BayesRep

BayesRep is an R package for the analysis of replication studies using Bayes factors.

Pawel, S., Held, L. (2022). The sceptical Bayes factor for the assessment of replication success. Journal of the Royal Statistical Society Series B: Statistical Methodology. 84(3): 879-911. DOI:10.1111/rssb.12491

Installation

## from GitHub
## install.packages("remotes") # requires remotes package
remotes::install_github(repo = "SamCH93/BayesRep")

Usage

library("BayesRep")

## Original effect estimate and standard error
to <- 0.2
so <- 0.05

## Replication effect estimate and standard error 
tr <- 0.15
sr <- 0.04

## Compute and format sceptical Bayes factor
scepticalBF <- BFs(to = to, so = so, tr = tr, sr = sr)
formatBF(scepticalBF)

#> [1] "1/20"

## Compute and format replication Bayes factor
repBF <- BFr(to = to, so = so, tr = tr, sr = sr)
formatBF(repBF)

#> [1] "1/521"

## Plot posterior distribution of effect size
repPosterior(to = to, so = so, tr = tr, sr = sr)

Plot of effect size posterior distribution based on original and replication data.