search

cran / metaumbrella

:exclamation: This is a read-only mirror of the CRAN R package repository. metaumbrella — Umbrella Review Package for R
9 stars 2 forks source link

In small meta-analyses, Egger´s test and 3-level structure cause an error #1

Open cyplessen opened 1 year ago

cyplessen commented 1 year ago

Hi all, thanks for the great package you provide! ✌️

I get Error in egger_reg$coefficients[2, 3] : subscript out of bounds for a meta-analysis that contains 5 effect sizes from 3 primary studies.

I could circumvent the problem by averaging the data first and using only the 3 studies without nested data.

The data for this specific meta-analysis looks like this:

structure(list(study = c("Enock 2014", "Enock 2014", "Pham 2016", "Pham 2016", "Stolz 2018"), measure = c("SMD", "SMD", "SMD", "SMD", "SMD"), factor = c("Weisel 2020", "Weisel 2020", "Weisel 2020", "Weisel 2020", "Weisel 2020"), meta_review = c("Weisel 2020", "Weisel 2020", "Weisel 2020", "Weisel 2020", "Weisel 2020"), value = c(0.645, -0.008, 0.0276355552083566, -0.084, 0.694 ), se = c(0.226, 0.14, 0.249, 0.249, 0.228), rob = c("high", "high", "unclear", "unclear", "low"), n_cases = c(53, 110, 32, 32, 42), n_controls = c(53, 110, 32, 32, 23), multiple_es = c("outcome", "outcome", "outcome", "outcome", "outcome")), row.names = c(NA, -5L), class = "data.frame")

umbrella <- umbrella(data, mult.level = TRUE)

gaborcsardi commented 1 year ago

Hi, this is a read only mirror of CRAN, please see the package authors in the DESCRIPTION file. Look for Maintainer, BugReports and URL. Thanks!