what statistical test to choose (large smple size)

#1
Hi all,
I need to perform a test which shows if there is a statistical difference between males and females in getting diagnosis. I believe I should perform a chi square. But, is it OK to do (since the sample is large). Attached is the data.


males females
total 12,340,524 17,104,707
diagnosed 374,332 438,832
undiagnosed 11,966,192 16,665,875
rate of diagnosed 3.03 2.57


Please help. Any help would be appreciated.


Thanks!
 

hlsmith

Omega Contributor
#2
Yup, that would be the test to use. Without a computer in front of me - it is likely it may be significant. Though, is a < 0.5 clinically significant?
 

rogojel

TS Contributor
#3
h,
actually a two-sample proportions test should work as well. I just ran the test, it is highly significant at p<0.001 but the difference is rather small, 0.5%. This looks like a case of statistical significance without much practical significance. By the sample sizes you have any small effect will be highly statistically significant BTW.

regards