Janda and Gillies: "How well does 'region' explain political party characteristics?" --> back to first page
Summary and conclusion

This study was prompted by scholars' tendencies to generalize about the characteristics of parties in given regions of the world. The assumption behind these generalizations is that nations in the same region share, to some extent, a common political culture. This is an implicit form of the argument that environment affects organizational form and purpose [see also Lammers and Hickson (1979: especially chapter 22) and Childs (1981) for cultural effects on organizations in general]. A more precise test of the effect of environment on party characteristics would focus on the countries individually rather than on countries lumped into regions. As Harmel and Janda have shown (1982), country effects on party characteristics tend to be substantially larger than the regional effects detected in this paper. For example, the national environment accounts for about 68 per cent of the variance in centralization of power compared to 24 per cent for regional effects, and nation explains 57 per cent of organizational complexity compared to 20 per cent for region (Harmel and Janda, 1982).

Their study, however, addresses a different question from the one addressed here, which was constrained to study environmental effects only as they operate within cross- national culture areas. As long as parties in 10 homogeneous regions were studied, that theoretical framework was adhered to. The authors moved outside that framework when seeking to maximize success in classifying parties according to their configurations of characteristics. This study shows that the three worlds of development, which are strikingly heterogeneous, are indeed characterized by different types of party politics. Whatever causes their similarity in party characteristics, it cannot be the influence of common domestic political culture, especially in the diverse countries of the Third World. More likely, the similarities in parties, as a form of political institution, are due to the influence of international politics. Stauffer (1971) holds that some features of domestic politics in Third World countries are dictated by East-West tensions--especially governmental capacity to maintain stability through coercion. Alternatively, the 'world system' approach to analyzing domestic politics (Wallerstein, 1974; Hopkins et al., 1982) attributes differences in domestic politics to the position of nations at the 'core or 'periphery' of the international capitalist economy (see Snyder and Kick, 1979).

How good are regional explanations of party politics? The simple answer is that regional explanations are rather powerful. Although there are substantial variations among parties within regions on major concepts, region has a significant effect on each of 11 party characteristics measured. Moreover, region explains more than 20 per cent of the variance for eight of the concepts and more than 33 per cent for five concepts. These effects are observed when the world is divided rather finely into 10 regions and attention is focused on single party characteristics. When attention shifts to configurations of several characteristics simultaneously, these regions retain their explanatory potential in discriminant analysis, but they do not achieve as much success in classifying parties as does a global grouping of parties into the Western, Eastern, and Third Worlds of development. In sum, parties in the same regions do tend to share the same characteristics, and there is evidence for even gross generalizations about party politics at the global level. But these generalizations probably owe more to international politics than to common cultural factors.


Janda and Gillies: "How well does 'region' explain political party characteristics?" --> back to first page