I'm conducting a meta-analysis and assessing publication bias using trim-and-fill method. The amount of studies I've included is small (~40), and we are further breaking them into sub-groups depending on the metrics they use. For the subgroups that have only 3-6 studies, trim-and-fill will often remove all the studies in the subgroup, resulting in an empty funnel plot. What do i do?
Hello Tor,
The trim and fill method presumes that there will be a symmetric funnel plot in the first place (which may or may not be true). With very small numbers of cases/ES values with which to work, that may not be the case (regardless of whether publication bias was or was not in effect).
So, you're probably disaggregating your data set too finely for this procedure to be very helpful. I'd suggest avoiding the trim/fill procedure with such small case sets.
Good luck with your work.
The number of studies is small that is why it didnt work, in this case, just dont use this method in your sup-groups analysis.
As a rule of thumb, at least 10 studies are needed to achieve Trim fill method. However, I have seen 7 studies to attain this. The lesser number of studies leads to lower statistical power and increase false negative error. I think you should try Egger's test which is more powerful.
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Meta-Analysis