Reputation: 1
I am trying to create a forest plot in R using forest()
in the metafor
package. Since I am conducting a meta-analysis of proportions, I would like to use an alternate means of calculating confidence intervals (such as the Wilson score interval) to avoid negative lower bounds. I haven't been able to find a way of specifying an alternate method of calculating CIs when calling 'forest', and I also don't see a way to pass in a vector of CIs I calculate myself. Does anyone have a way of easily accomplishing this? Thanks in advance!
Previously I tried passing vectors of manually calculated CIs into the ci.lb and ci.ub parameters in forest():
forest(all_studies.rma, ci.lb = all_studies.df$ci_lb_wilson, ci.ub = all_studies.df$ci_ub_wilson)
however these appear to no longer be valid parameters since I get this error:
In segments(...) : "ci.lb" is not a graphical parameter
Upvotes: 0
Views: 20