I used the samples' mean ages as a moderating variable for a meta-analysis on the relationship between two variables. I used the robumeta package and I found the mean age in 20 of the 21 studies (the number of effect sizes is 158). Yet, the degrees of freedom were lower than 4, and therefore, results were not trustworthy. Can someone provide me with a valid explanation of when results are not trustworthy and why? Might it be because the mean ages are all around 40 and variation is too low?

Thank you very much! Ellen

  • $\begingroup$ The package vignette explains that with low df the Satterthwaite approximation is not believed to be valid. Perhaps investigate that section of the vignette again? $\endgroup$ – mdewey Feb 14 '19 at 15:33

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.