Dear colleagues,
How to compare frequencies in a 3x2 contingency table?
We've observed frequencies of 3 categorical variables within two independent groups, and we would like to test the null hypothesis that frequencies are not different across the groups.
How to perform this analysis in SAS Project?
Thanks!
Dear Paulo,
I think you need to use a generalized linear model but I don't have any experience with it.
Use Proc Freq:
For ordered categories:
data paulo;
input treat $ outcome count ;
cards;
test 0 3
test 1 8
test 2 14
placebo 0 6
placebo 1 9
placebo 2 5
;
proc freq;
weight count;
tables treat*outcome / chisq cmh nocol nopercent;
run;
The Chisq is the test for no association, the Cochran-Mantel-Haenszel Row Mean Scores Differ tests wether the means scores differ.
If the categories are nominal, drop the cmh option.
Cochran Mantel-Haenszel test will work in your situation.
SPSS video will guide you how to perform it:
https://www.youtube.com/watch?v=ilKZG4nmuLU
Dear Paulo
The chi-square crosstabulation will be suitable for your data, please you can read in the following link:
http://www.indiana.edu/~educy520/sec6342/week_10/chi_sq_summary011020.pdf
Good Luck
- Badges
- Science topic
- Similar topics
- Phytochemistry
- Extracts