Standardizing and Visualizing Descriptive Summaries of Election Survey Data Huh, Myung-Hoe; Lee, Yong-Goo;
Survey reports of election opinions consist of numerous cross-tabulations between socio-demographic variables and political opinions including preferred candidates. Since socio-demographic variables are related each other, duplicate interpretations arise. The aim of this study is twofold: The first is to separate the effects of socio variables such as education, occupation and income from the effects of demographic variables such as region, sex and age. The second is the visualization of multiple cross-tabulations in low-dimensional space by extended doubling technique of correspondence analysis. Survey researchers may get some help from this study to present their survey results more lucidly and visually.
Opinion survey on election;direct standardization;correspondence analysis;doubling;
허명회 (1999). <다변량 수량화>, 자유아카데미
Greenacre, M. (1993). Correspondence Analysis in Practice, Academic Press, London
Greenacre, M. and Hastie, T. (1987). The geometric interpretation of correspondence analysis, Journal of the American Statistical Association, 82, 437-447