Picture of the week: Is fieldwork ecologically valid?
Many anthropologists are uncomfortable with the idea of performing experiments on the people - friends - they meet during fieldwork. This is partly for ethical reasons (which, one might add, are rarely seriously reflected on). But it is also because they doubt the ecologically validity of tasks which artificially disrupt the flow of ordinary life. This is a serious concern; but of course the same concern can be raised about anthropological fieldwork itself, which represents a dramatic intervention in the lives of our informants.
[PICTURE]
First picture: People in rural Southeastern Taiwan...
Second picture: ...with the anthropologist in their midst.
Rita Astuti 6 November 2008 (00:00)
Issues of ecological validity are clearly very important and anthropologists are right to be concerned, even if ”disrupting the flow of life” may not be the most significant issue to worry about. By contrast, I’m not sure that I understand what general ethical objections anthropologists might have regarding the use of experimental methods - unless of course a specifi experiment happens to be unethical.
Paulo Sousa 8 November 2008 (18:36)
Leaving aside the ethical issue, there seems to be two relevant issues suggested by the phrase ”disrupt the flow of ordinary life”:
(1) External validity (in particular, ecological validity): whether the results of a specific research methodology can be generalized beyond the exact research context. In relation to cognitive anthropologists doing experimental or quasi-experimental research in the field, the problem is whether the results coming from the research correspond to the cognition deployed in ordinary life. In fact, this is also a problem for anthropologists using a more qualitative type of methodology: do their results correspond to the cognition deployed in ordinary life?
(2) Internal validity: whether the results of a specific research methodology are affected by confounding variables. In relation to cognitive anthropologists doing experimental or quasi-experimental research in the field, confounding variables such as reactivity and absence of task familiarity should be addressed.