Hi,
I want to do a meta-analysis on say, education and its association on earnings.
Many studies report one parameter, say, the effect of an additional year of schooling on earnings. I can work with that. Many other studies however report essentially group differences, instead of reporting the linear coefficient for years of schooling they report the coefficients of dummy variables for schooling, e.g. one study reports results for 1) 12 years and 2) 13+ years compared to 3) 0-11 years, or another study reports coefficients 1) 8-9 years, 2) 10-11 years, 3) 12+ years, compared to 4) 0-7 yeas of education.
I learned that there are many ways to convert all sorts of estimates into standardized effect sizes, but what if I have several estimates for essentially the same thing?
I hope this question isn't stupid or inappropriate, I tried googling it and couldn't find anything (maybe I was using the wrong terms?), I work with Stata 15.
Thanks
Nora
I want to do a meta-analysis on say, education and its association on earnings.
Many studies report one parameter, say, the effect of an additional year of schooling on earnings. I can work with that. Many other studies however report essentially group differences, instead of reporting the linear coefficient for years of schooling they report the coefficients of dummy variables for schooling, e.g. one study reports results for 1) 12 years and 2) 13+ years compared to 3) 0-11 years, or another study reports coefficients 1) 8-9 years, 2) 10-11 years, 3) 12+ years, compared to 4) 0-7 yeas of education.
I learned that there are many ways to convert all sorts of estimates into standardized effect sizes, but what if I have several estimates for essentially the same thing?
I hope this question isn't stupid or inappropriate, I tried googling it and couldn't find anything (maybe I was using the wrong terms?), I work with Stata 15.
Thanks
Nora
