At present, my doctoral is a purely ethnographic study. In criminology, many such studies have been published as books (e.g. the works of Elijah Anderson, Dick Hobbs, Coretta Phillips). However, as regards academic papers/theses, even ethnographic studies seem to all have a component of stand-alone analysis: e.g. content analysis, discourse analyis, using NVivo or other such programs. In short, I'm trying to ask the following: if a thesis aims to purely be an ethnographic study, what explicit modes of analysis must be used (if at all)? And does this have to include computer-programs for data analysis?
Although grounded theory is now heavily identified with interviewing, all of the original studies by Glaser and Strauss (e.g., Time for Dying, Awareness of Dying, etc.) used what a sociologist would call participant observation, which is pretty close to what others would call ethnography.
From an ethnographic perspective, the most important component of grounded theory would probably be "theoretical sampling," where you use field notes and memos to keep track of your emerging theoretical insights, and then select further further types of observations (or interviews), based on what would do the most to advance your current insights.
Dev - in my experience (and you do highlight it) - the term 'pure' ethnography is a bit misleading. I personally would not use the term. One of the reasons might actually answer the question. Unlike other forms of qualitative approach i.e. grounded theory, IPA etc, which commonly subscribe to quite structured analytical step-wise processes - ethnographical approaches do not have to. It can adopt a more 'free-form' style. Subsequently, many ethnographical studies differ in their approach. This can be both an advantage and a disadvantage - especially I would argue at the doctoral level. I would discuss with your supervisors the types of examiners that you might end up with. They may be known to prefer certain approaches in certain disciplines. I would not leave it 'too loose' or 'free-form' though - as this would be harder to defend and ensure rigour. At the doctoral level - I would encourage the use of NVivo - not as you have to - but purely from the perspective that wider field work will generate very large narrative/observational data-sets - and it's easier to manage.
Dev, I agree with Dean. It is important to check with your supervisors if this is an appropriate approach to the subject you would like to study. One of the challenges of ethnographical research is how to manage huge amount of data, often with a very complex discursive "tissue". NVivo helps and I think Atlas.ti too.
Thank you, Dean and Carmen. As you say, Dean, I think I will have to bear in mind the level at which I am submitting - i.e. doctoral, and to make sure it's not too free flowing to hinder defending the work at the viva stage
Although grounded theory is now heavily identified with interviewing, all of the original studies by Glaser and Strauss (e.g., Time for Dying, Awareness of Dying, etc.) used what a sociologist would call participant observation, which is pretty close to what others would call ethnography.
From an ethnographic perspective, the most important component of grounded theory would probably be "theoretical sampling," where you use field notes and memos to keep track of your emerging theoretical insights, and then select further further types of observations (or interviews), based on what would do the most to advance your current insights.
If we understand ethnography as social relationship, in Colombia, the "method" is based on the purpose.
http://www.luguiva.net/admin/pdfs/Mauricio%20Pardo.pdf
http://www.luguiva.net/articulos/detalle.aspx?id=80
https://www.researchgate.net/publication/261496266_El_mal_de_ojo_de_la_etnografa_clsica_y_la_limpia_posmoderna
Article El mal de ojo de la etnografía clásica y la limpia posmoderna
if you wish to combine ethnography and grounded theory, I suggest to read the Handbook of Grounded Theory, Brayant and Charmaz (2007). This chapter is exactly about this topic: Timmermans S. and Tavory, I., 2007. Advancing Ethnographic research trough grounded theory practice. In: A. Bryant and K. Charmaz. eds. 2007. The Sage Handbook of Grounded Theory. London: Sage. pp. 493-512.
Are you trying to build a descriptive case study from your ethnographic method? Even if you seek primarily to "describe" the phenomenon or people you observe (either through observation or interviews or both), you need some way of pulling together the information; to sift and sort it so that eventually, you have a picture of what is unique about the case or what is most important to tell your readers. What kinds of questions do you hope to gain answers to from the ethnographic study? This provides the theoretical framework for your dissertation. Grounded theory is one approach to analysis (theory emerges from data), as is discourse analysis (analyzing the conversations, language, and interactions between participants reflected in your data). Other theoretical lenses may also be utilized depending on your data: feminist theory, race theory, social interaction theory, etc.- what you choose provides the perspective from which you analyze the data.
NVivo, on the other hand, is not an analytic lens (theoretical framework), but a tool to help you study and understand your ethnographic data. There are other computer software programs for this purpose, too; they each have advantages and disadvantages - but all are intended to help you organize and code your data. You also may do this manually but be warned, it can be very time consuming, even using software! Be careful not to conflate the analytic lens you may choose with the instruments you choose to help you in that analysis. The two work hand in glove but are separate entities.
Dear Dev,
I think that you must go back to your paradigm, as this will give you a clue.
I found the actual practical 'recipe' for data analysis a challenge when doing my PhD. You and your supervisor also need to decide what kind of 'pure' ethnography will you follow which is guided by your paradigm and then follow the work and suggested methods of a specific author. There are no generic 'recipes'. I followed Spradley and also used Saldanha, J - The coding manual for qualitative researchers to understand memoing, coding etc.
I hope this helps
Jane
I agree with Deborah that NVivo and other software programs are tools that can be used with a variety of different analysis methods, including content analysis and grounded theory. The strategy there is to enter your field notes, along with any primary documents etc., into the software and do your coding there.
Dev, you might find the netnography developed by Kozinets (2002) helpful. The full citation is as follows: Kozinets, R. V. (2002). The field behind the screen: Using netnography for marketing research in online communities. Journal of Marketing Research, 39(1), 61-72. This methodology has been widely used in consumer culture studies.
Ethnographic research includes three components that are often quite independent: method, theory, and resources. Ethnography is never pure method, even without a stated theory there needs to be some assumed theory to make the data understandable.
In deciding what is the best combination for your doctoral research you have to consider what you are researching, what the established approaches are in your discipline, and what your committee expects in terms of each of the three components.
Thematic analysis or content analysis might be used. Look for this article:
Content analysis and thematic analysis: Implications for conducting a qualitative descriptive study by Vaismoradi,M.; Turunen,H.; Bondas,T. (2013)
From My little experience on the field, you should not go to the field with rigid methodology as it may happen that you will have to adopt to situation as the geographers will term environmental determinism. read around most possible theories but be open minded especially if you will be researching in African or Asia.
your data analysis should follow the lead of your methodology - having established the tools that were used your ethnography, you can thematically organize your analysis in that order. For instance, if you have used non-participatory observation to observe the spending behaviour of one particular community in their natural environment, your analytical themes can emerge from the tools: what, where, how and why?
There have been many complementary questions on ethnography and data analysis on other threads, I personally believe that sure, there must be a data analysis strategy! but it can be never be completely conceived before-hand. the thesis, the amount of data and importantly the research questions should lead the way to making decisions on the available tools for data analysis.
Wish you all the very best Dev!
Hi I am doing ethnography in one of my doctoral studies, and using narrative method for analysis. I think any interpretative approach will be helpful, such as content analysis for example. It depends on your research question I guess, and what you want to highlight of the data you gathered.
Particulary useful has been publications of Charol Mattlingly and Donald Polkinghorne.
There is a huge bunch of methods in ethnography and there is a huge bunch of theories in ethnography. I resign from summing those up here in a short answer. Which to choose depends fundamentally from your study object, the field and - finally - from your own subjectivity.
As regards using a software for qualitative data analysis: of course it helps you a lot with the management of a huge amount of heterogeneous data (documents, audio, video, pictures, etc.) and gives you also some tools which might deliver interesting insights for the interpretation. But keep in mind: they are primarily tools for the management of the data, they will not do the analysis for you.
Dev, you may need to develop in-depth reflections of how you have approached the field, the actors, and their stories in the first place. Such reflections will give you the clues of what kinds of theoretical assumptions you carry while doing the fieldwork -- assumptions about knowledge, reality and how to approach them. I agree with Deborah -- the choice of theoretical framework is linked to the methods of data analysis. But it is also important to be clear that the process of collecting and interacting with data has great impacts on the choice of analysis methods. So think about what you actually mean by "purely" ethnographic study, before dancing with any techniques....
You have to identify the "native" categories of the informants of your study.
@Aimilia V: Please elaborate. All of my informants are 'native' although I will not be 'going native'....it is in a prison setting, FYI
No, I don't mean 'native' in that sense but I mean that you have to identify the concepts that give meaning to their lives. For example, if we are talking about a prison then the term 'penal system' might not mean so much to them as the term that they are using in their slang to refer to that system. That does not only mean that they are using a different 'language' but also that their term has a different meaning in their micro-society which in turn creates a new mental construction of what they are involved in as subjects of the penal system.
The classical text for what Almilla is talking about is James Spradley's book The Ethnographic Interview. More recent approaches are covered well by Naomi Quinn in her edited volume Finding Culture in Talk. Here is the citation: Quinn, Naomi. 2005. Finding Culture in Talk: A Collection of Methods. Edited by Douglas Holland, Culture, Mind, and Society. New York: Palgrave Macmillan.
Quinn argues that in examining interview data it is necessary to consider the combination of key words, metaphors, and lines of reasoning. This is based on her three centuries of social science research on marriage in the U.S.A.
I've little to add to what others have already said. But I'm a bit surprised you haven't (apparently) approached social anthropologists at Cambridge, including some who have published on ethnographic methodology. I would have thought that they could answer your questions.
Creo que un buen recurso es el metodo cualitativo ,basado en el estudio de casos individuales que despues puedas comparar para establecer ciertos patrones o llegar a conclusiones ,complementado con estudios academicos que puedan sustentar algunas teorias.
Look at beautifully written article written by Tim Ingold in recent HAU: journal of ethnographic theory. It is titled 'that's enough about ethnography' . The article introduces the debate and discussion about the method. it shall give you an idea how to go further.
I used ethnography in my doctoral research. There is no rule of thumb vis-a-vis your question. The primary question you have to ask yourself is: I am using ethnography as a methodological approach (which contains a proponent of analysis) or ethnography as a method? I take issue with the latter as students and scholars alike confuse ethnography as a method with participant observation. Nevertheless, this is a question you need to ask.
Also, I advise to not se grounded theory as a analytical framework of framework. This alludes that ethnography is purely inductive, which it isn't as the researcher will bring with them aprioi knowledge to the field. Ergo it is imperative to see ethnography as pulling the inductive with the deductive wags of doing research: that is, your empirical data will inform your theoretical knowledge which, in turn has some say over your data. For this I recommend Thematic Analysis. This way you are exploring theme in the fieldwork that derive from both your reading of the literature and data, as opposed to Grounded Theory where themes are solely drawn from the data.
I suggest that you have a look at one of my articles: Triangulating Research Methods In Case Studies: An Empirical Representation. The article is downloadable from: https://www.researchgate.net/publication/283709585_TRIANGULATING_RESEARCH_METHODS_IN_CASE_STUDIES_AN_EMPIRICAL_REPRESENTATION ; and http://www.hrodc.com/Research_Papers.htm
I have a second relevant article, which I will make available to you shortly. Good Luck – kind regards. Prof. Dr. Crawford
Conference Paper TRIANGULATING RESEARCH METHODS IN CASE STUDIES: AN EMPIRICAL...
An ethnographic study by its nature captures volumes of 'evidence' that Quantitative studies cannot, because of the need for extended observation at its core; so to start, your research must take a Qualitative approach. Second, because of the volume of data you will no doubt capture, you may wish to employ 'computer programmes' to assist with organising data collection and making sense of general emergent trends from the data. Finally, however, this does not and should not mean that the same could not be done manually, using a well-designed set of spreadsheets and simple graphical analyses (e.g. to assist with frequency counts). From my brief experience of CAQDAS, they have more to say on the 'tin' than they can deliver when you are in need!
Estoy de acuerdo con el profesor Aasland,la etnografía ,como tal ,en una investigacion como una tesis debe sustentarse en una teoria,al menos,para que los datos observados puedan tener una cierta validez.
Otra cuestiosn es si la investigación es mas cualitativa o cuantitativa ,empleando programas informaticos que nos ayudan a mejorar nuestra investigación.
Thank you all for your advice, it has been helpful to me too. I have just returned from Zimbabwe to collect data for my PhD on the experiences of people living with schizophrenia in Harare. And am now doing the analysis. Thank you for your insights.
If anyone would like to get in touch to offer more advice or collaborations fell free to email me here [email protected]
Regards
Shep
http://www.utas.edu.au/profiles/staff/health/shephard-chidarikire
I dont have much to say except that I am currently analyzing family geographies collected as part of our study 'role of marginalized in health care'. I am using MAXQDA for analysis (a very user friendly software) and had started to wonder that the open code approach that I use is OK, and started going through material on analysis of ethnography data.. Skimming through the material I came across, and this exchange also, I gather that every analyst needs to write how he/she did the analysis, and share with others in the same domain. there is really no one formula...the core thrust, however,, is, I believe, to get as close as possible to how life is lived by those one interacts to understand their life.
I found it helpful to think about ethnography as telling a story, which I suppose links to the idea that ethnography 'gives voice to'. I did observational fieldwork with some interviews at two sites - so I characterised these as two case studies, and then I told the story at both sites. I thought about using NVivo as a way of organising my data, but this would not have worked for my hand written fieldnotes (unless I had typed them up, or produced them digitally in the first place). Also, I have quite a lot of experience of using NVivo and I couldn't quite envisage using it for ethnography which seems a deeply immersive experience. I think NVivo is great for some qualitative research but I think for my project I may have ended up forcing themes and categories on my data when really it seemed to be more important to let the story emerge.
I am now researching on the experiences and stories of settlements of people who migrated from one place to another in the 1950s. Therefore to get their stories I have interviewed a number of them. Stories unfolded and based on the data a common theme emerges, without the help of any software.
The question is: Can I just write the stories as they are or do I need to verify them with any analysis software? and as Suzanne Molesworth said, I might just 'ended up with forcing themes and categories' onto my story. Will this method be considered as a credible analysis?
I found that in both Spradley's books he describes how to go about writing the ethnographic stories. I found this FORMAT useful, especially as I found no step by step analysis and felt I needed it.
Because their is no specific, prescriptive step by step, way to analyse data in ethnographic studies, this makes ethnographic analysis difficult. This explains one reason why member checks are so important once the stories are written and during writing the stories. Jane Kerr
I think it is often the case that collecting the data reveals the key themes without the need for elaborate, subsequent analysis -- especially if you do memoing throughout your data collection.
Even so, I would take some steps to confirm your preliminary conclusions. In particular, I would re-read each interview and summarize it in terms of your proposed theme(s). That will help you ams sure that that theme(s) are consistently present, and potentially alert you to some things you may have overlooked.
Thank you so much, Jane and David for your replies. I will definitely look up Spradley's books and take the steps you both recommended,
Computer assisted analysis is not a way of verifying or documenting the existence of themes you find, you don't have to do that. If you want to verify themes, and if they are key you probably should, then take your tentative themes to your interviewees or informants, and ask them how they react to what you think you have found. they may disagree, or add or modify what you have found, deepening your analysis, as well as giving you some confirmation of the themes you have found.
The computer programs can be a way of suggesting themes to you, that you then have to examine and say if they are real or useful, and they can help you find instances throughout your data if you have a lot of it. But it is not better than you are and certainly not better than your interviewees/informants at knowing what is important or useful.
Irene
I used thematic analysis - see Jennifer Mason and Braun and Clarke (2006). I preferred to use diagrams on flipchart paper and code using spreadsheets.
Hi Dev,
Here's a link that may be useful for you:
https://books.google.com/books?id=bOLFDQAAQBAJ&printsec=frontcover&dq=cresswell+five+approaches&hl=en&sa=X&ved=0ahUKEwi73tXqou_cAhUG0VMKHUb5ABYQuwUINTAB#v=onepage&q&f=false
Have a great day!
--Adrian
I think we are requiring all data whether using a longitudinal or ethnographic or immersive experience, or indeed broadly quant or qual, to be able to be demonstrated as meeting soundness as an overriding principle. Guba and Lincoln suggest the criteria credibility, transferability, dependability and confirmability. Some form of triangulation such as from multiple informants or from different data sources, such as interviews and observations and field notes, can provide multiple viewpoints to enhance credibility. Member checking - or a technique of sharing the findings or insights with the participants - seems quite a persuasive soundness technique as the researcher is most often trying to understand and show to others the participant view of the world. Video ethnography techniques may be used along with traditional field diaries and notes. If using a thematic analysis then development of matrices to help you understand and build codes to sub-themes is also helpful. It isn't necessary to use any machine assisted / software program as part of the data analysis but one advantage is that there is less cognitive demand on you / may speed up analysis / may provide some external "validity" as that is perceived as arm's length / not as potentially biased.
Mi respuesta en un libro: http://www.institutomito.org/assets/la-infancia-masculina-en-itagui-(1950-1960).pdf
Depending upon the research questions a combination of interviews with unstructured thematically guided observation all of this reduced to the Narrative analysis or discourse analysis if critical theory is the basis of observing a distinct cultural structure . Themes , Codes and categories can be generated through manual methods ....
- Badges
- Science topic