The construction of explanatory theory about the determinants of regional integration is contingent upon our prior ability to measure integration in a typology or set of dimensions as a description and classification of that dependent variable. Previous attempts to conceptually differentiate the types of regional integration are reviewed and found to be contradictory. In order to generate empirically informed estimates about which typifications of integrative behavior are the most powerful and reasonable for theory construction, Joseph Nye's framework was selected for analysis. Construct validation was performed by operationalizing Nye's variables and confronting his scheme with data drawn from the Southeast Asian regional subsystem in order to ascertain the scheme's region-specific applicability. Factor analysis provided partial support for Nye's construct. The evidence derived here suggests that regional integration is a multidimensional phenomenon and that efforts to construct a single, multipurpose index of integration are not warranted. The major dimensions of regional integration emergent were (a) societal interdependence, (b) attitudinal integration, and (c) intergovernmental cooperation. The findings suggest that it is most meaningful to formulate theories in terms of the sources of each distinct type of integration; moreover, it appears that the dimensionality of integration may vary across regions and/or types of national actors.
1 Charles W. Kegley, Jr. is an Associate Professor in the Department of Government and International Studies and Chairman of the International Studies Program at the University of South Carolina, Columbia, South Carolina. Llewellyn D. Howell, Jr. is an Assistant Professor of International Relations at the School of International Service, The American University, Washington, D.C.