[R-meta] Meta-analysis of mean differences

Dear Martin

I think the relevant part of the link to which Wolfgang directed you is

"In practice, one is likely to encounter difficulties in actually 
obtaining those correlations from the information reported in the 
articles. In that case, one can substitute approximate values (e.g., 
based on known properties of the dependent variable being measured) and 
conduct a sensitivity analysis to ensure that the conclusions from the 
meta-analysis are unchanged when those correlations are varied."

I suggest that you choose a range of plausible values for r and fit the 
model for each. If the scientific conclusion is unchanged over a range 
of values then you have your answer. If the exact value of r makes an 
important difference then you are a bit stuck unless you can persuade 
any of the authors to supply their r value.

Michael

[R-meta] Meta-analysis of mean differences

Thread (4 messages)