2
$\begingroup$

I have results from the same regression on two datasets, namely t statistics and standard errors $t_1,t_2,SE_1, SE_2$, what method would be appropriate for a meta-analysis?

$\endgroup$
1
  • 2
    $\begingroup$ Would you like to expand on this? What do you mean by saying you have values of t and their standard error? Do you mean you have the regression coefficients? And do you only have two data-sets to combine or was that just for illustration? $\endgroup$
    – mdewey
    Aug 15 '16 at 17:52
2
$\begingroup$

I got the effect sizes from $\beta_i = t_i SE_i$, and followed these formulas:

$\endgroup$

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.