You are here

metaSEM questions

2 posts / 0 new
Last post
GregoryVerleysen's picture
Joined: 02/26/2020 - 08:56
metaSEM questions
Binary Data Test.R3.83 KB

I am learning how to use metaSEM, but I am running into some issues.
Right now I only have 3 studies available to run the analysis (code with data is in attachment).

I am running into the following issues and was wondering if these would be caused by having only 3 studies
or am I doing something wrong here?

-When I perform the fixed effects version step one, I am getting negative AIC and BIC
-For the first step of random effects I keep getting the message to rerun it
-When trying to do mediation during step 2 I get the message:
"Error in solving the Hessian matrix. Generalized inverse is used. The standard errors may not be trustworthy."
Also TLI is minus infinity and AIC and BIC are zero.

Kind regards


Mike Cheung's picture
Joined: 10/08/2009 - 22:37
Hi Gregory,

Hi Gregory,

There is nothing wrong with negative AIC and BIC. See
With only 3 studies, I don't think that you can apply a random-effects model.