the sociologist today limits himself rather generally to the construction and conduct of questionnaires and interviews. In 1940–41, 25 per cent of the 85 empirical studies depended on interviews and questionnaires for their data; in 1965–66, 48 per cent did. However, if we consider studies based on secondary data that, in turn, derived from interviews, then 64 per cent of the 136 research papers in the latter two years were based on such verbal reports.

Increasingly, these verbal reports are limited to expressions of attitudes, feelings, and opinions rather than to factual accounts of past behaviour and interactions. In 1940–41, 8 of the 22 studies using questionnaires and interviews obtained statements about opinions or feelings, 6 focused on actual behaviour and 8 gathered information about both past behaviour and attitudes. In 1965–66, 49 of 66 studies in which interviews were collected dealt only with opinions and sentiment, 6 with behaviour and 8 with both behaviour and attitudes.