Dear all
Apologies for cross-posting
I would be most grateful to receive advice on any simple algorithms for adjusting confidence intervals for odds ratios in multiple comparisons.
To illustrate the idea, suppose I have three categories of smoker ('non-smoker', 'ex-smoker' and 'current smoker') and wish to estimate the difference in odds of having ulcerative colitis if a person falls
into one of these categories rather than the other, but I wish to calculate all three relative odds. How can I adjust the confidence intervals for the corresponding univariate odds ratios in a Bonferroni fashion in this and other cases where there may indeed be more odds
ratios to calculate?
I look forward to receiving your suggestions, all of which should be most welcome.
Best wishes
Margaret
---------------------------------
Yahoo! Photos – NEW, now offering a quality print service from just 7p a photo.
|