Dear all,
I would like to seek advice from a bio-statistician with experience in
Meta-Analysis on the issues of heterogeneity measures:
- I-squared:
- Does I-squared <50% rule out heterogeneity or can this be
misleading?
i.e. in what circumstances would I^2<50% not be reliable?
- How to interpret 95% CIs for I-squared?
- To what extend is I-squared a flawed measure of heterogeneity? Do
viable alternatives exists?
- Cochran's Q test:
- Can Cochran's Q test reliably detect heterogeneity? What is a
typical cut-off?
- Alternatives to Cochran’s Q test?
Any advice / insights would be much appreciated
--
Best regards
Dr Schneider-Luftman
UK Med-Bio fellow in Biostatistics
Epidemiology & Biostatistics dept.
St Mary's Hospital, School of Public Health, Faculty of Medicine
Imperial College London
You may leave the list at any time by sending the command
SIGNOFF allstat
to [log in to unmask], leaving the subject line blank.
|