Software

I am the author and developer of the R package puniform, which contains meta-analytic methods to correct for publication bias. Four methods are currently included in the package. The first is the p-uniform method (van Assen, van Aert, & Wicherts, 2015); van Aert, Wicherts, & van Assen, 2016), which can be used to estimate effect size, test the null hypothesis of no effect, and assess the presence of publication bias in a meta-analysis. The p-uniform method was later refined and extended in van Aert & van Assen (in press) resulting in an improved approach known as p-uniform*.

Another method in the package is the hybrid method. This method was initially developed as a meta-analysis method for combining one original study and one replication and while taking into account statistical significance of the original study (van Aert & van Assen, 2018). The method was later on extended to meta-analyze a mix of nonpreregistered original studies and preregistered replications (van Aert, in press).

The fourth method in the package is the Snapshot Bayesian Hybrid Meta-Analysis Method (van Aert & van Assen, 2017). This method computes posterior probabilities at four hypothesized effect sizes (e.g., no, small, medium, and large) based on an original study and replication while taking into account statistical significance in the original study. The method can also be used for computing the required sample size of the replication akin to power analysis in null hypothesis significance testing.

### Install puniform package
install.packages("puniform")
### Load puniform package
library(puniform)
### Access documentation
?puniform
?puni_star
?hybrid
?snapshot

### Install devtools package
install.packages("devtools")
### Install puniform package
devtools::install_github("RobbievanAert/puniform")
### Load puniform package
library(puniform)