Sampling in Design Research: Toward Ethnographic Segments
What people say and what they end up doing is different. It’s not just important to speak to them but you got to spend a day in the life of the customer and observe them.” For understanding the customer better, Jain suggests new methods such as ethnographic studies and tools like calculating customer lifecyclevalue. — Dipayan Baishya, THE ECONOMIC TIMES (India), July 2005.
Anthropological research teams are more and more often called upon to produce research results for design teams. Usually, the team has to devise some sort of formal research plan for the client. And often, the client is working from a segmentation scheme derived from traditional, questionnaire-based quantitative research. These schemes are usually based on questions that make sense--strategic sense--to the client. Sometimes they reflect the ways that consumers or end-users actually organize themselves. But more often than not, they miss important kinds of variation.
We were working, recently, for a client like this. Actually, we were almost working for them. (They did not accept our bid.) They had a segmentation scheme that they used for their global mobile-phone marketing and design. They wanted to know more about the Indian market. And they wanted us to build our proposal around their segmentation scheme. While we not win the contract, we used the bidding opportunity to think a bit about why we were not happy with traditional segmentation schemes.
Traditional segmentation schemes are means-based, and like other means-based approaches (see Maltz 1994), they don't do a good job of providing the contextual data that help designers imagine design scenarios. And because they are based on what people say, they may be based on lies that people are telling, as Professor Jain suggests in the quotation above.
A psychologists might draw a sample of individual product users or buyers, and study patterns in individual desires, attitudes, values or behaviors about a product. An anthropologist would discover the range of contexts in which groups of people learn about, acquire, transport, store, exchange, use, and talk about a product. Its a significant different.
And more than that: an anthropologist would want to participate as much as possible in product use, and interview the people using it. That approach calls for a rather different kind of sampling. It calls for theoretical sampling.
Works Cited Part I
1994 Deviating from the Mean: The Declining Significance of Significance. Journal of Research in Crime and Delinquency 31(4):434-463.