9.1. When and How to Use Ethnography

Anthropologists pioneered the use of ethnography through extended in-person observations of indigenous peoples. Notably, Franz Boas (1888) lived with the Inuit on Canada’s Baffin Island and then went on to study native populations in the Pacific Northwest. Bronislaw Malinowski (1922) spent years observing indigenous people living in the Trobriand archipelago of Papua New Guinea. Up to that time, anthropologists had largely studied groups of people by drawing upon secondhand accounts or inspecting artifacts. The compelling findings unearthed for Western audiences by these early ethnographies, however, popularized a new style of field research that involved living in or near the communities being studied and learning directly from them.

Soon a number of sociologists adapted ethnographic methods to advance their own research projects. The University of Chicago’s sociology department was incredibly influential in shaping this new approach to research in part because of its connection to Hull House, a settlement house that provided assistance to low-income immigrants (Deegan 1988). Many of sociology’s first ethnographers were former social workers who became interested in such research through their personal experiences working with the poor and advocating for social reforms. They developed what eventually came to be known as the “Chicago school” of sociology, characterized by its systematic use of ethnographic observation to understand the city’s neighborhoods—block by block.

In the 1970s, the anthropologist Clifford Geertz developed an ethnographic practice called thick description, which would influence qualitative researchers across disciplines. If “thin” description was a surface-level observation of what people were doing, “thick” description provided needed context, allowing the researcher to understand the social and subjective meanings of particular actions (Geertz 1973). We’ve mentioned thick description a few times in this textbook by now, but it merits a thicker description of its own.

Thick description generates rich and complex qualitative data. Researchers using this approach document even minute details of what is unfolding before them, focusing on a micro, and sometimes microscopic, level of analysis. They pay particular attention to any patterns and processes they’ve observed. They capture narratives of the social interactions and relationships they’ve witnessed. Yet they are not intent on producing merely a facsimile, transcription, or photorealistic rendition of what they’ve seen or heard. (If that was the case, ethnographers could just walk around with video cameras.) They also want to interpret the underlying meanings of what they see and hear.

This is the purpose behind going into the field: to study social phenomena within their natural setting, a specific social and historical context. (For this reason, sometimes social scientists describe qualitative approaches as naturalistic.) Proponents of thick description (and qualitative methods more broadly) emphasize the importance of context. You cannot truly understand what you observe, they argue, without viewing it against the background in which it is situated. You will miss important nuances and misinterpret basic truths.

Being embedded in a social context and gleaning rich qualitative data from it gives ethnographers the ability to understand complex processes and relationships and explain the meanings of what they observe—that is, the social significance of specific behaviors, symbols, appearances, and interactions. This in-depth, interpretative approach is very different from the one taken in quantitative analysis, which breaks down complicated phenomena into numbers for easier processing. Through quantification, researchers can compare numerous and diverse cases and identify larger patterns across them. Yet they lose something, too, because quantification strips away context and otherwise simplifies social reality. As you can guess, these different qualitative and quantitative approaches are useful in their own ways.

In Table 9.1 we quickly describe several celebrated ethnographies, the research questions they posed, and the types of data they collected. As we discussed in Chapter 4: Research Questions, ethnography—and qualitative research more generally—can answer “how” kinds of questions quite well: how the processes being studied occur, how the people being observed interact, and how events being witnessed unfold. In line with our discussions in previous chapters, such studies tend to adopt an inductive approach, generating plausible and nuanced theories that answer these “how” questions.

Table 9.1. Examples of Ethnographies

One thing to note is that all of these ethnographies were published as books. Some journals (such as Qualitative Sociology) specialize in qualitative work, and ethnographers routinely publish portions of their work as articles in top sociology journals. However, given their use of rich detail and narratives, ethnographies often lend themselves to a monograph format. Many journal tightly control the length and structure of their articles, and ethnographers often prefer the greater flexibility they have when writing up their research as books (which can themselves be called “ethnographies”). This means, however, that the book that results from an ethnographic study may only be published many years after its fieldwork has concluded.

Now that we have a better sense of the kind of observations involved in ethnography, let’s talk about the general approaches to conducting it and the strengths and weaknesses of the method.

The Participation Continuum

As we noted, many ethnographers choose not to be passive observers of social movements, companies, agencies, and the like. In fact, there is a storied tradition in sociology of actively contributing to the success of the organization or occupation under study—as a volunteer, a dues-paying member, or even an employee.

Because ethnographers get involved in the organizations they study so frequently, the term participant observation is sometimes used as a synonym for ethnography. This is a bit misleading, though, because the degree to which ethnographers “participate” in activities within the settings they observe actually varies quite a bit. You might say that there is a participation continuum in ethnography: complete participation lies at one end of that spectrum, and complete nonparticipation lies at the other.^[1] A common term for the latter—not engaging in the activities or groups being studied—is direct observation. This phrase, however, implies a level of objectivity and intensity that, as we discuss later, is not necessarily heightened at this end of the spectrum. Alternatively, you can think of this approach as bystander observation, where the researcher is an eyewitness, but not a participant, in relation to the action being observed.

In other chapters, we discussed two works that fall near either end of the participation continuum. Barrie Thorne’s (1993) observations of children in classrooms, school cafeterias, hallways, and playgrounds (mentioned in Chapter 5: Research Design) would lie near the “bystander” end of the spectrum. Rather than taking on the role of, say, an elementary school teacher, Thorne simply observed. The diametrically opposite approach was taken by Laud Humphreys (1970) in his controversial research on the “tearoom trade,” an informal but structured practice of anonymous and illicit sex taking place in men’s public restrooms (described in Chapter 8: Ethics). As you might remember, Humphreys participated as a “watch queen,” a role that actual participants in the tearoom trade also played.

One can argue that ethnographies always exist in some sort of middle ground between the poles of complete participation and nonparticipation. (And sociologists, as usual, disagree about what “participation” entails to begin with.) If you take a deep dive into the methodological debates among ethnographers, you’ll find different typologies and advice for where you might locate yourself on this continuum. Some argue that true “nonparticipation” might entail something like a video content analysis (discussed in Chapter 15: Materials-Based Methods), where the individual only observes a video recording of an event, with no physical participation whatsoever (Johanson and Williamson 2013). For our part, we recommend that you simply recognize the many gradations of participation that exist, however you categorize them—from standing aloof from, or peripheral to, the scene being observed (with the goal of being a more objective observer), to more or less actively engaging in some activities (yet still prioritizing your role as a researcher), to becoming (or already being) a full-fledged member of the group, accepted by all. You may even decide that sometimes you’ll take on one role and other times another, tweaking your degree of clinical detachment depending on what is appropriate within a specific social context.

In weighing the pros and cons of participation, it’s worth considering how journalism—a trade that also does its fair share of observation—would view such arrangements. Without question, volunteering or working for an organization you’re writing about would flagrantly violate ethical codes across the more reputable news outlets within the media industry. As we noted in Chapter 8: Ethics, however, journalism and sociology have different professional standards regarding bias and fairness, each of which can be justified on its own terms.

Compared to journalists, ethnographers are much more willing to occupy positions closer to the participation side of the continuum. The rationale is that participant observation provides an authentic view of the life of the group being studied. Participants may trust the researcher more, and as a result they will act more naturally, without the desire to hide certain behaviors or information from outsiders. The researcher, in turn, may gain esoteric or hidden knowledge about how the group operates or thinks, providing greater insight than an outsider could obtain. For these reasons, some ethnographers argue that participation is the only way to truly understand a particular social setting. Whether that’s true is debatable, but the tradeoffs of standing at a remove from those you’re studying is very real. Being an outsider observer looking in on the organization may mean that you may miss or misunderstand the meanings of what people say or do around you. You may also have difficulty building rapport with the members of the group you’re studying, perhaps preventing you from observing important aspects of the group’s social life.

The issue with being deeply embedded in the social worlds of those you study, of course, is ethnographic cooptation: you may start looking with an uncritical eye upon people’s beliefs, practices, and decisions. You may start agreeing with their claims or the righteousness of their cause while disregarding any contrary evidence, falling heavily into the trap of confirmation bias. You may forget how outsiders (the eventual readers of your published work) would look upon this group, which may be starkly different from how group members understand themselves—perhaps for good reason. These risks bring to mind a “parable-ish” joke recounted by the novelist David Foster Wallace (2009) in a commencement speech: “There are these two young fish swimming along, and they happen to meet an older fish swimming the other way, who nods at them and says, ‘Morning, boys. How’s the water?’ And the two young fish swim on for a bit, and then eventually one of them looks over at the other and goes, ‘What the hell is water?’” In a research context, you may need to step outside the water for a bit to recognize the importance of what everyone else seems to take for granted.

Another danger of getting in too deep (no pun intended) is that you may face heightened pressures to do something unethical or unsavory. For instance, a participant observer of criminal activities might find themselves witnessing—or even being urged to commit—something illegal. In Chapter 8: Ethics, we described a few high-profile cases of ethnographers who allegedly crossed one or more of these ethical lines, but we should point out that peer pressure is a common danger even when researching quite mundane organizations. Colleagues or higher-ups may push you to do things you’re not entirely comfortable with at a particular workplace or protest site, for instance.

Insiderism of this sort is seen as a serious problem in the news profession, which, at its best, seeks to avoid the partisanship and credulousness that allow for the manipulation and distortion of news reporting. Similar criticisms may be leveled against researchers who fall into this category of “going native,” as this phenomenon is sometimes called. Yet ethnographers, unlike journalists, are often willing to risk such observational overembeddedness in order to reap the advantages of full participation we’ve just described.

A more practical reason not to participate so much in the activities of the group being studied is that bystander observers can focus just on the task of observing. Observation can be exhausting and time-consuming in itself, depending on how much detail is being gathered. Not surprisingly, when you are actually involved with the organization you are studying, you can become too absorbed with your tasks to pay sufficient attention to what is going on around you. You may also run into issues with reactivity, in that interactions between you and your research participants may lead to certain outcomes that differ from what would have happened if the researcher wasn’t there. In this sense, nonparticipation—and even invisibility—is preferable. The British writer and aristocrat Jessica Mitford famously related how her governess would always whisper in her ear before any social occasion: “You’re the least important person in the room and don’t forget it” (Didion 2008). That savage putdown for an aristocrat can actually be good advice for an ethnographer, who generally does not want to be the center of attention. In fact, making yourself as inconspicuous as possible may embolden people to act more like their authentic selves in front of you.

So where should you locate yourself on the participation continuum? Some ethnographers argue that a happy medium is best: straddling the two roles of observer and participant, so that you gain greater access to the social world being studied but you refrain from “going native” (Chatman 1992). Fred Davis (1973) famously captured the equal value—and tradeoffs—in these two vantage points when he described the ethnographer as being either a “Martian” or a “convert.” Martians, he argued, kept some emotional and social distance from their participants in order to see them with fresh eyes. Converts dove deeply into the social life of their field site. One perspective assumed that dispassionate observation yielded a fuller truth; the other argued that total immersion led to true understanding. You may find yourself favoring one or the other approach in your own ethnographic observation, and you should do what fits best with your personal style. But remember the dangers of leaning too hard in one direction.

The debate over participation we’ve just had falls into a favorite pastime of ethnographers: reflexivity. This is a self-reflective process that sociologists engage in to understand how their own identity, beliefs, dispositions, actions, and practices may have influenced their research. A similar concept is positionality, which holds that one’s personal identities and perspectives shape how one interprets reality. The two terms are used somewhat interchangeably, but reflexivity is also attuned to the ways that a researcher’s actions (not just their identity) may influence their findings. For instance, reactivity—the ways that your presence in a setting may alter how participants act—is something sociologists also have to consider when they are being self-reflexive about their work.

Ethnographers routinely devote significant portions of their writing to this kind of critical evaluation of themselves as researchers. In separate reflexivity statements or positionality statements, or in discussions woven into the methods sections of their papers or books, they will describe in detail the specific roles they had in the research process, what implications their involvement may have had for what they ended up observing, how their group identities may have affected their research, and so on.

This continual self-examination can help ethnographers make sure that they are getting things right—that they aren’t imposing their outsider perspective upon a community they helicoptered into, for instance. It can also make them more sensitive to their treatment of their research participants and whether it is ethical and respectful. And it helps explain why many sociologists find it acceptable, and perhaps even preferable, to take on participant roles. Rather than trying to achieve some perfect state of objectivity, many sociologists would rather just acknowledge their biases and discuss what consequences they may have had for their analysis. (We’ll have more to say about reflexivity when we talk later about how an ethnographer’s personal identity affects their research.)

Regardless of whether you choose to observe as a participant or a bystander, remember that as a qualitative researcher, you are part of the measurement instrument. So much depends on your mettle as an individual existing in a particular social moment. In the field, you must rely on your observational skills, the trust you build with participants, and your ability to extract the correct information. Your highly subjective and personal experiences in the social context you are studying will allow you to interpret what is going on in a rich and full fashion. At the same time, if your goal is scientific—rendering an accurate portrayal of the phenomenon you’re studying—you need to maintain a fair-minded perspective. Keep your biases in check so that you don’t—consciously or unconsciously—twist the details you see to fit whatever narrative you prefer. Confirmation bias is very real.

The Strengths and Weaknesses of Ethnography

Ethnography allows researchers to gain firsthand experience and knowledge about the people and processes they study. Perhaps no other method offers an experience of everyday life that is this close-up and detailed. As we have noted, ethnography is a superior method for understanding the role of social context in shaping people’s lives and experiences. It excels at illuminating causal mechanisms as well—the specific pathways by which one variable influences another—given that thick description can fully capture complex relationships and interactions. In fact, as an ethnographer, you may be able to see these processes unfold before your very eyes. There are few kinds of data with greater face validity than what you can gather here: a clear and cogent story, directly observed.

Ethnography is also an excellent method to use for exploratory research. It is likely to generate results that surprise you, teaching you something about the phenomenon you’re studying that you didn’t know before you went into the field. With surveys and interviews, you have to know what questions to ask other people about a phenomenon of interest; with ethnographic observation, you can see it for yourself, which can draw your attention to things that the people regularly in those settings may not think about themselves. Furthermore, the freedom you have as a researcher to keep exploring the same sites or interacting with the same sorts of people can allow you to adapt to new information acquired in the field, test out your latest hunches, and seek out theoretically interesting cases (in the abductive approach we described in Chapter 4: Research Questions). As an ethnographer, you should take full advantage of the method’s nonjudgmental flexibility and openness to new possibilities. “I had a mentor who taught me that if you end up with the story you started with, then you weren’t listening along the way,” says the documentary filmmaker Matthew Heineman (McClennen 2022). ”It’s good advice for life and good advice for filmmaking.” It’s also good advice for ethnography.

The unmediated access that ethnographers have to the phenomena they’re studying also allows them to avoid some of the measurement problems that bedevil other methods. For instance, in-depth interviews and surveys ultimately depend on the hopeful notion that respondents will accurately relay their experiences and reasons for doing things. But what if their recall is faulty? What if they don’t know why they really do what they do? Or what if they feel uncomfortable about telling you what they truly think? Speaking again to this idea of being pleasantly and productively surprised, ethnographic observation can sometimes tell us things that our research participants don’t even know. And because ethnography typically occurs over an extended period of time, the researcher may discern certain social facts over the course of their study that were at first invisible or hidden from them.

Broadly speaking, ethnography involves two challenges. First, ethnographers need to be fully engaged with the social context they’re studying in order to generate useful data and theories. Ethnographic observation is a very labor-intensive form of research that requires researchers to think creatively and improvise constantly while in the field. It often takes a great deal of time to collect sufficient data on a few cases—even just a single site of study. Documenting ethnographic observations can also suck up a great deal of time and energy. In interview and survey research, respondents typically enter their responses themselves or have their spoken words recorded. Ethnographers generally have only themselves to rely on for documenting what they observe. This challenge becomes immediately apparent upon entering the field. As we discuss later, it may not be possible to take field notes while you are observing. Afterward, you won’t recall everything exactly as you saw it. Nor will you necessarily know which details to document out of the countless number that you could write down. The most revealing bits of ethnographic data for your study may very well turn out to be ones you thought were inconsequential when you first recorded them. This is the chaos and uncertainty that ethnographers regularly deal with—and that good ethnographers learn to take in stride.

Ethnography can also be emotionally taxing. As with in-depth interviewing, ensuring your observations yield useful data often requires establishing a rapport with the people you are studying; otherwise, participants may hide things or refuse to be part of the study in the first place. And while interviews and ethnography both require building such relationships, you might say that in-depth interviewing is like casual dating, while ethnography is like a long-term relationship. The latter generally requires a narrower focus and a heavier commitment—one often sustained over a much longer period than the hour or two it might take you to conduct a typical interview. As a result, the relationships you develop as an ethnographer tend to be more intimate. A number of ethnographers have described their close bonds—and, sometimes, complicated relationships—with research participants (Duneier 1999; Lareau 2011; MacLeod 2008). Gaining the trust and respect of your participants not only makes it more likely that your observation will yield rich data, but it can also be personally rewarding. It’s hard not to spend so much time with other people and not feel some connection to them and some enjoyment of their company. As in any relationship, however, you will experience emotional highs and lows, as you encounter the best and worst of the people and organizations you are studying. For a ground-level perspective of what this process of emotional immersion is like, watch Video 9.1, a companion documentary for Mitchell Duneier’s book Sidewalk (1999).

The second overarching challenge facing ethnographers is one we’ve described before (including in Chapter 6: Sampling) as a pitfall for all qualitative researchers: generalizability. Compared to other methods, it is a much heavier lift for ethnographers to convince other researchers that the social patterns they observed within their chosen setting or settings somehow apply more broadly. This is inevitable given the first drawback we discussed, the labor-intensive nature of ethnography. The approach is so detailed-oriented that it is by necessity somewhat narrow. Ethnographers are forced to focus on one or a handful of field sites, restricting themselves to a quite limited view of a larger phenomenon. They simply cannot gather data about as many individuals or locales as, say, a survey researcher can. Of course, this limitation of ethnography is also its strength, because ethnographers sacrifice breadth in exchange for depth. They retain the stories, context, and meanings that quantitative analyses by necessity strip away (Wang 2016). From the fine-grained and richly contextualized data their observations generate, they can piece together highly nuanced and sophisticated understandings of causal mechanisms, subjective viewpoints, and the underlying social meanings of what people do and say.