Democracy, Indices of
Democracy, Indices of
In order to document democratic development, a number of indices of democracy have been developed. These indices are formulated in a variety of ways, yet most are basically oriented to political, liberal democracy—what the political scientist Robert Dahl has called “participation and contestation” (Dahl 1971). Relatively high statistical correlations on the aggregated level also point to a far-reaching unity of viewpoint among those designing the different indices. The correlation coefficient (r) tends to lie at the 0.75–0.95 level (0.00 representing no concordance; 1.00 complete concordance). At the same time, there are differences—both conceptual and methodological—that make it significant which index is chosen. Experience has shown, namely, that in certain cases the choice of index can influence the results (Elkins 2000; Casper and Tufis 2003).
As a rule, extreme differences among countries (e.g., between Australia and Saudi Arabia) will register in much the same way no matter what index is used. Where more modest differences are concerned, however, the coding becomes more arbitrary, especially at lower or medium levels of democracy. It is likely, therefore, that changes over time will be registered differently by different indices. When studying gradual changes in the level of democracy, the results could, to a substantial degree, reflect which index is used (Hadenius and Teorell 2005).
The choice of index, in other words, is not trivial. Hence, it is important to account for the way they are constructed. One difference concerns what prime aspects of democracy are actually measured. Some indices are based on a broad range of criteria, which are aggregated into a scale. Freedom House (FH), which is a frequently used index, is a typical example. It includes an extensive checklist of political rights (related to elections) and civil liberties, which are certainly relevant. However, it also includes certain criteria that could be seen as less relevant, such as free enterprise, property rights, and lack of corruption. The other leading index, Polity, has a more limited focus. It accounts for a number of electoral requisites, but pays no heed to political freedoms.
A few other indices deserve to be mentioned. Reich applies criteria that relate both to elections and political freedoms, whereas Vanhanen and Alvarez et al. are more constrained. Vanhanen looks at electoral participation (which is a strongly contested indicator) and the share of parliamentary seats for the governing party, while Alvarez et al. is concentrated on a fairly narrow set of electoral criteria.
One question that has occasioned controversy has to do with dichotomous versus continuous measures of democracy. The main argument in favor of the dichotomous approach (which distinguishes only between democracies and nondemocracies) is that this divide is the most essential one. The main counterargument is that this is too rough an assessment, and that it therefore misses more graded differences. Among the indices mentioned, only Alvarez et al. is dichotomous in character.
Indices also differ with respect to methodological qualities, such as how the operational measures have been chosen and what rules of coding and aggregation have been applied. It is important that the transformation of the data be openly displayed, so that the process can be replicated and made the subject of both testing and alternative coding and aggregation.
Polity, Vanhanen, and Alvarez et al. are generally held to meet high methodological standards, especially with respect to transparency. The construction of the Freedom House scale is a more concealed process, while Reich is even more problematic in this respect, for it provides almost no information about the operationalizations and the coding rules applied.
On balance—considering both conceptual and methodological aspects—Polity and Freedom House are the most useful indices. Comparing these indices for possible biases (which have been argued to exist), it turns out that they do not differ with respect to region, religion, or colonial background. But the type of regime does matter, at least in some respects. Freedom House treats traditional monarchies more favorably, while Polity is generally more “dichotomous” in character (most cases are located at the top or the bottom of the scale). Polity therefore applies greater rewards to democratic improvements. In addition, Freedom House is more concerned about political violence and repression, while Polity puts stronger emphasis on electoral performance.
It seems that in most instances in the 1970s and 1980s, Freedom House tended to overestimate the level of democracy, whereas Polity underestimated it. During the 1990s, however, it was the other way around: Freedom House made relatively strict assessments, to the effect that democracy tended to be underrated. The opposite held for Polity. An average score, therefore, based on the two indices, seems to give the most accurate assessment.
SEE ALSO Democracy; Democratization
BIBLIOGRAPHY
Alvarez, Mike, José Antonio Cheibub, Fernando Limongi, and Adam Przeworski. 1996. Classifying Political Regimes. Studies in Comparative International Development 31: 3–36.
Casper, Gretchen, and Claudiu Tufis. 2003. Correlation versus Interchangeability: The Limited Robustness of Empirical Findings Using Highly Correlated Data Sets. Political Analysis 11: 196–203.
Dahl, Robert. 1971. Polyarchy: Participation and Opposition. New Haven: Yale University Press.
Elkins, Zachary. 2000. Gradations of Democracy? Empirical Tests of Alternative Conceptualizations. American Journal of Political Science 44 (2): 287–294.
Freedom House. 1994. Freedom in the World 1994. New York: Freedom House.
Hadenius, Axel, and Jan Teorell. 2005. Assessing Alternative Indices of Democracy. C&M Working Papers. International Political Science Association. http://www.concepts-methods.org/working_papers/20050812_16_PC%206%20Hadenius%20&%20Teorell.pdf
Marshall, Monty, and Keith Jaggers (2002). Political Regime Characteristics and Transitions, 1800–2002: Dataset Users’ Manual. Polity IV Project, University of Maryland. http://www.cidcm.umd.edu/inscr/polity.
Munck, Gerardo L., and Jay Verkuilen. 2002. Conceptualizing and Measuring Democracy: Evaluating Alternative Indices. Comparative Political Studies 35 (1): 5–34.
Reich, Garry. 2002. Categorizing Political Regimes: New Data for Old Problems. Democratization 9 (4): 1–24.
Vanhanen, Tatu. 2000. A New Dataset for Measuring Democracy, 1810–1998. Journal of Peace Research 37 (2): 251–265.
Axel Hadenius