Model-based Cluster Analysis Using Variables Characterizing Types of Democracy

<p> This thesis applies model-based cluster analysis to data concerning types of democracies, creating an instrument for typologies which attempt conceptual classification based on an explicit theory. We note several advantages of model-based clustering over traditional clustering methods, including model choice guided by the principle of likelihood. Rather than splitting the variables into two sets as Lijphart (1999) does, we fit a normal mixture model for types of democracy in the context of the majoritarian-consensus contrast using Lijphart's data for thirty-six democracies, which consists of 10 variables for two periods, from 1945 to 1996 and from 1971 to 1996.</p> <p>The model for the full period finds four types of democracies: two types for the majoritarian-consensus contrast, and two mixed ones lying between two extremes. The solution of the four-cluster model for the full period shows that most of the countries have high conditional probabilities of belonging to their respective groups, and the solution is found to be quite stable with respect to possible measurement error in the variables included in the model. The model for the recent period finds (except for for Austria, Switzerland, and Venezuela) most countries remain in the same clusters as for the full-period data. </p>

Name thesis:degree_name

M.S.

Level thesis:degree_level

Campus Access Thesis

Discipline thesis:degree_discipline

Statistics

Year dc:date.available

2011

Author dc:creator

• Jang, Jaewon

Contributors dc:contributor

• David Hitchcock

• Physical Sciences and Mathematics
• Statistics and Probability
• Model-based cluster analysis
• Types of democracy
• Typology

Statement dc:rights

• © 2011, Jaewon Jang

Repository record dc:identifier

https://scholarcommons.sc.edu/etd/2582

OAI identifier oai:identifier

oai:scholarcommons.sc.edu:etd-3659