Package: bayesqm 0.2.0

bayesqm: Bayesian Q Methodology: Exact Rank-Order Likelihood for Forced Q Sorts

A Bayesian analysis for Q methodology, alongside the classical one. Models the forced Q sort as an ordered partition of the statements through an exact rank-order likelihood (the design quotas fix the partition margins, so the likelihood of the observed sorting event is exact), fits it by a parameter-expanded Gibbs sampler in R with no compiled code and a convergence gate on rotation-invariant functionals, resolves rotational ambiguity via the MatchAlign post-processing of Poworoznek et al. (2025) <doi:10.1214/25-BA1544>, and returns the familiar Q tables as posterior summaries: credible intervals for bounded participant loadings, flag probabilities with an explicit unclassified state, quota-respecting factor arrays, distinguishing and consensus statements judged against a posterior critical difference and a grid-width equivalence region, one posterior false-discovery rule for all published claims, and a two-signal posterior-predictive workflow for the number of factors.

Authors:Raymond Dacosta Azadda [aut, cre], Henry Ofoe Agbi-Kaiser [aut], Hannah D. Robinson [aut], AK-ACE Team [aut], Karsten Hueffer [aut], Taa'aii Peter [aut], Stacy Rasmus [aut]