Dear Forum Members{
Do you agree that this is a mistake?
In Michael Borenstein's book, "Common Mistakes in Meta-analysis and how to avoid them " Section 8., Page 44, it is stated, " Some users have said that they use the fixed-effects model because the random-effects model gives equal weight to all studies …" I have been receiving promotion material for CMA 4.0 and a copy of this book is included in the package. .
One of the assumptions in his random-effects model is that the studies are a random sample from their universe (7.4.3 B, Page 26).
Is this really an error? Why or why not? I have asked Dr. Borenstein, President of Biostat Inc and developer of CMA 4.0 to weight in. I will give you feedback before the end of January/
Happy New Year,
Jon
------------------------------
Jonathan Shuster
------------------------------