These are some thoughts about anecdote as evidence, prompted by friend Carol's persistence. I suppose really this is about narratives in research. I would need to give further thought to how they differ from anecdotes.
Research is about gaining knowledge. Knowledge could be gained by discovering something that has never been found before. That something was there, it just needed finding, or 'proving' to be a fact.
But, really, facts are not that easy to establish, particularly in the study of humans. Even a nice neat survey (86% of people said they preferred to buy corned beef in square tins) becomes a nonsense, as soon as we look at the motives of the researcher, the flawed questionning techniques of the data-collector, and what the respondent actually wanted to say, but couldn't, because they weren't offered the right options.
If I ask a person to tell me their view, or give an account of an event, then I must accept that there is no such thing as a single truth that I can ascertain. As soon as I accept that, then I can allow myself to hear what they say in a different way.
Why are they telling me this story?
Why are they telling it in this way?
Would the other people in the account have told it differently?
If the person is telling me the story like this, then what does that tell me about what it meant to them?
For this person, where did the story begin and end?
If I ask them to tell me about the same event next week, will it be the same?
In this way I can find out more about the individual who tells the story, but it does not preclude me from finding out, also, about the event that happened, as long as I remain aware of the layers of representation of that event that are being introduced. (Catherine Riessman describes these well). The event is filtered through the perceiver's eyes (and history, and experience and motives) and then, as I hear about it, through my own. When I tell, or write about the individual's account of the event, the, again, I add a distorting layer of representation.
None of this makes the original account any the less interesting or useful. But I must stay aware.
I'm listening to, and trying to understand, health-service users' experiences of health care at the moment. What a wealth of information! And none of it is quantifiable (at least not if we want to keep the integrity of the tale that is told). To reduce any part of the tale to numbers would be to over-simplify and negate the impact of complex events, in complex contexts.
Thanks for doing this! I have been pondering a lot on these sorts of issues recently. I recently did an MSc in Epidemiology at the London School of Hygeiene and Tropical Medicine. It was a bit heavy on the mumbers side of things, whilst anecdote and narrative were almost taboo words. I love all the number crunching stuff, but I feel betrayed by it. I'm not going to continuously rant on about this on here - as you'd lose the will to live fairly quickly - but I think I need to give my thinking some context. So...my main role in recent years has been researching the causes of autism, and more specifically, why its changed from a prevalence of 4 in 10,000 20 years ago to 1 in 64 today. The big number crunching studies tell us there isn't a change in causality - they can't find a 'smoking gun'. They conclude, therefore that there IS no rise. The logic presumably being that if you can't find something, it doesn't exist. So, they say, we just didn't notice all these people before, or we didn't call them autistic, or we didn't count them. Or something. The arguments often sound quite frantic. The children were born with it, they've always been around, its genetic, nothing has changed, honest guv.
ReplyDeleteThe parents of many of the affected children, on the other hand, tell us that all was well until they took their child for routine vaccination, and then nothing was ever the same again.
As a scientist, I was with the numbers guys. Big, big studies have looked at the whole causality thing, and nothing pops out as a culprit (more of why that might be in later posts perhaps) so at the population level, no smoking gun. No smoking gun, no cause. No cause, then the thing can't be real. I'm sad to say it all made perfect sense to me.
And then I met the parents. Heard their narrative.
Something isn't right here, and a great deal is explained by the profit motive of pharmaceutical industry together with the genuine concerns about public health. But not everything in this medical catastrophe is explained by this, and its given me food for long and hard thought. About how we come to knowledge, what we actually mean by 'truth'
I guess there's the obvious point about the difference in reality at a population level, compared to the reality of an individual within it. Science is much better at helping us find out about the former, and next to useless at helping us learn more about the latter.
I think one thing this experience has taught me is that there are different truths, and different paths to each of them.
In the past I would have proudly declared myself a scientist and felt that in doing so I somehow added credibility and validity to any claims I might make. I know longer beleve that. I now see science as a process, a method towards one sort of truth. I'm not sure any more that there is an absolute truth. Science is in many ways a giant house of cards built on probability theory, and it can sometimes lead us far away from the real world.
I've got lots more thoughts to share - I'll try not to ramble. Must go and make a cup of tea ready to watch Poirot pursuing his own version of getting at the truth. Love it! xx
Nothing like a good Agatha Christie!
ReplyDeleteIt is, in my view (and that of others), a mistake to think that the word 'science' belongs to positivist number-crunchers. Science is about being systematic and thorough in one's investigation of phenomena. Well, you don't have to crunch numbers to do that. In qualitative, interpretive research, the research data can be analysed systematically and very thoroughly, with due regard being given to the range of factors that might interfere with, or distort the process and the findings. In this type of research, I can aim to enhance validity, although reliability is likely to be less controlled, because people's accounts do change according to context.
Oh - I sound like I'm in lecturing mode - time for another glass of wine.
Carol, it is so nice to get to know you a second time!
ReplyDeleteyes, you too. Agree with the above 'n awl. I'm s'posed to be writing a chapter in a book called Secret Ties which is about industry's influence on the scientific process but I'm so muddled about what science really is I can't get started.....
ReplyDeleteoh blimey it still thinks I'm polly
ReplyDelete