Help for loglinear analysis

Log-linear models provide a method for analyzing associations between two or more categorical variables. The method has become widely accepted as a tool for researchers during the last two decades. 

The visualization for log-linear models help to specify the possible models for a group of variables and assess the fit of these models. The list of terms (the variables and all the interactions between the variables) in the visualization gives a simple and intuitive way of indicating the terms to enter in a model. 

If the list of terms is in hierarchical model, all the terms below a selected one are automatically introduced in the model. If the list of terms is not in hierarchical model, terms can be selected individually. The rest of plots help to assess the fit of the model like mosaic plots for observed and predicted values or distance of Cook v. leverages scatterplots. The parameter plot sorts out the parameters of the model and provides a interpretation in terms of frequencies of the original data of each parameter. Finally, the history plot keeps information about the chi square statistic that can be used for returning to previous successful models 
