Minitab - Cross Tabulation and Chi-Square Test of Independence
Loading...
5,277
Uploader Comments (st8tistics)
see all
All Comments (4)
-
Quick questions - What is the difference between or significance of Pearson Chi-Square & Likelihood-Ratio Chi-Square.
Thanks for your help.
-
Oh my bad. I understand where you are coming from mathematically. Sorry to sound upset.
Loading...
I am telling you it is correct because I teach it. Read any intro statistics book - the understood location of the response variable is the column variable. It's always a possible question on the high school AP test it's that classic.
amoeba97 2 years ago
You sound upset - apologies - absolutely if that's what is tested you should teach it. I teach at university and have never asked that on an exam, mathematically it does not make a difference - it's a convention. Expected values are calculated if no relationship between the variables, so mathematically it doesn't matter. As I said before, if there is a preference because it makes things easier to understand, or because of convention that follow it. As many things are - like rejection below 0.05.
st8tistics 2 years ago
Usually the row variable is the explanatory variable; the column variable is the response variable (the variable of primary interest).
amoeba97 2 years ago
quite possibly that is correct! I find in practice it isn't always clear what is the response and what is explanatory - you just have two things you are interested in...and mathematically it doesn't make a blind bit of difference of course :-) can't remember what I said in the video but if it's a table to put in a report or whatever you can try it both ways and see which is easier to read...
st8tistics 2 years ago