On the Measurement of the Environment: An Assessment of the Lawrence and Lorsch Environmental Uncertainty Subscale

Abstract: The contingency approach is currently in vogue in organization theory. If certain important environmental parameters can be specified, the best form of organization can be determined. This approach is reminiscent of some of the criticism leveled at Taylor and Fayol, that is, they believed that there is one best way. Modern theorists associated with the contingency school now argue that there is one best way, but it all depends. Thompson noted that recent work searches for patterned variations, for the optimal choice of organization structure and administrative practices in the face of relevant situational factors (Thompson, 1967). Burns and Stalker (1961), Woodward (1965), and Lawrence and Lorsch (1969) all adopted a contingency approach. Their arguments, possessing considerable validity on the surface and generally presented in beguilingly simple form, centered on the existence of an optimal fit between organization structure and some contingency factor such as technology, situational favorability, environmental uncertainty, and so on. Lawrence and Lorsch (1969) argued, for instance, that organization success requires the maintenance of differentiation and integration consistent with the demands of the environment. Firms facing a stable, certain environment may find bureaucratic organization structure adequate for achievement of proper integration and differentiation, while a more uncertain environment would preclude effective use of the same form. Other researchers focused on different dimensions of this fit. Pugh (1963), for instance, considered interrelationships between entire sets of variables-contextual, structural, group, and individual. Morse (1970) suggested that proper fit may enhance not only productivity, but also worker "sense of competence motivation." Fiedler (1967) drew certain conclusions about appropriate and effective leadership styles based on the interaction of three situational variables. Central to this research is the adequate specification of moderator variable characteristics. The validity of prescriptive statements which all of these theorists set forth is especially sensitive to errors in measurement of the contingency variable, for example, situational favorability, environmental uncertainty, technology, and so on. It is therefore both significant and unsettling that recent research findings cast doubt on either the accuracy of measurement of certain of these measures, their underlying meaning, or their sufficiency in isolating relevant environmental factors. Harvey (1968) has argued, for instance, against Woodward's measures of technology in favor of technical diffuseness or specificity, taking account of both the form of technology and the amount of change within a given form. Hunt (1970) concluded that a potential basis for the similarity found by Woodward between unit production and process organizations may be resolved by conceiving of two basically different kinds of organizations-one geared chiefly to per-