[R-meta] see how much each study contributes to heterogenicity

[Guido Schwarzer]

Am 13.10.21 um 22:00 schrieb Martin Lobo:

> Hi every one,
> as I can see if any of the studies particularly influenced the high level of heterogeneity. I did the sensitivity analysis with meteinf but I don't know how to find this point.

If you are using study labels (argument 'studlab' in meta-analysis 
functions of R package meta), it should be clear which study has a 
"suspicious" confidence interval.

In addition you can produce a Baujat plot which plots the contribution 
of each individual study to the heterogeneity on the x-axis and the 
contribution to the overall effect on the y-axis. See example(baujat.meta).

Best, Guido