I wish I could recall the title and author of a paper that discussed the current tendency for sociology to disfavor the observation of real people in real encounters. Is that claim familiar to anyone? (how many qual studies are published by ASR compared to quant studies? )
David Silverman has made just this point recently. Silverman argues that qualitative researchers in particular have always favoured asking people about behaviour rather than observing actual behaviour. You may wish to look at his ‘Very Short, Fairly Interesting and Reasonably Cheap book about Qualitative Research’ for more on this.