Correct way of reporting Chi square with IV 4 levels results

#1
Sorry, if the question is silly. Cannot find anywhere what will be a correct way of reporting Chi Square results for the case when IV has 4 levels (4 different nationalities)? The study was checking differences between personal experiences descriptions for participants from 4 different countries, all DV are dichotomous.
 

hlsmith

Less is more. Stay pure. Stay poor.
#2
What makes reporting these results different than any other time?


If you want to partially parcel out which groups may be differing amongst themselves, you can report the chi-square residuals as well.
 
#3
What makes reporting these results different than any other time?
The simple fact that I don't think I've ever saw such results reported and wanted to be sure that there are no special tips, tricks and regulations for these cases.
Thanks for the idea about residuals, wasn't sure about it myself.
 

ondansetron

TS Contributor
#4
... wanted to be sure that there are no special tips, tricks and regulations for these cases.
...
Unfortunately, any such guidance is often dismissed as unimportant or inconsequential by researchers without a proper statistics background. Thank you for trying to make sure you have complete, ethical, and rational reporting of your results! :)

I'm not sure if this is the same thing as what hlsmith said, but you can try showing each cell's contribution to the overall chi-square value as a way to highlight where deviations from expectations may be more or less prominent. You could also just report the overall chi-square statistic for the table, but then report the percentage of total count in each cell-- that is, report the absolute cell frequency with the relative frequency for that cell. If cell 1,1 has 10 observations out of the 50 total, you could make a table with 10 (20%) in the first cell, for example.