ABSTRACT
Conventional wisdom in sociology dictates one sequence above all others: first theory, then data to test and/or elaborate that theory. In the task of survey design we were only too aware of this axiom. It was overshadowed, however, by the absence of any hypotheses, let alone a set of interrelated hypotheses amounting to a master theory, of overriding significance on gambling. 1 What did exist were a number of interesting theoretical leads and propositions which called out for further testing and elaboration, but none of which – taken alone – were of sufficient weight to form a pivot for the inquiry. The nearest to a full theoretical statement is that of Devereux, 2 but his structural-functional frame of reference predicted, for example, the persistence of gambling in a sub rosa status of illegality. The legalization of virtually all forms of gambling in this country in 1960 seemed prima facie evidence for rejecting his theory, whilst recognizing its suggestive powers and singling out from its theoretical corpus those propositions which seemed of continuing relevance. The situation of gambling theory, then, was one of the existence of interesting pointers rather than of any major theoretical statement around which to organize a full-blown study of gambling. We therefore determined to test these propositions as far as possible in a survey framework designed to reach other goals as well: namely, the gathering of primary information about a variety of forms of gambling, and the exploration of the relevance of differing commitments in leisure, socio-economic and attitudinal factors to gambling. In this way, we hoped to maximize the yield of the surveys: for to test one theory which falls flat is to waste the opportunity to test others and gather information which may enable us to generate yet more. 3 The former is worth-while only where a formidable theory has evolved which promises a greater explanatory yield than the sum of other theories and propositions available at the time which relates to the same phenomenon.
