Hi /sci/,
I'm trying to find the proper categorical analysis tool to test the independence two contingency tables (which you could see as a 3D contingency table I guess).
I have the following table:
CatA1,CatA2,CatA3
40,190,50
200,150,50
90,110,20
(See next post for how I'm visualizing it)
Now, this table is from some point in time.
At another point in time, I have a similar table with different data.
What I'm trying to figure out is how to test for independence of the two tables.
If I just wanted to test the table in itself, the answer would of course be to do a [math]\chi^2[/math]-test, but I don't think that's valid here.
Could you help me out?
Pic related, it's the book I'm using.
This is the format I'm using to interpret the data at one point in time.
I'm not too sure but my first intuition would be this
http://sphweb.bumc.bu.edu/otlt/MPH-Modules/BS/BS704-EP713_Confounding-EM/BS704-EP713_Confounding-EM7.html
I would need to read up. I have that text book available so a page or reference would help me.
>>7812297
Thanks, I'll give this a read through.
It's not actually something from the book, but my supervisor gave me this one to read because he thought it might contain something for it.
I haven't been able to find anything about this kind of comparison so far though.
>>7812297
Having read it, I think this is exactly what I need.
Thanks a lot!
>>7812307
That makes it harder. I thought you were taking a course with that textbook. I am currently out of my office but I'll do some reading once I return.
>>7812320
The Cochran-Mantel-Haenszel Test is introduced on page 227 of that book, if you still want to look it up
>>7812319
I thought so. Your bill for Statistical consulting service will arrive shortly.