Skip to content
Prev 3539 / 5632 Next

[R-meta] Pooling studies with binary and continuous outcomes

Dear Graham

It is not entirely clear to me what data each primary study is giving 
you. Comments below in-line
On 27/10/2021 21:13, Blackman, Graham wrote:
So what do they report, mean of the predictor in each group with its 
standard error?
Correlation between predictor and % change?

In general correlation and mean difference are measuring different 
things. Given two variables X and Y it is easy to construct an example 
where r(X, Y)=1 but |X - Y| is arbitrarily large.


  I would like to combine these different study designs to increase the 
statistical power.