Hi all,
Wondering if anyone can throw a few statistical pearls of wisdom my way!! I thought I had worked out what to do with my data....but having just read Andy Fields stats book (my bible!) I think I might have a problem....

I'll provide an example. If we were to categorise individuals by their stats knowledge.

1 - low (me!)
2 - medium
3 - high (hopefully you!)

then give a 6 month stats course to all individuals.
To see if there are significant changes in the level of knowledge I thought I could could categorise my time points (1=pre course, 2=post course), categorise my knowledge level (1=low, 2=medium, 3=high). Then run a Chi-square to see if there is a significant difference in category distribution from pre to post.
However....I've read that you can't use a Chi square for repeated measures?? I don't relaly understand why? I guess that each individuals score will be effected by their baseline score...but if Im literally interested in if there is a significant change from the number of people in a category from time x to time y is this really a problem? If so what's the alternative???????

Thanks inadvance if theres any one that can help!

:-)