In an attempt to elucidate some possible conditions for success in managerial decision making, data were analysed from 53 cases of decisions in eight British organizations, five business firms and three non-business organizations (two universities and a District of the National Health Service). No clear relationships between features of the processes of making the decisions, and their successfulness were found until the business firms and the non-business organizations were separated. Clear differences then showed up, relatively speaking, in the conditions conducive to success. In the business firms, a successful decision was more likely to result from a decision-making process in which resources were available. In other words, in business a successful decision is most likely when sufficient information and sufficient means of implementation are to hand. By contrast, in the universities and the Health District, a successful decision was associated more with the social qualities of the decision-making process itself. In other words, in non-business organizations a successful decision is most likely when the right people participate and the people at the very top do not interfere too much.
ORGANIZATION DIFFERENCES?Are the conditions conducive to a successful decision in one kind of organization the same as those appropriate in another kind, or are they different? Probably they differ, for no one argues that managerial decisions are made the same way and for the same reasons in every organization. It is taken for granted that what happens in different kinds of organization must differ, given differing conditions. Though Hickson et al. (1986a) firmly conclude from their extensive data that 'the matter for decision matters most', meaning that processes of decision making are patterned primarily by what is being decided, they emphasize that there are also considerable differences according to type of organization. This is because for the making of decisions an organization is the ruling framework governing how a Address for re@&. r: