Jesper N. Wulff


The goal of alphaN is to help the user set their significance level as a function of the sample size. The function alphaN allows users to set the significance level as function of the sample size based on the evidence and the prior features they desire. The function JABt and JABp converts test statistics and \(p\)-values into sample size dependent Bayes factors. JAB_plot plots the Bayes factor as a function of the \(p\)-value, and alphaN_plot() plots the alpha level as a function of sample size for a given Bayes factor.


You can install the development version of alphaN from GitHub with:

# install.packages("devtools")


Here is an example: We are planning to run a linear regression model with 1000 observations. We thus set n = 1000. The default BF is 1 meaning that we want to avoid Lindley’s paradox, i.e. we just want the null and the alternative to be at least equally likely when we reject the null.


alpha <- alphaN(n = 1000, BF = 1)
#> [1] 0.008582267

Therefore, to obtain evidence of at least 1, we should set our alpha to 0.0086.