orgtheory.net

organizational ethnography and pseudonyms: at what cost?

‘Tis the season to talk about behind- the-scenes techniques and public presentation of ethnography. 

At the Eastern Sociological Society meeting this past spring, Colin Jerolmack gave a compelling talk about the convention of ethnographic pseudonyms, based on a paper he’s writing with Alexandra Murphy, slated for a special issue of Sociological Methods and Research. They find much to criticize about the practice of making up names for people and places: that using pseudonyms creates a false sense of generalizability (‘Middletown is every town”), impedes the accumulation of knowledge by preventing revisits, and flat out is ineffective in many cases. They also argue that it is presumptuous of the researcher to assume s/he knows better than those in the field site if their own identities should be concealed. A key take-away is that pseudonyms should not be our default; we should have good justifications for using them.

While the talk/paper focuses on small group and community ethnography, some unique issues arise with naming organizations and organizational participants. At the talk, Fred Wherry raised the point that non-profit organizations risk losing funding if certain practices are revealed. Other issues arise. People’s jobs may be on the line. Potentially the researcher could be liable for defamation? Using the real name of the organization but pseudonyms for individuals is not ideal; we often are discussing people vis-a-vis their offices and positions, so it would be very easy for readers to identify them.

In my ethnography of diversity management at “Starr Corporation” a Fortune 500 multinational corporation, my written agreement with the company stipulated that I would disguise the identity of the company and individual employees. This made it easier for me to gain access initially. I think it also made employees more likely to open up to me, but then again, I’m not so sure how much comfort they took in my assurances of their anonymity. 

Although disguising an organization’s identity can help us get in the door, it can create frustrating limitations for the final written product. In my book, I discuss managerial and corporate power dynamics, but I could have made a stronger point if I had been able to explain just how large and powerful Starr Corporation is. The real name would have gone a long way. Crucially, concealing the company’s identity meant I could not elaborate the type of products the company sells, which is relevant for its diversity management strategies. There were some images from Starr’s annual corporate report I would have liked to include in my book, to illustrate the changing meanings of race and gender alongside shifts in corporate logic and goals. I don’t think any of this prevented me from doing a serious analysis, but the constraints have been irritating.

And how can we represent an organization’s public discourse while keeping its identity concealed, given how easy it is to search specific text on the internet? I set as my standard whether a sentence could be a) searched on Google and b) associated with the company c) at the time I wrote the book. In my writing, I have used only short phrases from Starr’s job advertisements. For the company’s business case for diversity, I used the internal policy and a version from the late 1990s pre-internet days that I got from the company’s private corporate archives; the message hadn’t changed in any meaningful way. But often it’s difficult to come up with those kinds of work arounds. 

What do you think about using pseudonyms for organizational ethnography? Does it not make a difference to you—as the researcher or a reader—or do you find that something meaningful is lost in translation? 

Written by ellenberrey

June 9, 2015 at 2:19 am

Posted in ethnography

11 Responses

Subscribe to comments with RSS.

  1. In my doctoral research (which was historical rather than ethnographic, but fits the general discussion here anyway), I did not promise anonymity to the organizations (colleges and universities) that I studied, though I did not need their approval anyway. I did promise anonymity to individuals I spoke within in relation to those organizations. At a late stage in the research, a particular department at one university asked me to remove them from my study–an interesting request, since I had never contacted them to ask for their participation and did not need their consent to write about them. However, in response to that request, my committee advised that I give the colleges and universities pseudonyms, and I kept that practice in the book that resulted. I believe, of course, that everything I wrote in the dissertation and the book is accurate, but the experience solidified my belief that using real identities for the organizations we study enhances intellectual honesty, and it allows for, as @ellenberry writes, the use of specific identifying data that is important to the understanding of the case. In addition, I suspect it is quite possible for people to determine the identities of at least some of the organizations I wrote about, making the exercise seem kind of silly. But since that’s the norm and I was a graduate student when I collected the data, I felt little choice but to go along with it.

    Like

    Mikaila

    June 9, 2015 at 3:16 am

  2. I cannot speak for other forms of organization, but as a corporate ethnographer I will say this: most corporate ethnography would not be possible, or would change dramatically, were anonymity not promised for the organizations and employees.

    Not one company I have studied would have allowed me my initial access without promises of anonymity to the corporation and its employees. These were terms they demanded. That is not to say you can’t find an example of a corporation that would allow itself to be named or its employees to be identified, but simply that, on average, these protections make this sort of work positive.

    Executives with large salaries will not often put their reputation or jobs – or their companies’ reputation and market value – at risk to allow a researcher in. Promising anonymity can be the one way to get around this and get access. If you do find companies willing to be named, you risk getting shallow access and the company-preferred story given concerns over reputation management. Many business trade books in which companies are named are like that.

    (Note that corporations sign non-disclosure and confidentiality agreements with nearly all parties they let into their backstage – partners, consultants, etc. Corporate ethnographers are often asking for access to, and the ability to write about, precisely what these agreements define as confidential information, so working out an agreement over anonymity can be the only way to get that access. Another data point is that in a large MBA class at my school, students do team projects with a company of their choosing. For a one-semester, student project, which will only be seen by the student team and faculty member, the companies almost all require students to sign confidentiality agreements and sometimes even then to anonymize the organization in anything written for the faculty member. The stakes are even higher when the organization believes you will be writing a book or articles for publication.)

    Regarding employees, if you want employees to share their honest impressions and perspectives with you, you need to protect their identities. People’s jobs and relationships are at stake. They know this and it will influence what they share if you do not promise to protect their identity. And, furthermore, you should want to protect their identities to ensure your research does not have adverse effects.

    I agree that these issues become even more complicated when studying a company’s public discourses, and that the constraints anonymity impose are not to be underestimated either. So, depending on the question, some scholars may weigh the tradeoffs and decide that identifying the company is best for the particular study. The point is that coming up with a blanket rule for things like this is ill-conceived.

    Sure, we can move to naming companies and employees, but fewer corporate ethnographies will then get done, and the ones that do may well be of a very different sort. We will likely see fewer and fewer rich studies that challenges aspects of corporate life like Kanter’s Men and Women of the Corporation, Kunda’s Engineering Culture, and Morrill’s the Executive Way, etc.

    As a final note, anonymity protects the researcher in allowing him/her to tell the story as they saw it. However, responsible corporate ethnographers share their work with the company and employees at the end, not as a condition for publication but to give back and to engage in dialogue about it. Knowing you will share your study with your subjects can be a powerful disciplining force to doing fair ethnography.

    Liked by 3 people

    anon

    June 9, 2015 at 7:46 am

  3. I agree with anon. Ellen, I get that you find the anonymity thing constraining, and that you are skeptical as to whether you can ever keep peoples’ identities truly anonymous, but I didnt get any sense of whether you think you could have ever gotten access to Starr corporation without it. If you don’t think that you could have, I think you should be more clear about that since that’s the big issue.

    Like

    Guest

    June 9, 2015 at 11:28 am

  4. My experience studying government officials in a foreign country is counterintuitive. The interviewees assumed they would be on the record and agreed to talk to me anyway. Whenever I mentioned, followibg protocol, that their real names would not be used, they clammed up. I could almost see the thought pop up in their heads: “oh, maybe I should be concerned about this!” In the end, I used pseudonyms throughout, but rarely told interviewees i would do it.

    Like

    Serave

    June 9, 2015 at 1:30 pm

  5. Mikaila- what a story! I find myself especially frustrated by qualitative studies of universities that try to disguise the organization’s identity. Often it seems pointless. In many cases it doesn’t work and as you say, is just plain silly. My position on this is probably biased by my experiencing doing an ethnography and historical research on affirmative action at the University of Michigan. It was impossible and undesirable to hide the university’s identity, given its involvement in two major Supreme Court cases. This suggests that our conventions for organizational pseudonyms should account for our objectives- whether they are illustrating generic social processes (but even then…) or accounting for historically and socially significant actions and events.

    These are great points, anon. I agree with most of what you’ve written in regards to corporate ethnography– except your statement that we have an obligation to share our findings with the company. I am doing as much, but when research has the objective of speaking truth to power, engaging that dialogue could prevent the researcher from doing a critical analysis. Sharing such findings may do no good and have the pernicious effect of making companies even less likely to allow ethnographers in. Also, the organization might find it more useful for the researcher to present different sorts of findings, such as a report or presentation on an specific issue that people within the company define as relevant.

    Guest- I will never know if I could have done the ethnography of Starr otherwise. At the outset, I proposed anonymity as an option, and the company took me up on that. That’s how I managed the negotiations, as I was very concerned they would turn down my request. I have wondered many times if I should have tried negotiating differently.

    I should be clear that while I found Jerolmack’s presentation to be persuasive on many counts, I remain skeptical that the argument works for organizational ethnography, esp. studies of companies and large non-profits. I think this is indicative of a broader issue within the field of ethnography, in which community and small group studies are treated as the norm (which is likely connected to the roots of sociology in the Chicago school); organizational ethnography gets less attention in the field, especially its insights into organizational structures (vs its insights into general social processes, such as dying).

    Liked by 1 person

    ellenberrey

    June 9, 2015 at 1:56 pm

  6. I have faced similar issues as the ones described by @ellenberrey. I studied a social movement that had received quite a bit of media attention, and some leaders were used to getting their name out there in the news. Some of them told me explicitly they wanted their name to appear in the study! But the ethical review board at my university insisted that I use pseudonyms instead of real names. In this case I feel that this requirement of anonymity was not doing anything to protect informants, it was simply the meaningless application of a taken-for-granted rule by an inflexible ethics review board. Part of the issue, I believe, is that some members of ethical review boards have little to no experience with field research and ethnography. They may be familiar with research designs like experiments, where the identity of the research subject is irrelevant, and thus they associate “real names” with “risk.” I think there is a need to educate ethics review boards on the limitations associated with the use of pseudonyms. Thank you Ellen for opening up this conversation.

    Like

    jib

    June 9, 2015 at 2:16 pm

  7. @ellenberry, with respect to your last question, I find myself of two minds. As a reader, I like transparency and knowing the subject organizations. I know I shouldn’t trust the ethnography more because of this, but I do. As someone who reports on this type of research in my classroom, I know that relating specific phenomena to specific contexts resonates better with the students — this is lost in pseudonyms. But as a researcher who is constantly doing field research in organizations, I understand (a) the need for confidentiality or anonymity to gain access and high-quality discourse with persons at different spots in the hierarchy, and (2) the hassle I will get from the IRB to describe any protocol without anonymity of the respondents, as they immediately invoke “social risk”. About ½ the time I get away with keeping real organization names and using pseudonyms or ersatz titles for persons lower in the hierarchy.

    I haven’t been inside an organization in more than 25 years without a promise for feedback. As many organizations will tell you, they get untold numbers of requests for interviews, surveys, and other data collection from researchers that want to study something amusing to themselves. They would like to offer access and time to researchers who give something back and ignore the rest.

    Like

    Randy

    June 9, 2015 at 3:27 pm

  8. Jib- that makes sense. I’ve found that marginalized activists want me to use their real names, as did Michigan administrators, whose admissions policy had been thoroughly vetted by lawyers. Perhaps they wanted to publicize their careful adherence to law following the 2003 Gratz and Grutter decisions.

    Randy- you’ve raised the important point of how pseudonyms compromise our ability to communicate with audiences other than fellow academic researchers. In addition to students, this applies to journalists, historians, law scholars, and I’m sure others.

    Like

    ellenberrey

    June 9, 2015 at 3:48 pm

  9. I remember an article a couple of years ago, maybe in the Chronicle though I can’t find it right now, concerning research on sex-positive activists who want their names associated with their stories since that is their whole point and the difficulties in explaining to them how the IRB requires you to keep them anonymous for their own protection.

    On Randy and ellenberrey’s point about other audiences: see Steven Lubet’s critiques of Alice Goffman, which focus, in part, on his belief that all responsible scholarship requires transparency of sources. Of course, Lubet as a law professor ought to know that there are laws governing our use of human subjects that required Goffman to act as she did, and presumably he has ulterior motives for making that point, but it does speak to the contextually-specific nature of anonymity as a research practice.

    Like

    Mikaila

    June 9, 2015 at 7:57 pm

  10. I’m not an ethnographer (although as an advisor I’ve worked with ethnographers), but logged some time on an IRB and can agree with those who say IRBs often impose senseless anonymity requirements. It is possible to push back against them with an informed consent protocol that has the option of being named by name. One problem, alluded to by ellenberrey, is that researchers often don’t know in advance how people will feel about having their names used and so cannot anticipate the desire to be known by name until they are already in the field and operating under an IRB protocol that requires anonymity.

    Like

    olderwoman

    June 9, 2015 at 9:24 pm

  11. This is an interesting and relevant discussion. I’d like to add two points, not so much from a practical standpoint of gaining and using access, but rather from a theoretical perspective on case studies and what we can learn from them.

    Basically, case studies should always tell us something going beyond the case itself. I.e. cases are cases of something–and what we find in one particular case should be broadly generalizable, within boundaries, of course. What this suggests is that while it’s true that using the real name would perhaps provide some interesting additional context information, a pseudonym should not compromise the theoretical insights a case study offers.

    Having said this, there is also an argument to make for using the real names of the organization(s), given that research should also be replicable, at least in theory. It’s only when we as readers/reviewers/audience know the real name of the organization that we can ‘test’ or rather critically assess to what extent we feel the argument holds.

    Like

    Johann Fortwengel

    June 22, 2015 at 5:28 am


Comments are closed.

%d bloggers like this: