babebi: An R Package for Bayesian Estimation and Validation in Small-N Two-Rater Pre–Post Designs

Description: Small pre–post rating studies are common in applied educational and psychological research, particularly in field settings where sample size is constrained before statistical planning begins (Shen et al., 2011). In such designs, limited sample size affects estimate precision and the interpretation of observed change (Faber & Fonseca, 2014; Lakens, 2022). babebi (Gianeselli et al., 2026) is an R package (R Core Team, 2026) that provides a workflow for complete two-time, two-rater pre–post rating designs in which recorded outcomes may be affected by rater-related differences. babebi standardises user-supplied data into a common long-format structure and analyses pre–post effects through difference scores in a linear model with a rater indicator as covariate. The package estimates pre–post change for a reference rater and rater-related differences in recorded change for a second rater. Posterior summaries for the regression coefficients are obtained from a large-sample normal approximation centred on the least-squares estimate, with plug-in covariance under a flat improper prior. Evidence for a non-zero pre–post effect, adjusted for rater differences, is summarised using a BIC-based approximation to the Bayes factor for the comparison between models with and without the pre–post effect (Kass & Raftery, 1995; Schwarz, 1978). The package also includes Monte Carlo validation calibrated from the observed design. The validation routine uses quantities estimated from the data, including sample size, mean pre–post change, and second-rater additive discrepancy. It summarises inferential performance through bias, root mean squared error, empirical 95% interval coverage, posterior tail probability summaries, and mean Bayes factor values.