Truth, Lies and Videotape
Imagine asking a group of eight year olds to present the results of a study on children's eating habits. Suppose you wrote in a report how annoying your teenage respondents were, because they wouldn't stop mucking about. Think about asking a team of single mothers to conduct an important social research project on attitudes to childcare. Or maybe show a new consumer typology to the respondents it was based on, and ditch it if they say they don't recognise themselves.
Commercial suicide for a researcher looking forward to early retirement, or a challenging, energising approach to inspire researchers and clients alike? These ideas come from recent thinking and controversy in academic ethnography and anthropology and I will argue that ongoing debates within academia are directly relevant to commercial market researchers in at least two ways. First, they address core problems of truth, objectivity and cultural understanding which are central to qualitative market research. Second, they also provide us with a source of new and more inspiring ways to help us conduct, analyse and present our research to clients.
Back to the future
Ethnography is not a new method, although it may be novel to some commercial researchers. For a good general introduction see Hy Mariampolski’s paper, ‘The Power of Ethnography’ (1999). Its beginnings are usually traced to the work of the social anthropologists Bronislaw Malinowski (1922) and Franz Boas (1911; 1927) in the early 20th Century.
Malinowksi conducted fieldwork among the Trobriand Islanders of the Western Pacific and his major contribution was to establish participant observation as key to ethnographic research. He also emphasised the importance of the relationships between different elements of a social system.
Boas, on the other hand, stressed the meticulous collection of data while emphasising the differences and particularities of cultural groups. By the 1920s the key aspects of ethnographic method were well established. These included long-term immersion in an alien culture; learning the language of the people; participating in as well as observing their social lives; and analysing all aspects of culture, to produce a complete account of the group being studied. (Clearly, commercial ethnography is different, with researchers spending less time with participants and having a more focused research topic... but let’s leave that to one side for now.)
Since the 1980s, however, these ethnographic conventions have been increasingly questioned from within anthropology and from cultural studies and literary analysis. Key criticisms include these:
- Ethnography has a colonial heritage and, historically, was
used by the powerful to exploit the weak
- The knowledge produced was biased because the
researchers adopted a colonial perspective
- Male ethnographers ignored the role of women, producing
patriarchal accounts of the societies researched
- Research subjects were construed as passive and were unable
to influence how they were portrayed
- Ethnographic reports were written as objective, scientific accounts, although this is not what they truly were
For some academics, particularly those within linguistics or cultural studies (Barthes 1977; Foucault 1979), these problems are inherent to the method and cannot be resolved. Other ethnographers, however, have developed a range of approaches to address these problems, approaches that can be adapted usefully for commercial researchers.
Who's in charge here?
The relationship between researcher and subject has often been portrayed as exploitative (e.g., Stacey 1988; Lather 1986; Oakley 1984; Finch 1984). Researchers don’t inform participants about the research; they don’t reveal personal information; they obfuscate and sometimes lie if pressed on the purpose of the study. Yet they expect participants to be open, trusting and honest with them.
Feminist researchers such as Stacey and Lather have, however, criticised this inequality of power, developing an approach which is termed ‘participative enquiry’ or even ‘emancipatory research’. This suggests that researchers will gain better results, and operate more ethically, if they are completely open and honest with participants. This means telling them the real research agenda, allowing them to influence the questions asked, and asking for their co-operation in seeking a solution jointly. It means treating them as intelligent equals, not as passive ‘respondents’.
If you read Oakley, Finch or Lather on the subject, you may feel that their portrayal of research as sisterly, feminist emancipation is far removed from commercial realities. There are, however, insights in their work which we could use in commercial ethnography. Why not allow participants to set the agenda and give them a full explanation of the research’s goals? What about involving them in research design, so they can suggest the most appropriate methods, where to conduct the observations, and who to include? Not received wisdom for most of us, but we stand to gain stronger engagement and commitment, as well as some good ideas, from those closest to the problem.
Who should conduct ethnography?
Ethnographic research was long deemed best conducted by someone from outside the culture being researched. Outsiders don’t make assumptions about what is normal and can enquire into the most obvious aspects of social life.
So, much early ethnographic research was conducted in this fashion, often by white, middle class ethnographers, researching non-white, poor subjects. From the 1970s onwards, however, the assumption that outsiders make the best ethnographers has been strongly criticised. Particularly in the post-colonial context, the idea of white men going to study primitive people is increasingly questioned.
Ethnographies conducted in the 1940s and 1950s are criticised as biased and partial accounts, driven by the needs of the colonial administration. These criticisms were strongly influenced by Edward Said’s brilliant book, Orientalism: western conceptions of the orient (1978). Said’s book is fascinating and accessible, and he makes a compelling argument that western scholars created an image of a static, primitive, exotic ‘orient’ as a justification for colonial domination.
Through such thinking, a strong argument was made in favour of ‘native ethnographers’ or ‘insider accounts’, research conducted by those within the culture. Insiders, it is argued, are better positioned to understand a culture’s subtleties. They are, in addition, better placed to develop trust and rapport, and are less likely to exploit the participants. Some post-colonial theory is dense and is not written in the most accessible style. (Spivak 1987; 1990). But practical anthropological work, such as Renato Rosaldo’s book, Culture and Truth: the re-making of social analysis (1989), can be inspiring.
Rosaldo suggests that traditional ethnography served the interests of the colonial states and oppressed the people it studied. He argues convincingly, however, that the method can recover from its tainted past if ethnographers acknowledge its limitations and realise that what they create is an interpretation of reality, not an objective reflection of it.
How do these insider/outsider debates affect us? In the commercial world we are accustomed to thinking that we can interview anyone, or observe any social group. We don’t believe only women can interview women, only black people can interview black people, or teenagers can interview teenagers. We could, however, think about training members of a community being researched to conduct their own observations or interviews.
Most of us will have noticed that when respondents in groups talk to each other, the quality of their conversation is different from when they talk to the moderator - more natural, more relaxed, more genuinely interested. So might this be even more relevant to ethnographic observations? Perhaps the best ethnographer for a project researching eight year olds is an eight-year-old child. Maybe a project researching drug prevention is best conducted by an ex-heroin addict, rather than a middle class researcher with no experience of addiction.
Obviously, you can’t hand over the whole project to schoolchildren or drugs workers, but we could certainly make greater use of the insights they might gain and which we could not access. We might consider a more collaborative approach to ethnography, working more closely with individuals genuinely immersed in the culture of interest.
What does it all mean?
The question of who should conduct ethnographic research leads us to the vexed issue of how to analyse and explain other cultures, whether we are discussing the Trobriand Islanders or the shoppers at our local supermarket.
There is a school of thought within anthropology, deriving from linguistics, that cultures can only be analysed and explained in their own terms. This is sometimes called 'cultural relativism', the belief that cultural systems make sense internally and that trying to analyse or criticise them in other terms will always distort and create a biased image. This is the logical conclusion of the argument of the linguist Peter Winch (1970), who suggests that we are all constrained by our own cultures and languages, condemned never to understand each other.
Anthropologists, however, have tried to find practical solutions to this dilemma. Like commercial ethnographers, they first try to understand a culture in its own terms, but then need to place this understanding in a wider context, perhaps using analytical categories from outside that culture. The question is how to do this without distorting the original cultural phenomena under analysis.
Various anthropologists have engaged with this question, perhaps most usefully Clifford Geertz (1973; 1983) and George Marcus (1994). Geertz suggests that anthropologists can pay more useful attention to ‘the quality of the space between us’, i.e., the differences between the perspectives of our research participants and our own or our clients’ analytical categories. George Marcus even suggests a solution might lie in ‘redesigning the observer’, i.e., scrutinising our own analytical concepts with as much rigour as those of our participants.
In commercial practical terms, these ideas perhaps describe a process for bridging the gap between how clients understand a market or category, and how it’s understood by consumers. We don’t need to privilege one account over the other; rather, both are analysed and some third position developed, taking account of client and consumer perceptions.
Beyond this, how many times do we feel, when we have analysed our data and presented it to our clients, that we have lost touch with the reality of consumer experience? Maybe we should present our interpretations back to research participants as well as to clients? Perhaps we should involve consumers in our analysis and interpretation, to see if they recognise the picture we have painted of them - if they can’t see themselves in the mirror, we must have lost something important along the way!
New Style Reporting
- Include yourself explicitly in the narrative
- Describe your feelings during the research
- Allow different viewpoints in the text
- Show how you drew your conclusions
- Explain which interpretations you rejected and the reasons why
Fact and fiction at the debrief
Finally, the area of ethnographic writing has been subjected to intense criticism over the last 20 years, in what has been called ‘the literary turn’ in anthropology (Back 1998). In classic ethnographic monographs, research findings were written up in an objective, quasi-scientific fashion. The focus was on the structure of society and culture, and they were often in the present tense, suggesting an eternal truth rather than mere snapshot.
These reports followed a linear logic and had a single authorial voice, the omniscient researcher guiding the reader through the alien culture. There was little reference to the researcher’s personal experience or indeed to the actual process of doing the research, and methodology chapters often conveyed a feeling of orderly planning rather than chaotic reality. This picture will probably be recognised by most commercial researchers, as it is pretty much the model that is always used for commercial research reports. We don’t sit down and tell our clients what really went on during the research - they don’t want to know, we tell ourselves.
This style of ethnographic writing has been subjected to illuminating analysis from the field of literary criticism. Analysing ethnographic reports as if they were works of fiction, critics have demonstrated that the conventions of this literary genre are just as strong as those of the romantic novel or the pastoral poem.
The most important text in this field is Writing Culture: the poetics and politics of ethnography (1986), edited by James Clifford and George Marcus. By applying the techniques of literary analysis to ethnographic writing, Clifford suggests that at best ethnographers offer ‘partial truths’ or ‘true fictions’. He means that the conventions of ethnographic writing allow certain truths to emerge but hide others, and that these processes are not fully within the writer’s control - the genre applies its own constraints.
There is, for example, little space within conventional ethnography to mention the boredom and loneliness that characterise much overseas fieldwork. Neither is it the done thing to admit to disliking one’s participants. But might these not have some bearing on the research findings? Why are we not allowed to mention them?
Since Clifford and Marcus’ book, many anthropologists have experimented with different ways of writing up and presenting research findings. Some have included ‘multiple voices’ within their texts, allowing conflicting points of view expressed by participants to stand side by side, forcing the reader to confront the complex reality. Others have tried to develop an ‘experiential’ style of writing, attempting to convey the complete and complex nature of cultural life without breaking it down into categories such as ‘kinship’, ‘ritual’, or ‘religion’.
Ethnographers have also attempted to tackle the criticism head on and have attempted consciously to place their personal experience, attitudes and opinions at the heart of the research process.
They refer to their own views on the subject, admit their prejudices and potential biases, and outline the ways in which they have attempted to deal with these. This is known as ‘reflexivity’, where the researcher him or herself is as much a topic of analysis as the research participants.
Les Back, for example, is an ethnographer who has worked on racism among young men in South East London. Writing as a white man from this background himself, in his work (1993) he includes discussions about his own feelings on researching this controversial issue, and the emotional difficulties it sometimes caused him.
Another approach is to include an explicit account of how the analytical ideas were developed, which ones were rejected and why, and how researchers arrived at their final conclusions rather than presenting them as a fait accompli.
I can already hear you thinking that your clients don’t want a confusing, contradictory presentation with a report emphasising how tired you were at the end of the evening. But there are other ways these ideas could be used. We could explain how the research was conducted in a more realistic fashion, rather than providing the usual cursory sample structure in the appendix. We could certainly include a wider range of points of view in our reporting, highlighting those that we ended up prioritising and saying why. Maybe our clients might even respect us more if they formed a credible picture of how, in reality, we arrive at our conclusions.
With clients regularly complaining about how boring research reports and debriefs can be, this alternative approach might be more inspiring, memorable, compelling, challenging... and even more truthful.
What do we need to know?
Ethnography is a powerful research tool that has survived the last 100 years and is unlikely to disappear. Indeed, it has emerged revitalised from the academic debates of the last 20 years and has an enhanced credibility, legitimacy and integrity as a result.
Ethnography in the commercial world can, we know, help us gain insights and inspirations that are not accessible from interview-based research. Recent advances within academic thinking, meanwhile, show how we can continue to develop this approach, in terms of how we relate to participants, how we analyse and interpret the data, and how we present it to our clients.
These debates do concern us. Their relevance extends beyond ivory tower academics and literary theorists. If as commercial researchers we want to make claims about the truthfulness and integrity of our work, in the face of competition from all around, we need answers to these questions, too.
Back, L (1993a) ‘Gendered Participation: masculinity and fieldwork in a south London adolescent community’. In D Bell, P Caplan, & W J Karim (Eds.), Gendered Fields: Women, Men and Ethnography (pp. 215-232). London: Routledge
Barthes, R (1977) Elements of Semiology. New York: Hill and Wang
Boas, Franz (1911) The Mind of Primitive Man, New York. Macmillan
Boas, Franz (1927) Primitive Art, New York. Macmillan
Clifford, James and Marcus, George (Eds) (1986), Writing cultures : the poetics and politics of ethnography, Los Angeles:University of California Press
Finch, J (1984), ‘"It’s Great to Have Someone to Talk to": The ethics and Politics of Interviewing Women’, in Bell, C and Roberts, H (Eds), Social Researching: Politics, Problems, Practice, London: Routledge and Kegan Billy
Foucault, M (1979) The History of Sexuality: Volume 1. Harmondsworth: Penguin
Geertz, Clifford (1973), The Interpretation of Culture, New York: Basic Book.
Geertz, Clifford (1983), Local Knowledge: Essays in Interpretative Anthropology, New York: Basic Books
Lather, P (1986) ‘Research as Praxis’, Harvard Educational Review, 56:257-275
Malinowski, Bronislaw (1922) Argonauts of the Western Pacific, London: Routledge
Marcus, George (1994) ‘After the Critique pf Ethnography: Faith, Hope and Charity, but the Greatest of These is Charity’, in Borofsky, R, (Ed), Assessing Cultural Anthropology, New York: McGraw-Hill, pp. 40-54
Mariampolski, Hy (1999) ‘The Power of Ethnography’: Journal of the Market Research Society, Volume 41(1), pp.75-86
Oakley, A (1981) ‘Interviewing Women: a contradiction in terms’. In H Roberts, (Ed), Doing Feminist Research (pp. 30-61). London: Routledge
Rosaldo, Renato (1989) Culture and truth: the re-making of social analysis, Boston: Beacon Press
Said, Edward (1978) Orientalism: Western Conceptions of the Orient, London: Penguin
Spivak, G C (1987) In Other Worlds: Essays in Cultural Politics, London: Routledge
Spivak, G C (1990) The Post-Colonial Critic: Interviews, Strategies, Dialogues, New York: Routledge
Stacey, J (1988) ‘Can there be a feminist ethnography?’, in Women’s Studies International Forum, 11:21-27
Winch, Peter (1970) ‘Understanding a primitive society’, in B R Wilson (Ed), Rationally, Oxford: Blackwell
Director, Turnstone Research
This article was first published in InDepth magazine, September 2004
Copyright © Association for Qualitative Research, 2004