비교정치사회연구에 있어서 혼합체계분석안의 적실성

김웅진 (1986), “比較分析技法의 再照明: 空間的 統制樣式을 중심으로.” 『한국정치학
회보』 20:1.
김웅진 (1993), 『비교정치연구의 논리』. 서울: 전예원.
김웅진·김지희 (1994), “비교정치연구의 分析戰略과 디자인: 通則生産의 基本 規準에
관한 재성찰.” 『한국외국어대학교 논문집』 27.
Almond, G. and Genco, S. (1977), “Clouds, Clocks, and the Study of Politics.”
World Politics 29:4.
Benjamin, R. (1977), “Strategy versus Methodology in Comparative Research.”
Comparative Political Studies 9:4.
DeFelice, E. (1986), “Causal Inference and Comparative Methods.” Comparative
Political Studies 19:3.
Frendreis, J. (1983), “Explanation of Variation and Detection of Covariation: The
Purpose and Logic of Comparative Analysis.” Comparative Political Studies
16:2.
Gayle, D. (1988), “Applying ‘Most Different Systems’ Designs: Comparing
Development Policy in Alabama and Jamaica..” Comparative Political
Studies 21:2.
Hage, J. and Meeker, F. (1988), Social Causality. Winchester, Mass.: Allen &
Unwin.
Lijphart, A. (1971), “Comparative Politics and Comparative Method.” The
American Political Science Review 65:3.
__________. (1975), “The Comparable-cases Strategy in Comparative Research..”
Comparative Political Studies 8:2.
Meckstroth, T. (1975), “‘Most Different Systems’ and ‘Most Similar Systems’: A
Study in the Logic of Comparative Inquiry.” Comparative Political Studies
8:2.
Przeworski, A. and Teune, H. (1970), The Logic of Comparative Social Inquiry.
New York: Wiley.
Ragin, C. (1987), The Comparative Method: Moving Beyond Qualitative and
Quantitative Strategies. Berkeley and Los Angeles: University of California
Press.
Ragin, C. and Becker, H., eds. (1992), What is A Case? Exploring the Foundations of Social Inquiry. New York: Cambridge University Press.
Teune, H. (1968), “Measurement in Comparative Research.” Comparative Political
Studies 1:1.
_________. (1975), “Comparative Research, Experimental Design, and the
Comparative Method.” Comparative Political Studies 8:2.
Sartori, G. (1970), “Concept Misformation in Comparative Politics.” The American
Political Science Review 64:4.


□ ABSTRACT □

A Methodological Assessment of the `Mixed Systems Approach` in
Comparative Sociopolitical Inquiry: Levels of Analysis, Detection of
Covariation, and Causal Inference.

Ungjin Kim and Ji-hee Kim


`The Mixed Systems Approach` in comparative sociopolitical inquiry, suggested by John Frendreis in 1983, is employing the Joint Method of Agreement and Difference in detecting cross-cases causal conjuncture, or covariation, between independent and dependent variables. This approach, Frendreis argues, effectively solves the problems stemming from what Arend Lijphart calls `inherent weaknesses` of the comparative method - `Too many variables, Too small number of cases` - by systematically integrating the logic of the Most Similar Systems Design(the Method of Difference) and the Most Different Systems Design(the Method of Agreement). More specifically, when the Mixed Systems Approach is applied, one is free from finding `matched cases` in terms of the variability of dependent variables. In this respect, Frendreis continues, generalizations drawn from the analysis may show higher degree of spatial-temporal applicability than those produced by the previous designs.
This strategy, however, suffers from the severe lack of specific theoretical guidances for selecting the levels of analysis in measuring experimental variables; measurement levels for both independent and dependent variables are virtually `open`. Therefore, if a number of variables are selected from different measurement levels, serious theoretical, and then methodological, problems arise in empirically detecting multi-levels and cross-levels relationship (a) among the independent variables, and (b) between independent and dependent variables, within a single analytic model. In other words, detection of covariation among the variables operating at different levels should be done on the basis of unsubstantiated theoretical hypotheses, or pure conjectures, thereby violating the basic assumptions of empirical causal inference. In sum, the Mixed Systems Approach is nothing more than a simple covaration analysis strategy; it merely shows how the logic of the Joint Method of Agreement and Difference may be applied in cross-cases studies. It needs, more than anything else, theoretical sophistications for selecting proper levels of measurement to gain methodological relevance in comparative sociopolitical analysis.



CONTRIBUTORS

□ UNGJIN KIM(Ph. D., University of Cincinnati) is Professor of Political Science at Hankuk University of Foreign Studies. His fields of teaching and research interest are methodology/epistemology of the social sciences, comparative politics(comparison strategies and designs), and sociology of knowledge. He has published a substantial number of scholarly articles appeared in such professional journals as The Korean Political Science Review and The Korean Journal of International Relations. He is also the author or editor of several books including Methods, Methodology, and Political Science in Korea(1994), The Logic of Comparative Political Inquiry(1993), Political Science Methodology: An Introduction(1992), and Readings in Comparative Politics, Vol. 1, 2, 3(1992, 94, 95). His current research focuses on the sociopolitical nature of paradigmatic formation in the social sciences.

□ JIHEE KIM(Ph. D. Candidate, Hankuk University of Foreign Studies) is Instructor of Local Politics and Administration at Hollym Junior College. She is currently working on her doctorate in political science at Hankuk University of Foreign Studies. Her research interests include multivariate statistical analysis, Boolean Algorithms for comparative political inquiry, and voting behavior in Korea. She is the coauthor of `Strategies and Designs in Comparative Political Analysis` appeared in The Journal(Hankuk University of Foreign Studies, 1994), and contributed a translation of H. Ross & E. Homer`s article `Galton`s Problem in Cross-National Research`(1976) to Readings in Comparative Politics, Vol. 1, edited by U. Kim and others(Seoul: Hanul, 1995).