Moderation doubt

Posted on
No user picture. JuanJMV Joined: 07/20/2016
Hi everyone,
I am trying to test if there is a difference in one variable (quantitative) due to another variable (binary but different for each twin).
These variables are related and I want to test if there are different “subtypes” of that disorder (quantitative variable) depending on the value of the other variable (binary variable).
I was thinking to do a moderation analysis (as Rob suggested me in a previous post) as follows:
1-Quantitate variable
2-Moderation variable coded 1:concordant type A; 2: concordant type B; 3: discordant

I would like to know if this approach is correct and If I get significant different values of heritability between type A and Type B, could I conclude that for that disorder could exist different etiologies

Thank you so much in advance.

Replied on Fri, 08/03/2018 - 14:29
No user picture. JuanJMV Joined: 07/20/2016

In reply to by AdminRobK

Sorry! I am going to try to explain it better.

I want to test if the heritability of one variable (quantitative) is different due to another variable (binary).

For example, if the heritability of neuroticism is different due to sleep quality (poor sleep quality vs adequate sleep quality)
Then I have thought to do a moderation analysis with one continuous variable (neuroticism) and 1 binary variable as moderator (sleep quality).
But like the moderation variable is different for twin 1 and twin 2 I have thought to code the moderation variable as follows:
1: concordant for poor sleep quality
2: concordant for adequate sleep quality
3: discordant for sleep quality

I think this is similar to a sex-limitation analysis.
I would like to know if this approach is correct for doing a moderation analysis with one binary variable.
Thank you so much!