Estimated reading time: 13 minutes, 20 seconds

The literature on CAE (Collaborative Autoethnography)

Estimated reading time: 13 minutes, 20 seconds

Image by Robbie via Flickr CC-By-SA
Image by Robbie via Flickr CC-By-SA 

I’ve been reading a bit lately about collaborative autoethnography (and it’s quite what I expected it to be). Mainly this article on mothering & this one by Geist-Martin et al) and this book (Chang et al), though I also have plans to read this open access book on (non-collaborative) autoethnography recommended by Penny. And also this open access article by Ellis et al on autoethnography (only skimmed it). I’d like to share some of what I’ve learned. I think we’re on the right track, but the readings raised a few important questions I’d like to share. Apologies in advance for what I know will be a long post. It’s not a research paper, but slightly tidied up research NOTES. So forgive the blandness & linearity of it, and bear with me as I jot these down publicly 🙂

Disclaimer: I’m not a methodological purist, I’m an omnivore & a quilt-maker. I don’t even think ethnography believes in methodological purity; the researcher is the instrument even more so if it’s auto. My approach is: I have a research question (and that question is influenced by my worldview and my research epistemology/ontology of course) – and I’ll do whatever I can (within ethical and logistical bounds) to answer it. This means that collaborative autoethnography might turn out to be only part of what answers my question. The narratives in the doc are only part, and it’s not a full collaborative autoethnography yet.

So what was MY question? It was, how are people experiencing rhizo14? What went well for them, what didn’t? [this is the open question, but I am interested in sub-topics of making connections and building community] What was yours, btw? 

Image by Torley via Flickr CC-By-SA
Image by Torley via Flickr CC-By-SA 

Why am I interested? I would like to understand how other experienced this MOOC, and it’s important to note the diverse ways in which the course was perceived by different people. I’m interested in what didn’t work. But I am also interested in what did work, and for whom. I’d like this knowledge to help influence future designers of connected courses by highlighting the participant experience. It’s important for me that this research is as participatory as possible so that no one is interpreting on behalf of anyone else; however, I realize that not everyone wants to participate and you can’t honestly have a research project negotiating with 30 people anyway, so it will always be partial. And we need to find different ways of dealing with that. Will come to it later. Why were YOU interested, btw?

Geist-Martin et al cite Ellis (2004, p. 30) on autoethnography, and it captures how I feel about this approach, why I wanted to do it:

“The goal is to practice an artful, poetic, and empathic social science in which readers can keep in their minds and feel in their bodies the complexities of concrete moments of lived experience”

As Geist-Martin et al say, doing collaborative autoethnography rejects the traditional approach of disembodied academic research.  AE “challenges the hegemony of objectivity or the artificial distancing of self from one’s research subjects” (Change et al p. 18), because analyzing oneself is something no other research method does.

One of the interesting things that came out of Chang et al is that there are three broad types of autoethnography; the type that emphasizes the auto (closer to autobiography, more narrative), OR a type that focuses on the ethnography part (more analytical, relating one’s own experiences to the wider culture) – but any AE contains elements of both. I *think* in #rhizo14 we’re attempting something closer to the latter, but what we have at the moment is closer to the former. What do others think? In any case, the practice needs to move beyond mere storytelling in order to be research, and of course that makes total sense. Autoethnography needs to “use personal stories as windows to the world, through which we interpret how their selves are connected to their sociocultural contexts and how the contexts give meanings to their experiences and perspectives” (Chang et al, p. 18-19).

I liked Geist-Martin et al’s & Chang et al’s critiques of their own process – here are some parts I wanted to highlight:

  • They looked for themes across their stories
  • They helped each other clarify certain aspects of each other’s stories
  • They critiqued and recognized ways in which their stories reproduced cultural stereotypes
  • They struggled with how to “cut” parts of their stories in order to make this paper
  • They mention how social activities they participated in, in each other’s lives, influenced how they wrote together
  • They talk about community-building that occurs because of the collaboration on the autoethnography itself
  • They raise ethical issues about how personal narratives actually refer to people outside the narrative itself and the ethics of such story-telling that will get published and scrutinized
  • Clearly, doing autoethnography collaboratively is meant to diversify the viewpoints on a topic, making the interpretation richer and more complex than just one person’s autoethnography. It also, of course, makes it more complicated to do. Easier to start than to finish 🙂 Sound familiar?

Chang et al mention 4 key dimensions of CAE:

  1. Self-focused
  2. Context-conscious
  3. Researcher-visible
  4. Critically dialogic

(I think the first three would seem tautological with the term CAE – the last, though, is worth pointing out – the more “critically dialogic”  work is, the more it tends towards an analytic/ethnographic rather than evocative/biographical type of research; I think it makes sense to  do evocative research on emotionally sensitive topics, where over-analyzing it might actually lose the essence of what is being researched; I can see it for tales of abuse, illness, etc., but not for #rhizo14 which is less of an emotionally taxing thing to talk about  – I hope!).

Some more stuff about CAE:

  • Alternation between solo and group work (already sort of happening, yeah?)
  • I don’t think I’d necessarily follow the steps exactly, but loosely (and Chang et al call it an “iterative process”), there’s data collection at the beginning (which can keep happening as gaps are found via group negotiation); there’s data analysis and interpretation (where we seem to be at – and I think that might raise areas of gaps to go find data about or to re-write our narratives about – will explain later); and of course writing.

This part in Chang et al made me laugh because of its vagueness:

“we argue that CAE as an emerging research practice should not be limited to a particular approach or style of representation as long as it holds true to the salient aspects of methodological rigor. Rather, it should yield to the demands of the research community a it is shaped by the pragmatists of social inquiry and lends itself to the true understanding of social phenomenon”

Does anyone else see the humor in the above??? But anyway, for me, what matters is that I can basically do whatever I want, call it CAE, and set my own criteria for rigor 🙂 I’m only half-kidding.

The authors suggest the following benefits of CAE  (p. 25):

  1. collective exploration of researcher subjectivity
  2. power-sharing among researcher-participants (I find this particularly important to discuss as we should never assume equal power throughout the process and need to be sensitive to this)
  3. efficiency an enrichment in the research process (by that, they mean that since the “subjects” are also the researchers, it’s easy to ask oneself for clarification, to dig up that old memory, than it would be to ask a different person outside the researchers; it also means the person can keep finding little bits and pieces to add throughout the research process up until writing time – this reminds me of sthg I read in Sarah’s narrative that resonated with my own but that I had not written in the narrative I wrote in the CAE – I had written it elsewhere on my blog – so I know it’s not that Sarah’s writing made me think something that was never in my head; it’s just something I had no written for THAT CAE. I always felt the prob with interviewing another person is that they might say something different tomorrow; you don’t need to worry about that with CAE if all the participants continue to collaborate on it – which is not the case for #rhizo14 but I’m coming to that later. The authors also talk about the benefits of working in interdisciplinary teams, each bringing a different lens to the research)
  4. deeper learning about self and other
  5. community-building (and I so believe in this – there are some people I was not close to during rhizo14’s set 6 weeks but got closer to during the process of doing this autoethnog).

I loved this quote (p. 26):

“CAE offers us a scholarly space to hold up mirrors to each other in communal self-interrogation and to explore our subjectivity in the company of one another”

Image by FlossyFlotsam via Flickr CC-BY
Image by FlossyFlotsam via Flickr CC-BY

I’ve seen this a lot in rhizo14 as we think about the autoethnog. Some of us will drift into a certain way of thinking about it, others will bring us back to something we might have missed or forgotten. Some of those others are people actively working on the autoethnog, others are not working on it but helping provide a critical eye (thanks, Frances; Alan).

I also love this quote (p. 28):

“This kind of collaborative meaning-making requires that each team members be willing to be vulnerable and open with co-researchers in order to enable the deeper analysis and interrogation that enriches the final product”

But this brings us to the challenges of CAE:

  • Risk of incomplete trust to lead to premature consensus-building that compromises the data
  • Apparently quite difficult to do at a distance because of degree of closeness needed. I know there are times I’ve been closer to the folks here at rhizo14 than I am now, so that’s something worth considering
  • Interdependency of research efforts (I think they just mean that none of us can finish this if some of us are slacking, too busy with other stuff, etc.)
  • Mutlivocality can make each researcher influenced by the voices of others (though I think that’s common in any kind of interpretive research; at least with collabo autoethnog everyone’s still there to reign others back? It’s something the researchers need to stay conscious of throughout)
  • Team effort (takes time; people need to decide on roles, commitments, etc – we have not done so)
  • Ethics & confidentiality (this prob deserves a post on its own, but I’ll just give it a section here for now)


Authors ask if CAE needs to go through IRB? Ours went through IRB. Not sure if they really understood the extent of what we were doing, but they approved it.

The biggest ethical issue I see is that when only indirectly reference others, we may be broaching on their confidentiality. It’s more obvious when someone talks about their parents, child, boss, that that person will be deduced. It’s less obvious in a large community but in many cases can still be deduced, right?

We also need to be clear on who gets  access to the data after we write our “report”, and how they can use it… individually, collectively, etc?

We as individual autoethnographers also need to recognize the need to protect ourselves – how much are we revealing about ourselves and is it OK that all of that becomes open to public scrutiny as we publish it? The incident over the use of our data during #et4online by Jen Ross and Amy Collier was a case in point – it is not that simple.

Image by Page Dooley  CC-BY via Flickr
Image by Page Dooley CC-BY via Flickr

I’m now reading Ch 5 of that book about the data analysis side of things. I’ve read the part about using an emerging coding approach, such as what we are doing now.

I won’t blog about that chapter as I think I’ve already come close to violating copyright of the first chapter of that book by blogging the main ideas here. But I’ll write notes about it in our private google doc.

I said throughout this post there were issues I’d come back to later. My brain is fried right now, and I can only find one of them, but I want to publish this post, so I’ll just come back to one MAIN point that’s running through my mind (well, points, plural, but they are all related):

  • CAE implies that only the authors’ stories are told. Now the authors could react to stuff that happened by and with other people, but there are ethical issues in getting to deep with that as it exposes the others implicated in our stories
  • How do we incorporate  the views of people who wrote narratives in the autoethnog but who are not part of the team currently analyzing the data? It’s not autoethnog if we’re using their stories – they’re THEIR autoethnogs
  • Can we get multiple autoethnogs out of this, with every 5-8 people writing their own version? That’s be wayyyy cool but not everyone wants to ever write that, so…
  • Can we use some of the other data in the narratives DIFFERENTLY? So not as autoethnog, but as narratives, interpreted by researchers (like regular research)
  • The inherent “connectdness” of it all makes it almost paralyzing to imagine how we can tell our own stories (6-7 of us) without either implicating others, or needing to reference others
  • I usually do ethnography by using any and all data I can; this would mean referencing public blogs, etc. We can get permission to do that. But it’s not auto if those blogs are not our OWN, of the people WRITING the CAE.
  • I keep circling back to the same thing, right? There power questions, there are questions of who can tell whose story? There are multiple “others” in the “we” of autoethnography, and what do we do by telling our story and leaving out theirs? What about the people who do not wish to be researchers but whose stories are in the narratives? What about the people who didn’t contribute to the narratives but were rhizo14ers? What about the people who didn’t even blog visibly or at all, and so have no easy “trace” to find even if we wanted to incorporate their views?

So where I stand right now is this: the CAE looks like it would focus on the narratives of the people who are now analyzing the data. The rest of the data needs to be treated in some other way that we need to negotiate with the people who wrote it.

For me, trying to answer our research question is more important than whatever methodology, but the methodology matters, the participatory nature of it matters.

So I’m seeing maybe a paper that comes out of a “proper” CAE where only writers’ narratives are included, with analysis and interpretation of course; and I see some other negotiated decision about how to deal with the rest of the data and who has access to it.

The plus side of all this is that this narrows our dataset down to just the few of us working on analysis, and sounds more manageable, as in, I think we can actually finish this and publish it someday this century 🙂

The downside is, it won’t really be the story of rhizo14, it’ll be “our” stories of rhizo14. But it’s better than none, and it’s definitely worth sharing with the world. And there’s no need to “stop” there. It’s just a milestone.

But my views are not final on this, it’s just where I stand right now. So I’d love to hear what others think

[Note: this post is CC-BY-SA unlike the rest of my blog – due to the use of images that require me to share it this way]

24 thoughts on “The literature on CAE (Collaborative Autoethnography)

  1. wow. I was quickly browsing twitter and came across your tweet to this post and parts of your convo – your app thingy to auto-post tweets to the comments worked! That’s pretty cool : ). It gets around the transient nature of tweets, capturing more of these conversations in one place (maybe now you just need to do the same for facebook…although I wonder if that has different ethics associated with it, since FB is a closed network…)

    Anyway, great to get some more insight into the autoethnography, and the questions being considered, as well as insight into the methodology. I always found the idea of producing something multidimensional that goes towards reflecting the experience of rhizo14 quite intriguing…but I guess that may just be too hard..? I think essentially it’s the challenge of trying to communicate a non-linear, multi-dimensional experience in a linear, single dimensional format, in order to make sense of it, and have others make sense of it. It’s a similar issue trying to storify a really fast and manic tweet chat…just on a much more complex scale….

    Look forward to hearing more about it as it progresses…..happy to help if and when I might have time (now is not great which is why I haven’t involved myself…still keen to follow though).

    1. Hey Tanya

      I am thinking there should be two directions at least: one multimedia dimension that attempts to capture data in non-linear ways, links to blogposts, makes, etc.
      And a second scholarly option which analyzes the data in the autoethnog narratives (which can produce a paper) but continues also beyond that to explore the experience further using data beyond the narratives of the researchers.
      How that latter part might complement the research by Frances,Jenny, and Mariana, is something we’ll have to discuss with them (as I suspect their data is confidential)

  2. “The goal is to practice an artful, poetic, and empathic social science in which readers can keep in their minds and feel in their bodies the complexities of concrete moments of lived experience”
    As Geist-Martin et al say, doing collaborative autoethnography rejects the traditional approach of disembodied academic research. AE “challenges the hegemony of objectivity or the artificial distancing of self from one’s research subjects” (Change et al p. 18), because analyzing oneself is something no other research method does.

    This is what I have been banging on about with my feeling about ‘science bound’. It is a question of choosing a means by which to tell or not ones’s story. For me all expressions of research or other expressions say more about the author and his/her context than the others studied. Who has access to the tools/methodology/money/communities to publish? How is that defined?

    What concerns me in all of that is to what extent research in social sciences and education in particular is simply a reinforcement of a power structure rather than a means to critically evaluate and transform it. – LInks to Freire here.

    The whole question of what constitutes ‘history’. The whole question of the limits of knowledge of what can be known.

    How do we incorporate the views of people who wrote narratives in the autoethnog but who are not part of the team currently analyzing the data? It’s not autoethnog if we’re using their stories – they’re THEIR autoethnogs

    As soon as I see/read/connect with others they become part of my story even if it is not explicit in what I express. I can see constant mirroring of expression/thought going on in rhizo14. Much of it is unconscious. When I learn a new language – can I cite each person with whom I learnt a new expression?

    Example Parse – Jennifer Maddrell.

    What is the continuum of authorship? When do I reference an influence? This is where I personally would like to go back and see how I am/have been mapping interaction in this dialogue space.

    I have the strong feeling that this is about becoming researcher.

  3. It just occurred to me from from a trip to the hospital that we are all learners within learners within learners. For instance, doctors learn in school and then become diagnosticians who learning aids them in a higher level of learning to extract information from patients in order to treat them. In my situation, the doctors have reached the level of “learned” in order to practice medicine but have lost the learning skill and just pronounce whatever it is that comes into their fat heads. They reason from a position assumed to be qualified but guarantee they are guessing–and maybe not even that.

    In other words the learning has ended and the pronouncing has begun. In the auto ethnog I suspect most of us remain in a state of what might be called “incompleteness” in order to access our learning spirits. Not sure this makes sense but to approach the “individual in progress” that is represented by our ethnographic statements brings up the odd image of catching someone in a kind of ethereal context that escapes conclusions?

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.