Developing an understanding of interorganizational systems: Arguments for multi - level analysis and structuration theory Shirley Gregor Department of Computing and Information Systems Central Queensland University Rockhampton QLD 4701 Australia Robert B. Johnston Department of Information Systems The University of Melbourne Parkville VIC 3052 Austra lia Abstract-Strategies and policies for the adoption and development of interorganizational systems require further understanding of the theoretical background to these systems. An argument is made for development of theory that is multi-level, processual and has an emergent perspective. Such theory is needed to deal with a context where environmental influences are important in addition to complex interactions between organizational activities at the micro-level and industry structure at the macro-level. The use of structuration theory as a vehicle to advance further understanding is explored. An illustration is given of application of this theory in the development of supply chain management in the beef industry. I. INTRODUCTION With developments in telecommunications there is an increasing trend for information systems to span boundaries between countries, organizations and the relatively separate components of large, geographically dispersed corporations. These interorganizational systems (IOS) include electronic data interchange (EDI), supply chain management (SCM), electronic funds transfer, electronic forms, electronic messaging, and shared databases [1]. Such systems provide the foundation for electronic business (e -business) -- a matter of great economic concern in today’ s world. A number of questions arise concerning strategies and policies for the adoption and development of interorganizational systems. What conditions within an industry or organizational grouping pa rticularly favour adoption of e-business? What are the points of leverage that can be exploited to help an industry in the introduction of e - business? Why have certain industries been able to adopt e - business technologies to reform supply chain management while others have not? Yet when we try to turn such questions into research agendas we are hampered by the lack of theory that can account for action at this broad level of analysis. Interorganisational systems research has dealt with the issues that aris e when systems cross -corporate boundaries, the difficulties of partnerships, and so forth. This research, however, has focused on interorganisational interactions to a limited degree. Action research projects in which the authors are engaged have indicated the need for theory to assist with answers to the questions asked above. In the apparent absence of a suitable theoretical base we have turned first to other empirical work to identify apparent influences on the development and use of interorganization al systems – with the aim of identifying forces and activities that appear to either encourage or inhibit the development of interorganizational systems. These empirical studies suggest that multi-level analysis is needed to deal with the problem. In particular, it appears theory is needed that includes industries or industry-clusters as a unit of analysis, as well as the enterprises within the industry. Theory is needed that encompasses the activities of a large group of firms and support organisations, which includes firms in the direct value chains, infrastructure providers, regulators, and trade organisations that have a business interaction focussed on a particular product. The theory also needs to encompass the forces in the external environment tha t act on the industry and the enterprises within it. Using the terminology of [2], we expect that theory needs to be developed that is mixed -level, in that the problem requires analysis at several different levels – both macro and micro. At the macro -level, we need theory that considers the “ industry-as-a whole” as a unit of analysis. For example, when we consider the impact of external forces on the industry we are considering the industry as an entity with properties and behaviour of its own – such as the price of its products or a response to legislation. At this level the relationships that characterize the structure of the industry appear to be particularly important. At a lower level we can look at the properties and behaviour of the individual enterprises that constitute the industry. Iacovou and Benbasat’ s study [3] of the adoption of electronic data interchange by small firms, with the individual firm as the unit of analysis, is an example of a study focusing on this level of analysis. It wou ld be