Stratification of Sex and Quartiles... 2 way ANOVA?

#1
Hi all,

I'm conducting research and am looking at a large list of continuous variables as outcome measurements. My exposure is quartiles (4 groups), but I want it stratified by sex.

Would this be a simple 2 way anova? or is there a way to I'm really new to statistics so sorry if this is quite apparent.

For my categorical outcomes, I think it'd be best to go with a chi square but how to stratify for sex?

Any help would be much appreciated
 

Karabiner

TS Contributor
#2
My exposure is quartiles (4 groups), but I want it stratified by sex. Would this be a simple 2 way anova?
I would think so. You can decide whether or no you also want to include the sex*group
interaction in your analysis.

For my categorical outcomes, I think it'd be best to go with a chi square but how to stratify for sex?
What do you mean by categorical outcomes - maybe binary? This could be done using
binary logistic regression.

With kind regards

Karabiner
 
#3
Thanks for your reply.

The issue is I am able to tell whether sex or quartiles have an effect on my continuous outcome of choice, but I am not necessarily "stratifying" by sex. This is to say, I am interested in seeing outcomes values for simply male vs female based on quartiles.

Would it be easier to just divide men and women and run the anova without sex as an exposure?