'Introduction' In their path-breaking article, Low & MacMillan (1988) suggest that entrepreneurship be defined as the 'creation of new enterprise'. The purpose of entrepreneurship research should be to 'explain and facilitate the role of new enterprise in furthering economic progress' (p. 141). Such a delineation, they hold, would encourage researchers to consider both micro and macro perspec-tives. They argued that researchers must acknowledge that entrepreneurship studies could and should be carried out at multiple levels of analysis and that these analyses complement each other. The reasons for studying entrepreneurship on multiple levels of analysis lie in the charac-teristics of the entrepreneurial phenomenon itself. Entrepreneurship takes place and has effects on different societal levels simultaneously. Schumpeter (1934) already linked the entrepreneurial initiatives of individuals to the creation and destruction of industries as well as to economic de-velopment. Several other scholars have contributed to increasing our understanding about entre-preneurship on different levels of analysis, ranging from the individual to the economy-at-large. The following paragraph highlights some of the levels of analysis that have been identified. In doing so it illustrates the richness of approaches. It is individuals who carry out entrepreneurial initiatives (Schumpeter, 1934). These initia-tives take place in organizational contexts (Moran & Ghoshal, 1999, Shane & Venkataraman, 2000), often resulting in the formation of new firms (Gartner, 1988; Schumpeter, 1934) or the rejuvenation and improved performance of established firms (Covin & Slevin, 1991; Lumpkin & Dess, 1996; Wiklund, 1998; Zahra, 1991). Entrepreneurial initiatives often result in innovations, which in turn may alter existing industries (Schumpeter, 1934), or create new ones (Aldrich & Martinez, this issue). The belief that such processes have profound effects on employment and economic growth on the societal level (Baumol, 1993; Birch, 1979; McGrath, 1999) is one of the major reasons for the increased interest in entrepreneurship. The above does not only illustrate that studies on different levels of analysis can be valu-able, but clearly shows that these levels are intimately entwined. Therefore, as Low and MacMil-lan suggested (1988, p. 152), there may be reason to integrate different levels of analysis in em-pirical research......(continues) In the remainder of the article, level of analysis refers to the hierarchy of aggregation in terms of micro and aggregate level. More fine-grained categorizations of micro (e.g. individual, team, firm) and aggregate (e.g. region, nation) levels are possible as well as alternative hierar-chies (e.g., firm, industry, economy-at-large vs. firm, region, nation). The level on which the principal research questions are posed and analyses carried out rather than the level at which data are collected determines the level of analysis. It is, for instance, common to first collect and then aggregate data from individuals in regional studies of entrepreneurship. If the analyses compare regional differences in entrepreneurial activity based on the aggregation of individuals, this would be a study at the regional level even though data were collected from individuals.
