mfrmr
Estimates many-facet ordered-response measurement models with arbitrary facet counts, rating-scale and partial-credit parameterizations, a generalized partial-credit extension with discrimination varying across levels of one selected facet. Marginal maximum likelihood permits a different facet to own category steps; joint maximum likelihood requires the same facet for both roles. Conditional maximum likelihood fits many-facet rating-scale and partial-credit models by conditioning on person totals, with structural contrast intervals and no assumed ability distribution. Experimental two-family estimation and conditional scoring retain explicit uncertainty limitations. The fit, diagnose and report workflow covers anchoring, linking, bias and differential-functioning screening, with saved results for replay. Observed-score generalizability and decision studies support assessment planning under declared rating designs. Agreement, rater feedback and conditional classification summaries connect analysis to assessment decisions. See 'Andrich' (1978) <doi:10.1007/BF02293814>, 'Masters' (1982) <doi:10.1007/BF02296272>, and 'Muraki' (1992) <doi:10.1177/014662169201600206> for the underlying ordered-response models.
This page was built for software: mfrmr