I think you can safely include gender in your model as a covariate. Rather than 1/-1 you should mean correct this column.
I would not encourage you to examine gender differences or interactions in such a small and skewed sample.
Colin Hawco, PhD
Neuranalysis Consulting
Neuroimaging analysis and consultation
www.neuranalysis.com
[log in to unmask]
-----Original Message-----
From: SPM (Statistical Parametric Mapping) [mailto:[log in to unmask]] On Behalf Of Mahsa
Sent: April-04-16 3:37 PM
To: [log in to unmask]
Subject: Re: [SPM] Use of covariates
Hi
I came across this post when trying to figure out the best model in a similar situation. Here it's been suggested to use "gender" either as factor or as covariate with a vector of 1s/-1s for each gender. My question is what would be the best option when there are only few subjects of one gender. In my case, I'm comparing 3 groups, each group including around 15 subjects where male subjects consist 20 – 37% of all subjects. Here's the exact numbers:
Group 1: 11 F / 5M
Group 2: 9 F / 2 M
Group 3: 10 F / 6 M
Would choosing "gender" as a factor in this case affect statistical power?
Many thanks in advance
Mahsa
|