Analysis of nominal data – multi-way contingency table

• Authors: Stanimir Agnieszka
• Languages of publication: EN

Abstracts

EN

Presented in this paper the method of graphical presentation of the relationship between nominal variables and their categories gives the opportunity for an extensive diagnosis of dependence variables. Correspondence analysis and mosaic plots are based on the same grounds, i.e. contingency table or multi-way contingency table. Correspondence analysis can be used in the study of relationships between two or more nominal variables without limiting the number of categories. In the case of many variables, the multidimensional contingency table is used very often. Only the difficulty of construction of such a table and the combined variables can affect the decision of a researcher about the validity of using this solution. For mosaic plots the situation is different. These graphs represent very well the relationships between two categories of nominal variables with few categories. The introduction of another variable to the study, which is described by two or three categories, is also not too problematic, and the graph is easy to interpret. However, if in a multi-way contingency table variables are a combination of several primary variables, described with many categories, the mosaic plot is no longer as clear as the projection made in correspondence analysis.

Keywords

EN

nominal data; multi-way contingency table; mosaic display; correspondence analysis

• Publisher: Wydawnictwo Uniwersytetu Ekonomicznego we Wrocławiu
• Journal: Mathematical Economics
• Year: 2011
• Issue: 7 (14)
• Pages: 229-240

Contributors

author

Stanimir Agnieszka

References

• Aktywności ekonomiczna ludności Polski. II kwartał 2011 (2011). GUS. Warszawa.
• Andersen E.B. (1997). Introduction to the Statistical Analysis of Categorical Data. Springer-Verlag. Berlin.
• Blalock H.M. (1975). Statystyka dla socjologów. PWN. Warszawa.
• Clausen S.E. (1998). Applied Correspondence Analysis. An Introduction. Sage. University Paper 121.
• Friendly M. (1992). Mosaic displays for loglinear models. In: Proceedings of the Statistical Graphic Section. Pp. 61-68.
• Friendly M. (1994). Mosaic display for multi-way contingency tables. Journal of the American Statistical Association. Vol. 89. No. 425. Pp. 190-200.
• Goodman L.A. (1963). On Plackett’s test for contingency table interactions. Journal of the Royal Statistical Society. Series B. Vol. 25. No. 1. Pp. 179-188.
• Greenacre M. (1993). Correspondence Analysis in Practice. Academic Press. London.
• Heijden van der, P.G.M. (1987). Correspondence Analysis of Logitudinal Categorical Data. DSWO Press. Leiden.
• Jobson J.D. (1992). Applied Multivariate Data Analysis. Vol. II. Categorical and Multivariate Methods. Springer-Verlag. New York.
• Hofmann H. (2000). Exploring categorical data: Interactive mosaic plots. Metrika. Vol. 51. Pp. 11-26.
• Stanimir A. (2005). Analiza korespondencji jako narzędzie do badania zjawisk ekonomicznych. Wydawnictwo Akademii Ekonomicznej. Wrocław.
• Walesiak M. (1996). Metody analizy danych marketingowych. PWN. Warszawa. Yule G.U, Kendall M.G. (1966). Wstęp do teorii statystyki. PWN. Warszawa.