This vignette explains how mfrmr fits MML
models and how to interpret the newer strict marginal diagnostics.
For a response vector \(\mathbf{x}_n\) and parameter vector \(\beta\), the current MML path
targets the marginal likelihood
\[ L(\beta) = \prod_{n=1}^{N} \int p(\mathbf{x}_n \mid \theta, \beta) g(\theta) \, d\theta \approx \prod_{n=1}^{N} \sum_{q=1}^{Q} w_q \, p(\mathbf{x}_n \mid \theta_q, \beta), \]
where \((\theta_q, w_q)\) are
Gauss-Hermite nodes and weights. In mfrmr, the integral is
approximated with Gauss-Hermite quadrature, the marginal log-likelihood
is optimized from the same shared kernel, and person summaries are
computed post hoc from the posterior bundle. When a latent-regression
population model is active, the package uses person-specific transformed
nodes derived from the same quadrature basis rather than one
unconditional fixed grid.
The posterior weight for person \(n\) at node \(q\) is
\[ \omega_{nq} = \frac{w_q \, p(\mathbf{x}_n \mid \theta_q, \hat{\beta})} {\sum_{r=1}^{Q} w_r \, p(\mathbf{x}_n \mid \theta_r, \hat{\beta})}. \]
Expected a posteriori (EAP) scoring then uses
\[ \hat{\theta}_n^{\mathrm{EAP}} = \sum_{q=1}^{Q} \theta_q \, \omega_{nq}. \]
This is the kernel that now feeds logLik, the gradient,
EAP summaries, and strict marginal expected values.
For the RSM / PCM MML branch:
mml_engine = "direct" uses gradient-based direct
optimization of the marginal log-likelihoodmml_engine = "em" and
mml_engine = "hybrid" are also available for
RSM / PCM; bounded GPCM uses the
direct engine regardless of that optionThis is the implemented MML-engine scope for this release.
The strict marginal branch is not based on plugging \(\hat{\theta}_n^{EAP}\) back into the response model. Instead, it works with posterior-integrated expectations. For a grouped summary \(g\) and category \(c\),
\[ \mathbb{E}_{\hat{\beta}}(N_{gc}) = \sum_{n=1}^{N} \sum_{q=1}^{Q} \omega_{nq} \, I(n \in g) \, P(X_n = c \mid \theta_q, \hat{\beta}). \]
The corresponding residual compares the observed count to that
latent-integrated expectation rather than to an EAP plug-in
prediction.
For pairwise local-dependence follow-up, the package keeps the same
posterior weights but replaces the one-category event with agreement or
adjacency events for the relevant pair of facet levels. That is why
top_marginal_cells and top_marginal_pairs are
conceptually related but not numerically comparable.
diagnose_mfrm() now keeps two evidence paths
explicit:
legacy: residual/EAP-oriented diagnostics inherited
from the earlier stackmarginal_fit: strict latent-integrated first-order and
pairwise screensboth: returns both without collapsing them into one
decision ruleThe object returned by summary(diag) exposes
diagnostic_basis so the two paths can be interpreted
separately.
The current design is deliberately aligned with five strands of the IRT fit literature.
Limited-information item-fit logic. Orlando and Thissen (2000,
2003) show why grouped or score-conditioned comparisons can be more
stable than full-information contingency-table statistics in realistic
IRT settings. The current package borrows that limited-information
logic, but it does not implement S-X2 or S-G2
literally. Instead, it applies posterior-integrated grouped residual
screens to many-facet cells and levels.
Generalized residual logic. Haberman and Sinharay (2013) define a generalized residual for a summary statistic \(T\) as
\[ r = \frac{T - \hat{\mathbb{E}}(T)}{\hat{s}_D}, \]
where \(\hat{\mathbb{E}}(T)\) and
\(\hat{s}_D\) are computed under the
fitted model. This is the clearest template for thinking about the
current marginal_fit outputs. The current pairwise
local-dependence summaries are informed by the same
observed-versus-expected logic, but they should still be read as
exploratory agreement screens rather than as formal Haberman- Sinharay
generalized residual tests.
Multi-method fit assessment and practical significance. Sinharay and Monroe (2025) review limited-information statistics, generalized residuals, posterior predictive checking, and practical significance, and recommend prioritizing fit procedures by intended use rather than treating one index as universally decisive.
Posterior predictive follow-up. Sinharay et al. (2006) treat
posterior predictive checking as a separate model-checking family built
around replicated datasets and discrepancy measures. That is the
intended follow-up role of the package’s
posterior_predictive_follow_up path, which is reserved for
a future release.
Many-facet reporting context. Linacre’s FACETS framework and applied MFRM studies such as Eckes (2005) remain the primary references for severity/leniency, mean-square fit, separation, and inter-rater agreement. The current strict marginal branch is designed to sit alongside that many-facet toolkit, not to replace it.
The strict marginal branch is currently a screening layer, not a fully calibrated inferential test battery.
This package therefore treats strict marginal diagnostics as structured evidence about possible misfit, not as a single definitive accept/reject rule. That design choice follows the broader review logic in Sinharay and Monroe (2025): use several complementary diagnostics, match them to the intended use of the scores, and examine practical significance before making strong claims.
The marginal_fit grid sums per-observation
posterior-integrated variances as
VarianceCount = sum_i w_i^2 * p_i * (1 - p_i), which treats
observations within a cell as independent Bernoulli draws given the
fitted parameters. This is the same independence assumption that
underlies conditional independence in the MFRM specification (Linacre,
1989). When the rating design violates it - e.g. multiple ratings by the
same rater on the same person-criterion combination - cell-level
posterior variances can be underestimated, which in turn makes the
reported standard errors and screening t-statistics slightly optimistic.
The documented follow-up is
diagnose_mfrm(..., diagnostic_mode = "both") combined with
strict_pairwise_local_dependence, which surfaces evidence
of dependence at the pair level. See Haberman and Sinharay (2013) for a
generalized-residual framework that models such dependence explicitly;
that path is not yet included in the current release scope.
For many-facet reporting, one additional boundary matters.
Facet-level separation/reliability and inter-rater agreement answer
different questions. High rater separation reliability can coexist with
weak observed agreement, and strong observed agreement does not imply
that raters are interchangeable on the latent severity scale. That is
why mfrmr reports diagnostics$reliability and
diagnostics$interrater as separate objects.
The current simulation-based validation covers:
These checks target RSM and PCM.
GPCM is now supported only within a bounded core route:
fitting, slope summaries, posterior scoring, information curves, direct
curve/category reports, exploratory residual-based follow-up, direct
recovery checks, and the slope-aware element-conditional
fair_average_table() and estimate_bias().
Historical fair-average SE columns remain scaled facet-measure SEs
rather than delta-method SEs of the fair-average value;
fair_average_table(fair_se = TRUE) adds distinct structural
delta-method fair-average SEs for non-person rows when the MML
observed-information Hessian is available. Bias SE / t /
Prob. columns are conditional plug-in screening quantities,
and bounded-GPCM rows also carry profile-likelihood follow-up columns.
Summary-table appendix export is available for supported direct outputs.
The APA writer, fit-based report/export bundles, FACETS-style score-side
exports, QC pass/fail pipelines, linking synthesis, and planning /
forecasting helpers remain out of scope for GPCM in this
release. See gpcm_capability_matrix() for the full
per-helper support contract.
GPCM is the current upper supported scope for three
reasons.
MML kernel and the response-probability core
already generalize to the bounded GPCM branch without
changing the main package architecture.This is a narrower but more defensible claim than saying the whole package is uniformly generalized to free-discrimination many-facet work.
Robitzsch and Steinfeld (2018) are helpful because they separate two arguments that are often conflated in applied many-facet work.
RSM/PCM fit.If the intended score interpretation requires equal contributions of
items and raters, then the Rasch-family route remains substantively
attractive even when a slope-aware model fits better. mfrmr
therefore treats RSM / PCM as the
equal-weighting reference models and bounded GPCM as a
supported alternative for users who explicitly want to inspect or allow
discrimination-based reweighting.
This is also why score-side FACETS output-contract routes (for
example, score-side facets_output_file_bundle()),
manuscript-grade APA text, and fit-based report/export bundles remain
out of scope for bounded GPCM in this release: their
published forms are Rasch-family score transformations, and the
slope-aware analogue that would replace them in a free-discrimination
context requires careful delta-method standard-error handling.
fair_average_table() and estimate_bias()
themselves are now available under bounded GPCM via the
slope-aware element-conditional kernel. For fair averages,
fair_average_table(fair_se = TRUE) adds structural
delta-method SEs for non-person rows when the MML Hessian is available;
the historical SE columns remain measure-level SEs.
One additional distinction matters for implementation. The
weight argument in fit_mfrm() is an
observation-weight column. It changes how rating events enter estimation
and summaries, but it is not the same thing as the equal-weighting
versus discrimination-weighting question discussed above.
For users deciding among RSM, PCM, and
bounded GPCM, use this reading order:
RSM; if thresholds may vary by a designated step
facet, begin with PCM.method = "MML" and read summary(fit) plus
summary(diagnose_mfrm(fit, diagnostic_mode = "both")).GPCM only when discrimination-based
reweighting is a meaningful sensitivity question, not as a routine
replacement for the equal-weighting model.GPCM fits better, report what changed:
slopes, information redistribution, bias-screening rows, fair averages,
or conclusions. Do not let fit improvement alone decide the operational
model.This sequence keeps the interpretation aligned with the validation
boundary: RSM/PCM support the full
manuscript/reporting route, while bounded GPCM supports the
documented direct and caveated routes.
Posterior-predictive checking, MCMC engines, and heavier
runtime infrastructure remain future extensions. They are not required
for the current quadrature-based MML route or for the
bounded GPCM support described here.
For the current release, the most defensible interpretation sequence is:
summary(fit) for estimation status and precision
basis.summary(diag) with
diagnostic_mode = "both" to keep legacy and strict evidence
separate.marginal_fit and marginal_pairwise
as screening layers for first-order and local-dependence follow-up.