[R-meta] Output value reported from metaprop function

[Danyang Dai]

Dear R Mata group members

Thank you for developing and contributing to the meta package. I enjoy
using R for my research work and it is great to have packages that help us
to do meta analysis.

I am currently conducting a meta analysis on prevalence using the metaprop
function (not for randomised control trials). For the transformation
method, I am using Freeman-Tukey Double arcsine transformation (sm="PFT"). I
am using both random effect and fixed effect. The printed output from the
metaprop function shows the proportion for random effect and fixed effect
as the *backtransf* in the *metaprop* function was set to be true by
default. However, when I try to use $ to extract the values, they are not
the same value as the printed output.  I looked up the documentation for
the metaprop function. It seems that the test_meta$TE.random the
untransformed value and if I want the proportion reported from the printed
meta output, I would need to use the *backtransf()* function. When I did
try to use the *backtransf() *to obtain the transformed proportions for the
fixed effect model and random effect model, I got somewhat similar values
but not exact values. I suspect that it is probably due to rounding errors
but I am wondering if there is a way to obtain consistent proportion values
from the printed values and the meta objects? I am also a bit confused with
the outputs and the meta estimate. If metaprop(*backtransf = TRUE*), then
it will print the proportion, if metaprop(*backtransf = FALSE*), then it
will print the transformed proportion. Would you please explain more about
which proportion and transformed proportion please?

I appreciate all your suggestions and help and I look forward to hearing
from you all! Thanks!

See below: (Please see the sample text_data and code attached for
reproducibility. )

[image: Screenshot 2024-08-16 at 10.54.03.png]
[image: Screenshot 2024-08-16 at 10.53.30.png]


Kind regards
Daidai
Github: https://github.com/DanyangDai

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://stat.ethz.ch/pipermail/r-sig-meta-analysis/attachments/20240816/573577ed/attachment-0001.html>

-------------- next part --------------
A non-text attachment was scrubbed...
Name: Screenshot 2024-08-16 at 10.54.03.png
Type: image/png
Size: 134810 bytes
Desc: not available
URL: <https://stat.ethz.ch/pipermail/r-sig-meta-analysis/attachments/20240816/573577ed/attachment-0002.png>

-------------- next part --------------
A non-text attachment was scrubbed...
Name: Screenshot 2024-08-16 at 10.53.30.png
Type: image/png
Size: 131388 bytes
Desc: not available
URL: <https://stat.ethz.ch/pipermail/r-sig-meta-analysis/attachments/20240816/573577ed/attachment-0003.png>

-------------- next part --------------
A non-text attachment was scrubbed...
Name: test_data.csv
Type: text/csv
Size: 255 bytes
Desc: not available
URL: <https://stat.ethz.ch/pipermail/r-sig-meta-analysis/attachments/20240816/573577ed/attachment-0001.csv>

-------------- next part --------------
A non-text attachment was scrubbed...
Name: test_meta.R
Type: application/octet-stream
Size: 438 bytes
Desc: not available
URL: <https://stat.ethz.ch/pipermail/r-sig-meta-analysis/attachments/20240816/573577ed/attachment-0001.obj>