poll: is ethnography completely different than other kinds of research?
Click on all answers that apply.
Although I have published an ethnography article and a book of interviews and historical research, I do not have the habitus of the typical qualitative sociologist. In talking to ethnographers and other qualitative researchers, I often get the feeling that they are openly hostile or critical to the ideas that motivate quantitative research. Samples, inferences, and general conclusions are anathema. Is this just a posture? Or do qualitative researchers think that they are doing something completely different? If they aren’t looking for applicable lessons, are they just looking for well documented “just so” stories? Or is it merely an exercise in distinction, where ethnographers and other qualitative scholars use a different rhetoric to bolster their academic standing?
Adverts: From Black Power/Grad Skool Rulz
Qualitative researchers who are openly *hostile* to (as opposed to merely critical of) quantitative methods feel guilty about their math abilities. This doesn’t mean they aren’t or couldn’t be as skilled as those using quantitative methods, but for whatever reason they’re intimidated. This is why their attacks are so presumptuous: “they don’t capture the richness of the social” or “you can’t quantify the social”. They won’t even acknowledge the possibility that statistical models, whether of survey data or qualitative content, could help us learn something about the social world. The smallness they feel because of their math guilt makes them attack those who feel more proficient. I’d actually be interested in hearing OrgTheorists ideas about why people tend to have identities about their math proficiency “I suck at math/I’m good at math” but not “I suck at interpreting things/I’m good at interpreting things.”
This does NOT mean that all qualitative researchers are like this. And it does not mean there are no critiques of quantitative methods to be made, although the most persuasive ones I’ve read (Andy Abbott + Andrew Gelman + John Martin) tend to be by quantitative researchers.
That said, my guess is that there are also quantitative researchers who feel social guilt — they’re terrified of spending significant amounts of time with strangers, so they don’t want to believe you can learn anything from interviews or ethnography. And you’ll hear from them stuff like, “qualitative work is just bullshit” or “you can make up anything you want”.
In case you’re wondering, I’d say I’m a 70% qualitative/30% quantitative researcher.
NotYourTypicalHabitus
February 28, 2013 at 12:52 am
The poll excludes respondents who disagree with all four statements.
Pretendous
February 28, 2013 at 1:09 am
As someone who uses whatever methods best suit my research question, I go back to Bernice Pescosolido’s response to John Levi Martin at an IU Sociology colloquium, “Aren’t we beyond this debate already?” Kristin Luker’s (2008) take on it is helpful. Qualitative research is good at examining categories which are in flux and changing. Such categories are impossible to study quantitatively because they mean different things to different people. Quantitative research is better at studying cultural categories which are more stable and more or less mean the same thing to respondents. From qualitative research we aim to make theoretical inferences. Making some kind of larger inference is the point that makes it worthwhile afterall isn’t it? You also have Vaisey’s (2009) perspective that different methods may better capture different ways people cognitively function in the world.
That being said, certain audiences respect qualitative research less than quantitative research. For example, some high level businessmen I’ve come across in my ethnographic work or interviews do not think qualitative research is valuable, and such encounters force you to defend your methods and explain why they have to be used with certain research questions. I think to many audiences within and outside of the social sciences, you have to explain why you are using qualitative methods (and that it’s Not that you are bad at math). Quantitative researchers don’t have to do this.
jaimekucinskas
February 28, 2013 at 9:07 am
This is some quality trolling, Fabio. Next up, should sociology move away from culture? Inquiring minds just want to have a conversation about it. Here’s a poll.
Peter
February 28, 2013 at 2:10 pm
Fabio: what you should have asked is — “How does it feel to be an ethnographic researcher? Very bad, somewhat bad, neither bad nor good, somewhat good, or very good?”
Steve Vaisey
February 28, 2013 at 3:23 pm
^Indeed. And then assigned a cardinal value to those responses and run a single-variate regression to “revolutionize the field” and alienate colleagues.
Graham Peterson
February 28, 2013 at 3:42 pm
@NotYourTypicalHabitus: The most prominent and vocal methodological critics of formal and quantitative (they’re *very* different) methods are people whom have extensive mathematical training and are well qualified to do so. Ken Arrow is a good example. A similar situation applies: Jesus was a smart and downright classy guy. His followers are not always.
But math phobia is well worth exploring more. When one finally gets to advanced mathematics courses for instance and starts doing problem sets with peers because they’re so crushing hard, one discovers quickly that everyone else whom wasn’t raising their hand either for the last two/three years were similarly intimidated by the often times belligerent arrogance and commonly abusive and mocking pedagogy of mathematics instructors. In response to a question a graduate student had, a forerunning topologist once responded, “That’s a stupid question. Do not ever ask me a question again.”
Graham Peterson
February 28, 2013 at 4:07 pm
The book that comes closest to articulating what this crude poll is aiming at is Gary Goertz and Jim Mahoney’s new Tale of Two Cultures. Though in their case they are comparing comparative-historical, qualitative inference with maximum likelihood, quantitative inference, they make a strong case that these two research traditions are essentially parallel cultures in terms of their underlying logic of inference (set-theoretic versus probablistic) and methods (process tracing versus regression). I can imagine something similar, a Tale of Three Cultures which puts stats, comp-hist, and ethnography as three different research traditions side-by-side.
DR
February 28, 2013 at 7:16 pm
Or, maybe, they’re hostile to the study of the sorts of phenomena you must study — constrained as you are by your own methodological choices — and use a different rhetoric not to bolster their standing but, rather, to mark the very difference between the phenomena to which they attend and those to which you do.
MDH
February 28, 2013 at 8:07 pm
Article from the atlantic the other day worth a read
http://www.theatlantic.com/magazine/archive/2013/03/anthropology-inc/309218/
Pablo
February 28, 2013 at 8:11 pm
^That’s actually how Nike developed their wildly successful SB Dunk line of boutique skate shoes. They sent NYU anthropology grads to take ethnographies at skate parks, notices all the kids were wearing vintage Dunks they’d found at second hand shops, and sparked a critical mass threshold by reissuing the shoe immediately and only selling them at small skateshops – cuz skaters hate the man. Interestingly, Nike had previously failed to enter the skate shoe market because of the “corporate” symbolism attached to their brand. They’d buy a small skate shoe manufacturer and as soon as everyone found out Nike had bought it – they’d stop buying the shoes. Two leasons: ethnographic research is useful stuff, and skaters have incoherent ideas about markets.
Graham Peterson
February 28, 2013 at 9:12 pm
One item that made me hesitate is the extent to which one seeks or can generalize such findings. Ethnographic research traditionally focuses on studying a single culture or subculture, so generalizing findings beyond that (sub)culture would be inherently problematic.I also think there is a tendency towards equating ethnography and qualitative research, i.e., not all qualitative research is ethnographic.
The trolling about qualitative researchers not being very rigorous is a trope; increasingly, with the use of statistical software analysis, quantitative researchers can provide a wealth of analysis without having to justify its merit or utility. Similarly, qualitative analysis software programs have become increasingly rigorous and sophisticated.
candrews1
February 28, 2013 at 10:07 pm
Wow, just wow. I don’t even know where to start. How can so many smart people be so lunk-headed and ignorant?
Joey
March 1, 2013 at 12:26 am
Potentially because people sympathetic to ethnographic research throw around pejoratives like lunk-headed and ignorant.
Signed,
Sympathetic to Ethnographic Research
Graham Peterson
March 1, 2013 at 12:29 am
–> two questions
1. doesn’t the bigger picture, that one that contains this conversation, have to do with what you aim to do with research and what aims research can serve?
2. is it a problem, per se, if for some research “Samples, inferences, and general conclusions* are anathema”?
to the first question, i offer a short answer below. to the second question (informed by analysis of the first question, and discussed broadly below), i offer an even shorter answer: no.
–> re: question one
i start with a little piece from McGrath, cite below. (i find the first ten pages from ‘dilemmatics’ helpful and would be happy to provide more information on why if questions arise… there is, as well, much more to extract from the ten pages than what is explicitly stated. of course the piece has some weaknesses too, but i overlook/gloss that for succinctness)
McGrath, J. E. 1981. Dilemmatics The Study of Research Choices and Dilemmas. American Behavioral Scientist, 25(2): 179–210.
every piece of social science research is concerned with Actors performing some kind of Behavior within a Context (i.e., a place/time setting). figure two in McGrath’s piece is particularly helpful to show how different research approaches optimize on specific areas. i extrapolate from those ten pages, and assert that a first step (grossly simplified, of course) is to situate and see how the puzzle pieces of the three core research traditions that inform organization theory fit together; that is, how their central ‘thing’ of concern, and the methods of their craft all fit together. we have, most broadly, economics, psychology, and sociology that tend to inform our understanding of organizations. one is concerned with generalizability, another with precision of measurement of behavior, and the remaining one with system characteristics.
within each of these traditions, there are varying levels of concern with generality, precision, and setting. we have to have some of each three, but the way that each is or is not considered can lead to the wide range of research aims and methods we observe in the real world; not just in org theory, but social science most generally, that is, if we think of science as the our superordinate field, and physical and social science as being the two main subordinate levels underneath the umbrella of science. there are a multitude of subfields in social science that prize many different combinations of generality, precision, and setting. (logically, perfect generality would be perfectly imprecise and uncontextual?)
so it seems that there is not just *an ethnography method*… there is more than one way of doing ethnography (or in McGrath’s terms, more than one way of doing a field study). ethnography may range from the type of detailed study that encompasses a rich, time consuming, effort intensive ethnographic study which may optimizes very heavily on context and gives up as much as necessary on generality and precision. on the other hand, a researcher may engage in an ethnography (which necessarily must first be focused on context) that secondarily optimizes (as much as possible) on some balance between generalizability and precision (in theory, one could split the two; in practice, i cannot think of a good/satisfying example of work that does this). for example, we might have microsociologists who are very interested in bringing some precision to their work and apply some kind of field study methodology to do this (that is, some microsociologists may set out to do an ethnography that also pins down behaviorial attributes as precisely as possible) or we could study field level phenomenon and do macro-macro causation where any kind of precise behavioral assumption is suspect (a la Jepperson and Meyer’s argument) but generality is a welcome, perhaps even pursed, objective.
anyway, going deeper on this leaves room for some great discussion on the aims of research (or was our blogger more simply goading us for some trolling of the academic variety for his own wicked pleasures?). i do find that the substantive issues are time consuming to digest, but worthwhile to understand as deeply as possible… or is this just due to my own limitations cognitively and situationally, i.e., as a third year phd student? take that bit of context about me and generalize for what you will :-)
*Fabio – speaking of general conclusions, i recall another discussion you brought up not too long ago that started with institutional theory, and which I found particularly helpful when Jerry Davis brought up his piece (incidentally, which I have had several people now read, or at least skim over) about the place of theory. while I disagree with being atheoretical (is that even possible in our field? don’t our old organizational theories, as crappy as they may be, inform what the interesting questions are? as Popper has written, if all we are interested in are substantive findings about relationships in our world, then we could go measure the height, width, and depth of all the books in the London library and marvel at how well those dimensions correlate with the weights of those books. also, if we leave theory behind all together, then we leave behind the question of generalizability as well? at least that is what McGrath’s argument says to me). don’t get me wrong, i do think that theory is overwrought… and this is another discussion altogether.
pm
March 1, 2013 at 12:35 am
[...] poll: is ethnography completely different than other kinds of research? (orgtheory.wordpress.com) [...]
Qualitative research: Ethnography, and Netnography | Marketer737
March 1, 2013 at 3:10 pm
“increasingly, with the use of statistical software analysis, quantitative researchers can provide a wealth of analysis without having to justify its merit or utility”
This cynicism has no basis in reality.
Please show me an example of a published paper using quantitative methods, in particular one claiming to use a novel approach, that did not attempt to justify its merit or utility.
jamesdean@thefifties.edu
March 9, 2013 at 7:21 am