. Consensus coding emphasizes shared interpretation and can help make coding decisions more transparent and theoretically grounded.
Context: The cultural, historical, temporal, and situational conditions within which meaning is produced. Context is not merely background information; it is essential for understanding and interpreting data.
Data: The observable traces of meaning-making, including speech, text, actions, interactions, documents, digital logs, images, and other artifacts that can be analyzed to understand social processes.
Discourse: The patterned ways people communicate and interact that reflect how they understand and participate in a particular culture or community. It includes not only what people say, but also how they connect ideas, practices, values, identities, and ways of knowing within a specific context.
Discourse Model: A representation of how meaning is expressed, organized, and communicated through language, actions, interactions, and other symbolic forms. In QE, discourse models describe patterns in meaning-making by representing relationships among concepts, practices, identities, values, or ways of knowing within a particular context. Discourse models serve as an intermediate layer between coded data and models, helping researchers understand how participants construct and negotiate meaning over time. Discourse models are abstractions that highlight meaningful patterns in communication and participation. Examples include networks of codes, representations of epistemic frames, sequences of interaction, and other analytic structures that capture how meaning is organized within a community or system.
Emic Meaning: The meanings, understandings, and interpretations that emerge from the perspective of participants within a particular cultural or social setting. QE seeks to preserve emic meaning when constructing codes, models, and interpretations.
Epistemic Frame: A network of skills, knowledge, values, identities, and ways of knowing that characterize participation in a particular community of practice. Epistemic Frame Theory provides an important theoretical foundation for many QE studies and for Epistemic Network Analysis.
Epistemic Network Analysis (ENA): A QE method that models and visualizes relationships among coded elements in discourse, action, or interaction data. ENA examines how concepts, skills, values, identities, or ways of knowing are connected within a particular context. By representing these connections as networks, ENA helps researchers identify patterns in thinking, learning, participation, and other forms of meaning-making across individuals or groups.
Etic meaning: An analytic interpretation developed from the researcher’s concepts, categories, or theoretical framework. In QE, etic meanings are represented through codes and constructs that must remain demonstrably grounded in participants’ emic symbols and the relationships among them in the original data.
Ethnography: The systematic study of culture, meaning, and social practice through close attention to the experiences, interactions, and perspectives of individuals and communities.
Fairness: The extent to which analytic procedures, representations, and interpretations adequately reflect the perspectives and experiences of those being studied. In QE, fairness is not a single analytic step but a commitment that spans the entire research process, from engaging communities and selecting theoretical frameworks to collecting data, conducting analyses, and interpreting findings. By embedding fairness throughout each stage, researchers can produce more rigorous, transparent, contextually grounded, and equitable representations of meaning.
Horizon: The portion of a context that is available for a given unit of analysis to engage with at a particular time, whether or not that engagement actually occurs. The horizon defines the set of possible concepts, practices, people, resources, or interactions that could enter the unit’s experience.
Interrater Reliability (IRR): A measure of the degree to which independent coders apply codes consistently to the same data. In QE, interrater reliability is often used as one form of evidence that a coding scheme is being applied systematically. Common measures include Cohen’s kappa and Krippendorff’s alpha.
Metadata: The contextual variables that provide information about participants, settings, time, data collection, and study conditions, allowing researchers to situate, interpret, and compare empirical findings.
Meaning-Making: The process through which individuals and groups construct, communicate, and negotiate meaning through discourse, actions, interactions, and artifacts.
Models: Structured representations of phenomena that help researchers understand, explain, and interpret patterns in data.
Networks: Visual representations of relationships among codes, concepts, practices, or participants. Networks enable researchers to examine how elements of meaning-making are organized and connected.
Pattern: A recurring configuration of codes, relationships, practices, or forms of participation identified through qualitative and quantitative analysis.
Ordered Network Analysis (ONA): A QE method that incorporates the temporal ordering of events. Rather than examining only whether concepts were connected, ONA investigates the order in which concepts are used in response to others.
Quantitative Ethnography (QE): An interdisciplinary field of research that unites qualitative and quantitative approaches to study meaning-making. QE uses systematic coding, computational modeling, and contextual interpretation to understand patterns in discourse, actions, interactions, and artifacts.
Reflexivity: The practice of critically examining how researchers’ assumptions, identities, experiences, and decisions influence the research process and the interpretations that emerge from it.
Reliability: The consistency and stability of a measurement, coding process, or analytic procedure. In QE, reliability refers not only to whether coding decisions can be reproduced across coders or time, but also to whether analytic procedures are applied systematically and transparently.
Representation: A theory-informed depiction of some aspect of a phenomenon, produced through decisions about what data to collect and how to segment, code, model, and visualize them. In QE, a representation foregrounds particular patterns of meaning and connections while necessarily omitting other details; its adequacy is evaluated by returning to the original data and context to determine whether it supports a warranted interpretation.
Segmentation: The process of dividing data into meaningful segments for analysis. Segmentation helps define the boundaries of observations, interactions, or discourse that will be coded and analyzed. The way data are segmented influences the patterns and relationships that can be identified in a QE study.
Statistical Warrant: Evidence derived from quantitative analyses that supports an interpretation or claim. In QE, statistical warrants complement, rather than replace, qualitative warrants and are evaluated within the broader context of a study’s theoretical and interpretive framework.
Theoretical Saturation: The point at which additional data no longer contribute substantially new insights to an emerging interpretation. QE builds on this qualitative concept by using quantitative methods to provide additional evidence for theoretical saturation.
Theoretical Warrant: Evidence supporting an interpretation based on theoretical coherence, conceptual consistency, and alignment with existing scholarship. Theoretical warrants help explain why observed patterns are meaningful and how they relate to broader phenomena.
Transmodal Analysis (TMA): A QE method that examines how meaning is constructed across multiple modes of communication and action. TMA is grounded in the idea that meaning-making is often distributed across modes—speech, text, gesture, movement, artifacts, and other forms of discourse—and cannot always be modeled by analyzing one mode alone. By examining relationships across different forms of discourse, TMA helps researchers investigate complex social, cognitive, and collaborative processes.
Transparency: The practice of clearly documenting analytic decisions, including data collection, segmentation, coding, modeling, and interpretation. Transparency allows others to understand, evaluate, and potentially reproduce the analytic process.
Trustworthiness: The overall credibility, rigor, and defensibility of a study’s findings. In QE, trustworthiness emerges from systematic coding, transparent analytic procedures, theoretically grounded interpretations, and the integration of qualitative and quantitative evidence.
Units of Analysis: The entities being studied and compared. Units may be individuals, groups, teams, classrooms, organizations, events, or other meaningful categories.
alidity: The extent to which a study, coding scheme, model, or interpretation accurately represents the phenomenon being investigated. In QE, validity is established through the alignment of theory, data, coding, modeling, and interpretation, as well as through the iterative process of closing the interpretive loop.
Warrant: Evidence that justifies or supports a claim or interpretation. QE unifies qualitative and quantitative warrants to strengthen confidence in conclusions and ensure that interpretations are systematically grounded in evidence.
Window: The local analytic context within which coded elements are considered related. A window determines which coded observations are connected to one another for analysis.
No. We actually recommend beginners start with the ENA webtool, a free tool with an easy-to-navigate interface. However, if you are comfortable with R, there are various free R packages and tutorials are available.
Not necessarily. Familiarity with basic inferential statistics is helpful and can be learned alongside QE methods. Foundational resources are available for beginners of all levels.
Not necessarily. Familiarity with qualitative coding and interpretation is helpful and can be learned alongside QE methods. Foundational resources are available for beginners of all levels.
No. QE is used in diverse fields, including education, psychology, communication, health sciences, organizational studies, design, engineering, physics, and many other disciplines.
Many newcomers begin with epistemic network analysis (ENA), though the appropriate analytical tool depends on the research question and data.
Yes. QE methods can be applied to most forms of qualitative and mixed-methods data.
Most QE tools require your data to be organized in a tabular (spreadsheet) format, where each row represents an event (e.g., an utterance, sentence, message, gesture, &c.) and each column represents one ontologically consistent category (e.g., participant ID, gender, group ID, code A, code B, code C, &c.). At a minimum, your dataset should include:
The QE community is highly collaborative, and there are many ways to get help as you learn. You can:
| Foundational reading | ||
|---|---|---|
| Quantitative Ethnography | Shaffer, D. W. (2017) | The foundational book defining QE, its philosophy, assumptions, and core methodological framework. |
| What Good Are Statistics That Don’t Generalize? | Shaffer, D. W., & Serlin, R. C. (2004) | This paper explains how researchers can use statistics to strengthen their understanding of patterns within a particular qualitative study, even when they are not trying to generalize the findings to a larger population. |
| Is QE Just ENA? | Shaffer, D. W., & Ruis, A. R. (2022) | The paper examines whether QE is synonymous with ENA, concluding that although the two are closely connected, neither is exclusive to the other and their relationship is best understood through continued methodological reflexivity. |
| The Foundations and Fundamentals of Quantitative Ethnography | Arastoopour Irgens, G., & Eagan, B. (2023) | This paper outlines 10 steps in the iterative QE modeling pathway and 5 main practices in a QE analysis. |
| Embracing Mess: Reflection on How We Engage with QE. | Espino, D., & Onyewuenyi, A.C. (2025) | The paper examines how researchers’ diverse disciplinary and methodological backgrounds enrich QE, emphasizing reflexive, intentional, and community-engaged practices throughout the research process. |
| Introductory videos | |
|---|---|
| Brief QE overview | A short primer on what QE is and why it matters. |
| Quantitative Ethnography: Tools for Modeling Meaning in Learning Analytics | A more detailed look at how QE supports modeling meaning in learning analytics. |
| Introduction to Quantitative Ethnography | An accessible introduction to the key concepts and methods of QE. |
Browse our Zotero library for a comprehensive, continually updated collection of Quantitative Ethnography research.