Bayesian Informative Hypotheses Evaluation Web Applications
Researchers often have expectations about the relations between means
of different groups or standardized regression coefficients; using informative
hypothesis testing to incorporate these expectations into the analysis through
order constraints increases statistical power
Vanbrabant and Rosseel (2020) . Another valuable
tool, the Bayes factor, can evaluate evidence for multiple hypotheses without
concerns about multiple testing, and can be used in Bayesian updating
Hoijtink, Mulder, van Lissa & Gu (2019) . The 'bain'
R package enables informative hypothesis testing using the Bayes factor. The
'mmibain' package provides 'shiny' web applications based on 'bain'. The
RepliCrisis() function launches a 'shiny' card game to simulate the evaluation
of replication studies while the mmibain() function launches a 'shiny'
application to fit Bayesian informative hypotheses evaluation models from
'bain'.
mmibain
The Mighty Metrika Interface to BAIN (‘mmibain’) R package provides
Shiny apps to explore basic functionality of the
‘bain’
package for BAyesian INformative Hypotheses Evaluation.
Installation
You can install the released version of ‘mmibain’ from
CRAN:
install.packages("mmibain")
You can install the development version of ‘mmibain’ from
GitHub with:
# install.packages("devtools")
devtools::install_github("mightymetrika/mmibain")
Play RepliCrisis
‘RepliCrisis’ is a Shiny app game that simulates evalutating replication
studies based on the framework presented in Hoijtink, Mulder, van Lissa
& Gu (2019). Follow these steps to
play:
- Set your sample size (for groups within study), difficulty, alpha
level, and seed for reproducibility.
- Define thresholds for the Bayes Factor and Posterior Model Probability
to assess evidence in favor of the original study.
- Conduct the original study to generate data and form a hypothesis.
- Show diagnostics and descriptives to understand statistical results
and hypotheses.
- Conduct a replication study, using swap controls to match the original
study’s results.
- Run replication analysis to evaluate the results against the original
hypothesis.
- Start a new game by conducting a new original study.
To play, load ‘mmibain’ and call the RepliCrisis() function:
library(mmibain)
RepliCrisis()
mmibain Shiny App
The package also includes a Shiny app for running basic bain::bain()
models:
- Upload your data in CSV format.
- Choose your modeling engine (lm, t_test, lavaan).
- Input your model and any additional arguments.
- Fit the model and input hypotheses for evaluation.
- Adjust settings such as the fraction parameter, standardized
regression coefficients, and confidence intervals.
- Set a seed for reproducible results.
- Run the Bayesian Informative Hypotheses Evaluation.
Launch the app with:
mmibain()
References
Hoijtink, H., Mulder, J., van Lissa, C., & Gu, X. (2019). A tutorial on
testing hypotheses using the Bayes factor. Psychological methods, 24(5),
539–556. https://doi.org/10.1037/met0000201