Chapter 6. Reflexivity
Introduction
Related to epistemological issues of how we know anything about the social world, qualitative researchers understand that we the researchers can never be truly neutral or outside the study we are conducting. As observers, we see things that make sense to us and may entirely miss what is either too obvious to note or too different to comprehend. As interviewers, as much as we would like to ask questions neutrally and remain in the background, interviews are a form of conversation, and the persons we interview are responding to us. Therefore, it is important to reflect upon our social positions and the knowledges and expectations we bring to our work and to work through any blind spots that we may have. This chapter discusses the concept of reflexivity and its importance for conducting reliable qualitative research.
Reflexivity: What It Is and Why It Is Important
Remember our discussion in epistemology? Qualitative researchers tend to question assertions of absolute fact or reality, unmediated through subject positions and subject knowledge. There are limits to what we know because we are part of the social worlds we inhabit. To use the terminology of standpoint theorists, we have a standpoint from which we observe the world just as much as anyone else. In this, we too are the blind men, and the world is our elephant. None of us are omniscient or neutral observers. Because of this epistemological standpoint, qualitative researchers value the ability to reflect upon and think hard about our own effects on our research. We call this reflexivity. Reflexivity “generally involves the self-examination of how research findings were produced, and, particularly, the role of the researcher in their construction” (Heaton 2004:104).
There are many aspects of being reflexive. First, there is the simple fact that we are human beings with the limitations that come with that condition. We have likes and dislikes, biases, blind spots, preferences, and so on. If we do not take these into account, they can prevent us from being the best researcher we can be. Being reflective means, first and foremost, trying as best as possible to bracket out elements of our own character and understanding that get in the way. It is important to note that bias (in this context, at least) is not inherently wrong. It just is. Unavoidable. But by noting it, we can minimize its impact or, in some cases, help explain more clearly what it is we see or why it is that we are asking the questions we are asking. For example, I might want to communicate to my audience that I grew up poor and that I have a lot of sympathy and concern for first-generation college students as a result. This “bias” of mine motivates me to do the work I do, even as I try to ensure that it does not blind me to things I find out in the course of my research. [1]
A second aspect of being reflexive is being aware that you yourself are part of the research when you are conducting qualitative research. This is particularly true when conducting interviews, observing interactions, or participating in activities. You have a body, and it will be “read” by those in the field. You will be perceived as an insider or an outsider, as a friend or foe, as empathetic or hostile. Some of this will be wrong. People will prejudge you based on the color of your skin, your presented gender, the accent of your language. People will classify you based on the clothes you wear, and they will be more open to you if you remind them of a friendly aunt or uncle and more reserved if you remind them of someone they don’t like. This is all natural and inevitable. Your research will suffer if you do not take this into account, if you do not reflect upon how you are being read and how this might be influencing what people tell you or what they are willing to do in front of you. The flip side of this problem is that your particular body and presence will open some doors barred to other researchers. Finding sites and contexts where your presented self is a benefit rather than a burden is an important part of your individual research career. Be honest with yourself about this, and you will be more successful as a qualitative researcher. Learn to leverage yourself in your research.
The third aspect of being reflexive is related to how we communicate our work to others. Being honest with our position, as I am about my own social background and its potential impact on what I study or about how I leveraged my own position to get people to open up to me, helps our audiences evaluate what we have found. Maybe I haven’t entirely eliminated my biases or weaknesses, but by telling my audience who I am and where I potentially stand, they can take account of those biases and weaknesses in their reading of my findings. Letting them know that I wore pink when talking with older men because that made them more likely to be kind to me (a strategy acknowledged by Posselt [2016]) helps them understand the interview context. In other words, my research becomes more reliable when my own social position and the strategies I used are communicated.
Some people think being reflective is just another form of narcissistic navel-gazing. “The study is not about you!” they might cry. True, to some degree—but that also misses the point. All studies on the social world are inevitably about us as well because we are part of that social world. It is actually more dangerous to pretend that we are neutral observers, outside what we are observing. Pierre Bourdieu makes this point several times, and I think it is worth quoting him here: “The idea of a neutral science is fiction, an interested fiction which enables its authors to present a version of the dominant representation of the social world, naturalized and euphemized into a particularly misrecognizable and symbolically, therefore, particularly effective form, and to call it scientific” (quoted in Lemert 1981:278).
Bourdieu (1984) argues that reflective analysis is “not an epistemological scruple” but rather “an indispensable pre-condition of scientific knowledge of the object” (92). It would be narcissistic to present findings without reflection, as that would give much more weight to any findings or insights that emerge than is due.
The critics are right about one thing, however. Putting oneself at the center of the research is also inappropriate.[2] The focus should be on what is being researched, and the reflexivity is there to advance the study, not to push it aside. This issue has emerged at times when researchers from dominant social positions reflect upon their social locations vis-à-vis study participants from marginalized locations. A researcher who studies how low-income women of color experience unemployment might need to address her White, upper-class, fully employed social location, but not at the cost of crowding out the stories, lived experiences, and understandings of the women she has interviewed. This can sometimes be a delicate balance, and not everyone will agree that a person has walked it correctly.
Examples of Reflexivity in Practice
Most qualitative researchers include a positionality statement in any “methods section” of their publications. This allows readers to understand the location of the researcher, which is often helpful for gauging reliability. Many journals now require brief positionality statements as well. Here are a few examples of such statements.
The first is from an ethnographic study of elite golfers. Ceron-Anaya (2017) writes about his class, race, and gender and how these aspects of his identity and social location affected his interactions with research participants:
My own class origins, situated near the intersection between the middle and the lower-middle class, hindered cooperation in some cases. For example, the amiable interaction with one club member changed toward the end of the interview when he realized that I commonly moved about in the city by public transportation (which is a strong class indicator). He was not rude but stopped elaborating on the answers as he had been doing up to that point.…Bodily confidence is a privilege of the privileged. My subordinate position, vis-à-vis golfers, was ameliorated by my possession of cultural capital, objectified in my status of researcher/student in a western university. However, my cultural capital dwindled in its value at the invisible but firm boundary between the upper-middle and the upper class. The few contacts I made with members of the upper class produced no connections with other members of the same group, illustrating how the research process is also inserted in the symbolic and material dynamics that shape the field. (288)
What did you learn from Ceron-Anaya’s reflection? If he hadn’t told you about his background, would this have made a difference in reading about elite golfers? Would the findings be different had Ceron-Anaya driven up to the club in a limousine? Is it helpful to know he came by bus?
The second example is from a study on first-generation college students. Hinz (2016) discusses both differences and similarities between herself and those she interviewed and how both could have affected the study:
I endeavored to avoid researcher bias by allowing the data to speak for itself, but my own habitus as a White, female, middle-class second-generation college student with a few years of association with Selective State [elite university] may have influenced my interpretation. Being a Selective State student at the time of the interviews provided a familiarity with the environment in which the participants were living, and an ease of communication facilitated by a shared institutional culture. And yet, not being a first-gen myself, it seemed as if I were standing on the periphery of their experience, looking in. (289–290)
Note that Hinz cannot change who she is, nor should she. Being aware (reflective) that she may “stand on the periphery” of the experience of those she interviews has probably helped her listen more closely rather than assume she understands what is really going on. Do you find her more reliable given this?
These statements can be quite long, especially when found in methodological appendixes in books rather than short statements in articles. This last lengthy example comes from my own work. I try to place myself, explaining the motivations for the research I conducted at small liberal arts colleges:
I began this project out of a deep curiosity about how college graduates today were faring in an increasingly debt-ridden and unequal labor market. I was working at a small liberal arts college when I began thinking about this project and was poised to take a job at another one. During my interview for the new job, I was told that I was a good fit, because I had attended Barnard College, so I knew what the point of a liberal arts college was. I did. A small liberal arts college was a magical place. You could study anything you wanted, for no reason at all, simply for the love of it. And people would like you for it. You were surrounded by readers, by people who liked to dress up in costume and recite Shakespeare, by people who would talk deep into the night about the meaning of life or whether “beauty” existed out there, in nature, or was simply a projection of our own circumstances.
My own experience at Barnard had been somewhat like that. I studied Ancient Greek and Latin, wrote an undergraduate thesis on the legal standing of Vestal Virgins in Ancient Rome, and took frequent subway rides to the Cloisters, the medieval annex of the Metropolitan Museum of Art, where I sketched the courtyard and stared at unicorn tapestries. But I also worked full-time, as a waitress at a series of hectic and demanding restaurants around the city, as a security guard for the dorm, as a babysitter for some pretty privileged professors who lived in doorman buildings along Riverside Park, and at the library (the best job by far). I also constantly worried I would not be able to finish my degree, as every year I was unsure how I would come up with the money to pay for costs of college above and beyond the tuition (which, happily, was covered by the college given my family’s low income). Indeed, the primary reason I studied the Classics was because all the books were freely available in the library. There are no modern textbooks—you just find a copy of the Iliad. There are a lot of those in a city like New York. Due to my fears, I pushed to graduate one year early, taking a degree in “Ancient Studies” instead of “Classics,” which could have led on to graduate training. From there, I went to law school, which seemed like a safe choice. I do not remember ever having a conversation with anyone about how to find a job or what kinds of job one could do with a degree in Ancient Studies. I had little to no social networks, as I had spent my time studying and working. And I was very lucky, because I graduated with almost zero debt.
For years, until that job interview, I hadn’t really thought my Barnard experience had been that great or unusual. But now it was directly helping me get a job, about fifteen years after graduation. And it probably had made me a better person, whatever that means. Had I graduated with debt, however, I am not so sure that it would have been worth it. Was it, on balance, a real opportunity and benefit for poor students like me? Even now? I had a hunch of what I might find if I looked: small liberal arts colleges were unique places of opportunity for low-income first-generation working-class students who somehow managed to find and get in to one of them (no easy task). I thought that, because of their ethos, their smallness, the fact that one could not hide from professors, these colleges would do a fair job equalizing opportunities and experiences for all their students. I wanted to tell this story. But that is not the story that I found, or not entirely. While everyone benefits from the kind of education a small liberal arts college can offer, because students begin and continue so differently burdened and privileged, the advantages of the already-advantaged are amplified, potentially increasing rather than decreasing initial inequalities. That is not really a surprising story, but it is an important one to tell and to remember. Education doesn’t reduce inequality. Going to a good college doesn’t level the playing field for low-income, first-generation, working-class students. But perhaps it can help them write a book about that. (Hurst 2019:259–261)
What do you think? Did you learn something about the author that would help you, as a reader, understand the reasons and context for the study? Would you trust the researcher? If you said yes, why?
How to Do It
How does one become a reflective researcher? Practice! Nearly every great qualitative researcher maintains a reflexive journal (there are exceptions that prove the rule), a type of diary where they record their thinking on the research process itself. This might include writing about the research design (chapter 2), plotting out strategies for sample selection (chapter 6), or talking through what one believes can be known (chapter 3). During analysis, this journal is a place to record ideas and insights and pose questions for further reflection or follow-up studies. This journal should be highly personal. It is a place to record fears, concerns, and hopes as well. Why are you studying what you are studying? What is really motivating you? Being clear with yourself and being able to put it down in words are invaluable to the research process.
Today, there are many blogs out there on writing reflective journals, with helpful suggestions and examples. Although you may want to take a look at some of these, the form of your own journal will probably be unique. This is you, the researcher, on the page. Each of us looks different. Use the journal to interrogate your decisions and clarify your intent. If you find something during the study of note, you might want to ask yourself what led you to note that. Why do you think this “thing” is a “thing”? What about your own position, background, or researcher status that makes you take note? And asking yourself this question might lead you to think about what you did not notice. Other questions to ask yourself include the following: How do I know “that thing” I noted? So what? What does it mean? What are the implications? Who cares about this and why? Remember that doing qualitative research well is recursive, meaning that we may begin with a research design, but the steps of doing the research often loop back to the beginning. By keeping a reflective journal, you allow yourself to circle back to the beginning, to make changes to the study to keep it in line with what you are really interested in knowing.
One might also consider designing research that includes multiple investigators, particularly those who may not share your preconceptions about the study. For example, if you are studying conservative students on campus, and you yourself thoroughly identify as liberal, you might want to pair up with a researcher interested in the topic who grew up in a conservative household. If you are studying racial regimes, consider creating a racially diverse team of researchers. Or you might include in your research design a component of participatory research wherein members of the community of interest become coresearchers. Even if you can’t form a research team, you can reach out to others for feedback as you move along. Doing research can be a lonely enterprise, so finding people who will listen to you and nudge you to clarify your thinking where necessary or move you to consider an aspect you have missed is invaluable.
Finally, make it a regular part of your practice to write a paragraph reporting your perspectives, positions, values, and beliefs and how these may have influenced the research. This paragraph may be included in publications upon request.
Internal Validity
Being reflexive can help ensure that our studies are internally valid. All research must be valid to be helpful. We say a study’s findings are externally valid when they are equally true of other times, places, people. Quantitative researchers often spend a lot of time grappling with external validity, as they are often trying to demonstrate that their sample is representative of a larger population. Although we do not do that in qualitative research, we do sometimes make claims that the processes and mechanisms we uncover here, in this particular setting, are likely to be equally active in that setting over there, although there may be (will be!) contextual differences as well. Internal validity is more peculiar to qualitative research. Is your finding an accurate representation of what you are studying? Are you describing the people you are observing or interviewing as they really are? This is internal validity, and you should be able to see how this connects with the requirement of reflexivity. To the extent that you leave unexamined your own biases or preconceptions, you will fail at accurately representing those people and processes you study. Remember that “bias” here is not a moral failing in the way we commonly use bias in the nonresearch world but an inevitable product of our being social beings who inhabit social worlds, with all the various complexities surrounding that. Because of things that have happened to you, certain things (concepts, quotes, activities) might jump out at you as being particularly important. Being reflexive allows you to take a step back and grapple with the larger picture, reflecting on why you might be seeing X (which is present) but also missing Y (which is also present). It also allows you to consider what effect/impact your presence has on what you are observing or being told and to make any adjustments necessary to minimize your impact or, at the very least, to be aware of these effects and talk about them in any descriptions or presentations you make. There are other ways of ensuring internal validity (e.g., member checking, triangulation), but being reflective is an essential component.
Advanced: Bourdieu on Reflexivity
One researcher who really tackled the issue of reflexivity was Pierre Bourdieu.[3] Known in the US primarily as a theorist, Bourdieu was a very capable and thorough researcher, who employed a variety of methods in his wide-ranging studies. Originally trained as an anthropologist, he became uncomfortable with the unreflective “outsider perspective” he was taught to follow. How was he supposed to observe and write about the various customs and rules of the people he was studying if he did not take into account his own supposedly neutral position in the observations? And even more interestingly, how could he write about customs and rules as if they were lifted from but outside of the understandings and practice of the people following them? When you say “God bless you” to someone who sneezes, are you really following a social custom that requires the prevention of illness through some performative verbal ritual of protection, or are you saying words out of reflex and habit? Bourdieu wondered what it meant that anthropologists were so ready to attribute meaning to actions that, to those performing them, were probably unconsidered. This caused him to ponder those deep epistemological questions about the possibilities of knowledge, particularly what we can know and truly understand about others. Throughout the following decades, as he developed his theories about the social world out of the deep and various studies he engaged in, he thought about the relationship between the researcher and the researched. He came to several conclusions about this relationship.
First, he argued that researchers needed to be reflective about their position vis-à-vis the object of study. The very fact that there is a subject and an object needs to be accounted for. Too often, he argued, the researcher forgets that part of the relationship, bracketing out the researcher entirely, as if what is being observed or studied exists entirely independently of the study. This can lead to false reports, as in the case where a blind man grasps the trunk of the elephant and claims the elephant is cylindrical, not having recognized how his own limitations of sight reduced the elephant to only one of its parts.
As mentioned previously, Bourdieu (1984) argued that “reflective analysis of the tools of analysis is not an epistemological scruple but an indispensable precondition of scientific knowledge of the object” (92). It is not that researchers are inherently biased—they are—but rather that the relationship between researcher and researched is an unnatural one that needs to be accounted for in the analysis. True and total objectivity is impossible, as researchers are human subjects themselves, called to research what interests them (or what interests their supervisors) and also inhabiting the social world. The solution to this problem is to be reflective and to account for these aspects in the analysis itself. Here is how Bourdieu explains this charge:
To adopt the viewpoint of REFLEXIVITY is not to renounce objectivity but to question the privilege of the knowing subject, which the antigenetic vision arbitrarily frees, as purely noetic, from the labor of objectification. To adopt this viewpoint is to strive to account for the empirical “subject” in the very terms of the objectivity constructed by the scientific subject (notably by situating it in a determined place in social space-time) and thereby to give oneself awareness and (possible) mastery of the constraints which may be exercised on the scientific subject via all the ties which attach it to the empirical “subject,” to its interests, motives, assumptions, beliefs, its doxa, and which it must break in order to constitute itself. (1996:207; emphases added)
Reflexivity, for Bourdieu, was a trained state of mind for the researcher, essential for proper knowledge production. Let’s use a story from Hans Christian Andersen to illustrate this point. If you remember this story from your childhood, it goes something like this: Two con artists show up in a town in which its chief monarch spends a lot of money on expensive clothes and splashy displays. They sense an opportunity to make some money out of this situation and pretend they are talented weavers from afar. They tell the vain emperor that they can make the most magnificent clothes anyone has ever seen (or not seen, as the case may be!). Because what they really do is “pretend” to weave and sew and hand the emperor thin air, which they then help him to put on in an elaborate joke. They tell him that only the very stupid and lowborn will be unable to see the magnificent clothes. Embarrassed that he can’t see them either, he pretends he can. Everyone pretends they can see clothes, when really the emperor walks around in his bare nakedness. As he parades through town, people redden and bow their heads, but no one says a thing. That is, until one child looks at the naked emperor and starts to laugh. His laughter breaks the spell, and everyone realizes the “naked truth.”
Now let us add a new thread to this story. The boy did not laugh. Years go by, and the emperor continues to wear his new clothes. At the start of every day, his aides carefully drape the “new clothes” around his naked body. Decades go by, and this is all “normal.” People don’t even see a naked emperor but a fully robed leader of the free world. A researcher, raised in this milieu, visits the palace to observe court habits. She observes the aides draping the emperor. She describes the care they take in doing so. She nowhere reports that the clothes are nonexistent because she herself has been trained to see them. She thus misses a very important fact—that there are no clothes at all! Note that it is not her individual “biases” that are getting in the way but her unreflective acceptance of the reality she inhabits that binds her to report things less accurately than she might.
In his later years, Bourdieu turned his attention to science itself and argued that the promise of modern science required reflectivity among scientists. We need to develop our reflexivity as we develop other muscles, through constant practice. Bourdieu (2004) urged researchers “to convert reflexivity into a disposition, constitutive of their scientific habitus, a reflexivity reflex, capable of acting not ex poste, on the opus operatum, but a priori, on the modus operandi” (89). In other words, we need to build into our research design an appreciation of the relationship between researcher and researched.
To do science properly is to be reflective, to be aware of the social waters in which one swims and to turn one’s researching gaze on oneself and one’s researcher position as well as on the object of the research. Above all, doing science properly requires one to acknowledge science as a social process. We are not omniscient gods, lurking above the humans we observe and talk to. We are human too.
Further Readings
Barry, Christine A., Nicky Britten, Nick Barbar, Colin Bradley, and Fiona Stevenson. 1999. “Using Reflexivity to Optimize Teamwork in Qualitative Research.” Qualitative Health Research 9(1):26–44. The coauthors explore what it means to be reflexive in a collaborative research project and use their own project investigating doctor-patient communication about prescribing as an example.
Hsiung, Ping-Chun. 2008. “Teaching Reflexivity in Qualitative Interviewing.” Teaching Sociology 36(3):211–226. As the title suggests, this article is about teaching reflexivity to those conducting interviews.
Kenway, Jane, and Julie McLeod. 2004. “Bourdieu’s Reflexive Sociology and ‘Spaces of Points of View’: Whose Reflexivity, Which Perspective?” British Journal of Sociology of Education 25(4):525–544. For a more nuanced understanding of Bourdieu’s meaning of reflexivity and how this contrasts with other understandings of the term in sociology.
Kleinsasser, Audrey M. 2000. “Researchers, Reflexivity, and Good Data: Writing to Unlearn.” Theory into Practice 39(3):155–162. Argues for the necessity of reflexivity for the production of “good data” in qualitative research.
Linabary, Jasmine R., and Stephanie A. Hamel. 2017. “Feminist Online Interviewing: Engaging Issues of Power, Resistance and Reflexivity in Practice.” Feminist Review 115:97–113. Proposes “reflexive email interviewing” as a promising method for feminist research.
Rabbidge, Michael. 2017. “Embracing Reflexivity: The Importance of Not Hiding the Mess.” TESOL Quarterly 51(4):961–971. The title here says it all.
Wacquant, Loïc J. D. 1989. “Towards a Reflexive Sociology: A Workshop with Pierre Bourdieu.” Sociological Theory 7(1):26–63. A careful examination of Bourdieu’s notion of reflexivity by one of his most earnest disciples.
- Someone might ask me if I have truly been able to “stand” in the shoes of more privileged students and if I might be overlooking similarities among college students because of my “biased” standpoint. These are questions I ask myself all the time. They have even motivated me to conduct my latest research on college students in general so that I might check my observations that working-class college students are uniquely burdened (Hurst 2019). One of the things I did find was that middle-class students, relative to upper-class students, are also relatively disadvantaged and sometimes experience (feel) that disadvantage. ↵
- Unless, of course, one is engaged in autoethnography! Even in that case, however, the point of the study should probably be about a larger phenomenon or experience that can be understood more deeply through insights that emerge in the study of the particular self, not really a study about that self. ↵
- I mentioned Pierre Bourdieu earlier in the chapter. For those who want to know more about his work, I’ve included this advanced section. Undergraduates should feel free to skip over. ↵
The practice of being conscious of and reflective upon one’s own social location and presence when conducting research. Because qualitative research often requires interaction with live humans, failing to take into account how one’s presence and prior expectations and social location affect the data collected and how analyzed may limit the reliability of the findings. This remains true even when dealing with historical archives and other content. Who we are matters when asking questions about how people experience the world because we, too, are a part of that world.
The branch of philosophy concerned with knowledge. For researchers, it is important to recognize and adopt one of the many distinguishing epistemological perspectives as part of our understanding of what questions research can address or fully answer. See, e.g., constructivism, subjectivism, and objectivism.
A statement created by the researcher declaring their own social position (often in terms of race, class, gender) and social location (e.g., junior scholar or tenured professor) vis-à-vis the research subjects or focus of study, with the goal of explaining and thereby limiting any potential biases or impacts of such position on data analyses, findings, or other research results. See also reflexivity.
Reliability is most often explained as consistency and stability in a research instrument, as in a weight scale, deemed reliable if predictable and accurate (e.g., when you put a five-pound bag of rice on the scale on Tuesday, it shows the same weight as when you put the same unopened bag on the scale Wednesday). Qualitative researchers don’t measure things in the same way, but we still must ensure that our research is reliable, meaning that if others were to conduct the same interview using our interview guide, they would get similar answers. This is one reason that reflexivity is so important to the reliability of qualitative research – we have to take steps to ensure that our own presence does not “tip the scales” in one direction or another or that, when it does, we can recognize that and make corrections. Qualitative researchers use a variety of tools to help ensure reliability, from intercoder reliability to triangulation to reflexivity.
In mostly quantitative research, validity refers to “the extent to which an empirical measure adequately reflects the real meaning of the concept under consideration” (Babbie 1990). For qualitative research purposes, practically speaking, a study or finding is valid when we are measuring or addressing what we think we are measuring or addressing. We want our representations to be accurate, as they really are, and not an artifact of our imaginations or a result of unreflected bias in our thinking.
A method of ensuring trustworthiness where the researcher shares aspects of written analysis (codes, summaries, drafts) with participants before the final write-up of the study to elicit reactions and/or corrections. Note that the researcher has the final authority on the interpretation of the data collected; this is not a way of substituting the researcher’s analytical responsibilities. See also peer debriefing.
The process of strengthening a study by employing multiple methods (most often, used in combining various qualitative methods of data collection and analysis). This is sometimes referred to as data triangulation or methodological triangulation (in contrast to investigator triangulation or theory triangulation). Contrast mixed methods.