Analysis of emigration in Poland with the use log-linear models

Log-linear analysis is a statistical method used for the independence analysis of qualitative data in contingency table. Cell counts are Poisson distributed and all variables are treated as response. This method allows to analyze any number of variables in a multi-way table. Log-linear models, where interaction terms are included, enable to examine various types of association (conditional model, complete independence model, homogenous association model, saturated model, null model). In log-linear analysis we model cell counts in a contingency table in terms of associations among variables and marginal frequencies. For testing the goodness of fit the chi-square test, likelihood ratio test and information criteria AIC [3] and BIC [12] are used. Visualizing with the use of several plots designed for categorical data will be presented to enhance the interpretation. The data presented in this paper come from Demographic Yearbook of Poland 2012 and include report on emigration in Poland in 2011. All calculations will be conducted in R.