CHARACTERIZING ENVIRONMENTAL VARIATION.

A hypothesized relationship between environmental variation and organizational actions is central to organization theory. This study shows that conventional summary measures of instability do not capture all dimensions of environmental variation. This problem is addressed with frequency, amplitude, and predictability measures that characterize the pattern of environmental variation. The hypothesis that there is a relationship between longitudinal environmental variation and organizational structures, actions, and outcomes is common to most of the major perspectives of organization theory (Child, 1972; Hannan & Freeman, 1977; Pfeffer & Salancik, 1978; Thompson, 1967). Although there has been little agreement across theories about the causal mechanisms linking environmental variation with organizational characteristics, organizational theorists have generally agreed that there is some relationship between longitudinal environmental change and organizing problems. As a result, environmental variability is a central independent construct in models of processes as disparate as strategic choice, firm failure, and executive turnover. Organizations experience longitudinal environmental variation as variation in the availability of the required for organizational survival (Dess & Beard, 1984: 52). These resources include sales and knowledge about product design and manufacturing processes in private enterprises, budget allocations in the public sector, personnel, and other production factors. Researchers have typically characterized variation in the availability of such resources over time with measures of volatility, dynamism, or instability that attempt to capture total environmental variation with a single dimension. We propose a characterization of the pattern of longitudinal environ