Iain Murray is quite right to suggest that we should “never trust research that relies on what people say they want or like – it’s what they do when they have a choice that counts” (MW September 13). In fact, the whole issue reminded me of the problems faced by the early researchers in linguistics.
They discovered that they had a problem when studying speech as, when it came to the essential recording which formed the basis of their data, people automatically spoke differently once the recorder was on. Labov and a few others devised a clever approach whereby they would spend some time “interviewing” people, and then pretend to turn the recorder off and proceed to engage the subjects in ordinary conversation which was secretly taped. This was the real data they sought and used.
The problem with researching behaviour is that, in too many instances, people do not know what they do or why they do it. Or they will tell you that they do something that they patently don’t for a whole host of strange and obscure reasons.
Consequently, although it may be interesting to know what people think they do, or want you to think they do, it is always preferable to find out what they really do and work on that.
Aspen Business Communications