Bain stands for Bayesian informative hypothesis evaluation. It computes Bayes factors for informative hypotheses in a wide variety of statistical models. Just run your analysis as usual, and then apply bain to the output. A tutorial is available at DOI:10.31234/osf.io/v3shc.
Install bain from CRAN:
Add bain to your existing R workflow, and obtain Bayes factors for your familiar R analyses! Bain is compatible with the pipe operator. Here is an example for testing an informative hypothesis about mean differences in an ANOVA:
# Load bain
library(bain)
# dplyr to access the %>% operator
library(dplyr)
# Iris as example data
iris %>%
# Select outcome and predictor variables
select(Sepal.Length, Species) %>%
# Add -1 to the formula to estimate group means, as in ANOVA
lm(Sepal.Length ~ -1 + Species, .) %>%
bain("Speciessetosa < Speciesversicolor = Speciesvirginica;
Speciessetosa < Speciesversicolor < Speciesvirginica")
#> Bayesian informative hypothesis testing for an object of class lm (ANOVA):
#>
#> Fit_eq Com_eq Fit_in Com_in Fit Com BF PMPa PMPb
#> H1 0.000 0.447 1.000 0.500 0.000 0.224 0.000 0.000 0.000
#> H2 1.000 1.000 1.000 0.165 1.000 0.165 66166997632.868 1.000 0.859
#> Hu 0.141
#>
#> Hypotheses:
#> H1: Speciessetosa<Speciesversicolor=Speciesvirginica
#> H2: Speciessetosa<Speciesversicolor<Speciesvirginica
#>
#> Note: BF denotes the Bayes factor of the hypothesis at hand versus its complement.