The package VCD includes a function that computes some assoc stats for contingency tables, I belive that could be a good idea to include it in the section crosstabs > Crosstabs N to 1.
I would like to thank you for your suggestions. Unfortunately, most of the people contributing to RKWard (including myself) are a little busy, so don't expect miracles.
On the plugin: The main idea in developing RKWard is to use as less libraries as possible (memory and speed issues). This plugin N to 1 is based on table and chisq.test functions and also contains a barplot option with sufficient level parametrization. The vcd package will load itself and more three packages (MASS, grid and colorspace) and will give as return the likelihood ratio, Phi-Coefficient, Contingency Coeff. and Cramer's V. I really don't know where is the break-even point there and how useful it would be for the users.
Kind Regards,
Ilias
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
The package VCD includes a function that computes some assoc stats for contingency tables, I belive that could be a good idea to include it in the section crosstabs > Crosstabs N to 1.
There is a link to the package:
http://cran.r-project.org/web/packages/vcd/
and the function is assocstats. Ther is the example in the documentation:
data("Arthritis")
tab <- xtabs(~Improved + Treatment, data = Arthritis)
summary(assocstats(tab))
Also, I want to suggest the inclusion of prop.table, for the percentage of the cells in the table, respecto for the row, the column and the total
Congratulations for the program, is very good, I don't want to be demanding with you, I only want to help making suggestions.
Dear yayopoint,
I would like to thank you for your suggestions. Unfortunately, most of the people contributing to RKWard (including myself) are a little busy, so don't expect miracles.
On the plugin: The main idea in developing RKWard is to use as less libraries as possible (memory and speed issues). This plugin N to 1 is based on table and chisq.test functions and also contains a barplot option with sufficient level parametrization. The vcd package will load itself and more three packages (MASS, grid and colorspace) and will give as return the likelihood ratio, Phi-Coefficient, Contingency Coeff. and Cramer's V. I really don't know where is the break-even point there and how useful it would be for the users.
Kind Regards,
Ilias
I'm working in some changes for the plugin, the next link is for download my version:
http://rapidshare.com/files/122477126/crosstabplugin.tar.bz2.html
There have many errors, but it work
If you want i can mantain this plugins an make all the fixes, but I want that you test it and tell me if it good to you or not...
Please give me an answer to my mail
yayopoint@gmail.com
thanks...
Ahh, I'm starting to work in the relliability plug in, I submit it in a few days