facets_guide('coverage') summarizes how mfrmr maps
the main FACETS output-table, output-file, and graph-menu surface to package
functions. It is a surface-coverage guide, not a statement that estimands,
conditioning, extreme-score handling, degrees of freedom, or numerical
results are equivalent.
Use this helper before migration work when you need a public, user-facing answer to three questions:
which FACETS outputs have a close
mfrmrroute,which outputs are only partially covered by structured R objects,
which FACETS-specific outputs are not implemented or intentionally outside the current package scope.
Value
A data.frame with columns:
FACETSAreaFACETSFeatureFACETSReferencemfrmrRouteStatusSurfaceCoverageStatisticalContractValidationEvidenceOperationalStatusCapabilityLimitationAlternative
Details
The matrix is based on the FACETS 64-bit output index, which lists output
Tables 1–14, DIF/bias plots, R/Web plots, output files, and graph-menu
curves. mfrmr intentionally prioritizes structured R tables and reusable
plot data over exact FACETS line-printer output.
The executed comparison uses FACETS 64-bit 4.5.0. The online update history lists 4.5.3 (October 2026); current web help is not a version-pinned executable specification. In particular, Table 13 web df and executed df differ. The 4.5.1 history replaces the gtheory interface while its help still names gtheory; mfrmr's independent G-study does not depend on that menu. Numerical comparisons require their recorded executable version, design and estimand.
Status meanings:
implemented: a package-native route covers the substantive output surface; this status alone does not claim an externally matched statistical contract.supported_with_caveat: a package-native route exists, but the output must be read with explicit identification, validation, or scope caveats.partial: the concept is covered, but not the full FACETS formatting, option surface, file type, or external integration.not_implemented: a FACETS feature has no direct package-native route in the documented package scope.not_targeted: the feature is tied to FACETS UI, Web/Excel handoff, or another external program format and is outside the package scope.
The four contract axes are deliberately independent:
SurfaceCoveragerecords whether a corresponding package surface exists; familiar visual grammar belongs only to this axis.StatisticalContractrecords the package-native statistical scope and never implies a FACETS-matched estimand.ValidationEvidencerecords whether this table establishes matched external numerical evidence. It currently does not; that conclusion requires a separate comparison using a matched model and estimand.OperationalStatusrecords route availability. A package route being available does not makemfrmroperationally interchangeable with FACETS.
For observed-score assessment planning, mfrm_g_study() and
mfrm_d_study() support their documented univariate or
multivariate component designs. The older mfrm_generalizability() /
mfrm_d_study() pair has a different main-effects decomposition. Neither
route reproduces the FACETS interface to the external gtheory package.
For balanced ANOVA, confint.mfrm_multivariate_gstudy() adds component
SEs/intervals for a score or fixed composite under normal random effects;
two-score covariance intervals have explicit generalized and conservative
choices with signed bounds. Explicit reml_profile variance and covariance intervals
also accept MINQUE(0), with a separate REML center and no sampling SE;
two-score covariance profiles fit a joint normal model and retain signed bounds.
vcov.mfrm_multivariate_gstudy() separately supplies component sampling
covariance/SEs for balanced ANOVA and identifiable MINQUE sources using
normal product moments; these describe the original estimator, not REML.
Individual G/Phi intervals from balanced two-crossed-facet sources default
to approximate MLS null inversion. Undefined MLS widths are conservatively
retained and flagged; numerical availability does not guarantee coverage.
G/Phi and SEM have an explicit experimental reml_profile option for balanced or
identifiable incomplete sources; the separate REML fit retains original
ANOVA/MINQUE points and does not guarantee finite-sample or boundary coverage.
With reference, it maps one shared profile interval for one-facet plans
or nested plans with a common future child count, including incomplete sources.
Balanced crossed G also accepts explicit experimental score intervals,
retaining empty-set and numerical diagnostics; Phi is not supplied by this route.
mfrm_multivariate_d_compare()
supplies a separate, limited normal-theory paired-plan calculation. See
vignette("mfrmr-gtheory", package = "mfrmr") for a runnable workflow.
References
Linacre, J. M. (2026). A user's guide to FACETS, version 4.5.0. Winsteps.com. See the guide's output-table index for the documented FACETS files, plots, and graphs.
Examples
facets_guide('coverage')
#> FACETSArea
#> 1 Output table
#> 2 Output table
#> 3 Output table
#> 4 Output table
#> 5 Output table
#> 6 Output table
#> 7 Output table
#> 8 Output table
#> 9 Output table
#> 10 Output table
#> 11 Output table
#> 12 Output table
#> 13 Output table
#> 14 Output table
#> 15 Output table
#> 16 Output table
#> 17 Output table
#> 18 Output table
#> 19 Output table
#> 20 Output table
#> 21 Output table
#> 22 Output table
#> 23 R/Web plots
#> 24 R/Web plots
#> 25 R/Web plots
#> 26 R/Web plots
#> 27 R/Web plots
#> 28 R/Web plots
#> 29 R/Web plots
#> 30 Output file
#> 31 Output file
#> 32 Output file
#> 33 Output file
#> 34 Output file
#> 35 Output file
#> 36 Output file
#> 37 Output file
#> 38 Output file
#> 39 Graph menu
#> 40 Graph menu
#> 41 Graph menu
#> 42 Graph menu
#> 43 Graph menu
#> 44 Graph menu
#> 45 Specification/workflow
#> 46 Specification/workflow
#> 47 Specification/workflow
#> 48 Current scope boundary
#> 49 Current scope boundary
#> 50 Current scope boundary
#> 51 Current scope boundary
#> 52 Current scope boundary
#> 53 Observation weighting
#> 54 Current scope boundary
#> 55 Current scope boundary
#> 56 Current scope boundary
#> FACETSFeature
#> 1 Table 1: specification summary
#> 2 Table 2: data summary report
#> 3 Table 3: main iteration report
#> 4 Table 4: unexpected responses
#> 5 Table 5: measurable data summary
#> 6 Table 6.0: all-facet Wright map rulers
#> 7 Table 6.0.0: disjoint element listing
#> 8 Table 6.2: graphical facet statistics
#> 9 Table 7: facet measurement report
#> 10 Table 7: reliability and chi-square
#> 11 Table 7: agreement statistics
#> 12 Table 8.1: dichotomous/binomial/Poisson statistics
#> 13 Table 8.1: polytomous rating-scale/partial-credit statistics
#> 14 Table 8: scale-structure bar chart
#> 15 Table 8: scale-structure probability curves
#> 16 Table 9: bias-estimation iteration report
#> 17 Table 10: unexpected after allowing for bias
#> 18 Table 11: bias-calculation counts
#> 19 Table 12: bias summary report
#> 20 Table 13: DIF/bias detail report
#> 21 Table 14: pairwise bias report
#> 22 DIF/bias Excel plot
#> 23 Scatterplots and histograms from FACETS menus
#> 24 X-Y plot: R Statistics
#> 25 X-Y plot: Webpage
#> 26 X-Y-Z plot: R Statistics
#> 27 Histogram: R Statistics
#> 28 Generalizability Theory via R package gtheory
#> 29 Connectivity network graph via igraph
#> 30 Specification settings file
#> 31 Anchor output file
#> 32 Graph plotting file
#> 33 Output report file
#> 34 Residuals output file
#> 35 Score output file
#> 36 Simulated data file
#> 37 Subset group-anchor file
#> 38 Winsteps control and data file
#> 39 Category probability curves
#> 40 Expected score ICC/IRF
#> 41 Cumulative probability curves
#> 42 Test information function
#> 43 Category information function
#> 44 Conditional probability curves
#> 45 Full FACETS command-file parser and UI option surface
#> 46 Exact FACETS line-printer report emulation
#> 47 Raw FACETS report-text import
#> 48 FACETS or third-party frozen-calibration import
#> 49 General threshold or step anchors and starting-value import
#> 50 Multiple observed scales and scale-specific PCM
#> 51 Nominal/multinomial response models
#> 52 Binomial-trial and Poisson/count response models
#> 53 Row-frequency weights for ordered ratings
#> 54 Native multidimensional estimation and dimension-specific scores
#> 55 Unrestricted GPCM
#> 56 FACETS free-slope polytomous GPCM comparison
#> FACETSReference
#> 1 table1.htm
#> 2 table2.htm
#> 3 table3.htm
#> 4 table4.htm
#> 5 table5.htm
#> 6 table6.htm
#> 7 table6_0_0.htm
#> 8 table6_2.htm
#> 9 table7.htm
#> 10 table7summarystatistics.htm
#> 11 table7agreementstatistics.htm
#> 12 table8_1dichotomous.htm
#> 13 table8_1ratingscale.htm
#> 14 table8barchart.htm
#> 15 table8curves.htm
#> 16 table9.htm
#> 17 table10.htm
#> 18 table11.htm
#> 19 table12.htm
#> 20 table13.htm
#> 21 table14.htm
#> 22 difbiasplot.htm
#> 23 outputtableindex.htm
#> 24 xyplotr.htm
#> 25 xyplotwebpage.htm
#> 26 xyzplotr.htm
#> 27 histogramr.htm
#> 28 gtheory.htm
#> 29 networkgraph.htm
#> 30 specificationfile.htm
#> 31 anchorfile.htm
#> 32 graphoutputfile.htm
#> 33 outputfile.htm
#> 34 residualfile.htm
#> 35 scorefile.htm
#> 36 simulatedfile.htm
#> 37 subsetfile.htm
#> 38 winstepsfile.htm
#> 39 graphs.htm
#> 40 graphs.htm
#> 41 graphs.htm
#> 42 graphs.htm
#> 43 graphs.htm
#> 44 graphs.htm
#> 45 index.htm
#> 46 outputtableindex.htm
#> 47 outputtableindex.htm
#> 48 current mfrmr public contract
#> 49 current mfrmr public contract
#> 50 current mfrmr public contract
#> 51 current mfrmr public contract
#> 52 models.htm
#> 53 current mfrmr weight contract
#> 54 current mfrmr public contract
#> 55 current mfrmr public contract
#> 56 t7menu.htm
#> mfrmrRoute
#> 1 specifications_report()
#> 2 data_quality_report(); describe_mfrm_data()
#> 3 estimation_iteration_report()
#> 4 unexpected_response_table(); plot(x, type = "unexpected")
#> 5 measurable_summary_table(); describe_mfrm_data()
#> 6 plot(renderer = "facets", type = "wright", view = "unified"); plot(fit, type = "wright")
#> 7 subset_connectivity_report()
#> 8 facet_statistics_report(); plot(...)
#> 9 fit_measures_table(); diagnose_mfrm(); summary(fit)
#> 10 facets_chisq_table(); diagnose_mfrm()$reliability
#> 11 interrater_agreement_table(); rater_network_analysis(); rater_halo_network_analysis(); plot(x, type = "agreement")
#> 12 rating_scale_table() for two-category ordered scores
#> 13 rating_scale_table(); category_structure_report()
#> 14 category_structure_report()
#> 15 category_curves_report(); plot(fit, type = "ccc")
#> 16 estimate_bias(); bias_iteration_report()
#> 17 unexpected_after_bias_table()
#> 18 bias_count_table()
#> 19 summary(estimate_bias(...)); plot(x, type = "bias")
#> 20 estimate_bias(); bias_interaction_report()
#> 21 bias_pairwise_report(); build_fixed_reports()
#> 22 plot(plot = ..., type = "bias")
#> 23 plot_data(); package plot helpers
#> 24 plot_data(); user-defined R plotting
#> 25 none
#> 26 plot(fit, type = "ccc_surface"); plot_data()
#> 27 plot_data(); plot(fit, type = "wright"); plot(x, type = "qc")
#> 28 mfrm_g_study(); mfrm_d_study(); mfrm_multivariate_d_compare(); mfrm_generalizability(); mfrm_d_study()
#> 29 subset_connectivity_report(); mfrm_network_analysis(); rater_network_analysis(); rater_halo_network_analysis(); plot(..., type = "network")
#> 30 build_mfrm_manifest(); build_mfrm_replay_script()
#> 31 make_anchor_table(); export_mfrm_bundle(include = "anchors")
#> 32 facets_output_file_bundle(include = "graph"); read_facets_results(graph_file = ...); plot()
#> 33 export_summary_appendix(); build_fixed_reports()
#> 34 write_mfrm_residual_file(); diagnose_mfrm(); unexpected_response_table(); residual plot helpers
#> 35 facets_output_file_bundle(include = "score"); read_facets_results(); facets_fit_review(); plot()
#> 36 simulate_mfrm_data(); build_mfrm_sim_spec()
#> 37 write_mfrm_subset_file(); group_anchors; review_mfrm_anchors(); make_anchor_table()
#> 38 none
#> 39 category_curves_report(); plot(fit, type = "ccc")
#> 40 plot(fit, type = "pathway"); category_curves_report()
#> 41 category_curves_report(); plot(..., type = "cumulative")
#> 42 compute_information(); plot(type = "tif")
#> 43 category_curves_report(); plot(..., type = "category_information"); compute_information(); plot(type = "iif")
#> 44 category_curves_report()
#> 45 run_mfrm_facets(); fit_mfrm()
#> 46 build_fixed_reports() for selected tables
#> 47 read_facets_results() for delimited/fixed-field score extracts
#> 48 no external import; native route: mfrmr_output_guide("calibration_capabilities"); load_mfrm_calibration(); score_mfrm_calibration()
#> 49 not available in the current package
#> 50 not available in the current package
#> 51 not available in the current package
#> 52 not available in the current package
#> 53 fit_mfrm(weight = ...)
#> 54 not available in the current package
#> 55 not available for unrestricted GPCM
#> 56 none as a direct common-estimand comparison
#> Status SurfaceCoverage
#> 1 implemented available
#> 2 implemented available
#> 3 partial partial
#> 4 implemented available
#> 5 partial partial
#> 6 implemented familiar_visual_grammar_available
#> 7 supported_with_caveat available_with_caveat
#> 8 partial partial
#> 9 supported_with_caveat available_with_caveat
#> 10 supported_with_caveat available_with_caveat
#> 11 supported_with_caveat available_with_caveat
#> 12 partial partial
#> 13 supported_with_caveat available_with_caveat
#> 14 partial partial
#> 15 implemented available
#> 16 implemented available
#> 17 implemented available
#> 18 partial partial
#> 19 partial partial
#> 20 implemented available
#> 21 implemented available
#> 22 partial partial
#> 23 partial partial
#> 24 partial partial
#> 25 not_targeted out_of_scope
#> 26 partial partial
#> 27 partial partial
#> 28 supported_with_caveat available_with_caveat
#> 29 implemented available
#> 30 partial partial
#> 31 implemented available
#> 32 implemented available
#> 33 partial partial
#> 34 implemented available
#> 35 partial partial
#> 36 partial partial
#> 37 partial partial
#> 38 not_implemented unavailable
#> 39 implemented available
#> 40 implemented available
#> 41 implemented available
#> 42 implemented available
#> 43 implemented available
#> 44 partial partial
#> 45 not_targeted out_of_scope
#> 46 not_targeted out_of_scope
#> 47 partial partial
#> 48 not_implemented unavailable
#> 49 not_implemented unavailable
#> 50 not_implemented unavailable
#> 51 not_implemented unavailable
#> 52 not_implemented unavailable
#> 53 supported_with_caveat available_with_caveat
#> 54 not_implemented unavailable
#> 55 not_implemented unavailable
#> 56 not_implemented unavailable
#> StatisticalContract ValidationEvidence
#> 1 package_native_not_facets_equivalent external_match_not_established
#> 2 package_native_not_facets_equivalent external_match_not_established
#> 3 partial_package_native_contract external_match_not_established
#> 4 package_native_not_facets_equivalent external_match_not_established
#> 5 partial_package_native_contract external_match_not_established
#> 6 package_native_not_facets_equivalent external_match_not_established
#> 7 package_native_caveated external_match_not_established
#> 8 partial_package_native_contract external_match_not_established
#> 9 package_native_caveated external_match_not_established
#> 10 package_native_caveated external_match_not_established
#> 11 package_native_caveated external_match_not_established
#> 12 partial_package_native_contract external_match_not_established
#> 13 package_native_caveated external_match_not_established
#> 14 partial_package_native_contract external_match_not_established
#> 15 package_native_not_facets_equivalent external_match_not_established
#> 16 package_native_not_facets_equivalent external_match_not_established
#> 17 package_native_not_facets_equivalent external_match_not_established
#> 18 partial_package_native_contract external_match_not_established
#> 19 partial_package_native_contract external_match_not_established
#> 20 package_native_not_facets_equivalent external_match_not_established
#> 21 package_native_not_facets_equivalent external_match_not_established
#> 22 partial_package_native_contract external_match_not_established
#> 23 partial_package_native_contract external_match_not_established
#> 24 partial_package_native_contract external_match_not_established
#> 25 not_applicable not_applicable
#> 26 partial_package_native_contract external_match_not_established
#> 27 partial_package_native_contract external_match_not_established
#> 28 package_native_caveated external_match_not_established
#> 29 package_native_not_facets_equivalent external_match_not_established
#> 30 partial_package_native_contract external_match_not_established
#> 31 package_native_not_facets_equivalent external_match_not_established
#> 32 package_native_not_facets_equivalent external_match_not_established
#> 33 partial_package_native_contract external_match_not_established
#> 34 package_native_not_facets_equivalent external_match_not_established
#> 35 partial_package_native_contract external_match_not_established
#> 36 partial_package_native_contract external_match_not_established
#> 37 partial_package_native_contract external_match_not_established
#> 38 not_available not_applicable
#> 39 package_native_not_facets_equivalent external_match_not_established
#> 40 package_native_not_facets_equivalent external_match_not_established
#> 41 package_native_not_facets_equivalent external_match_not_established
#> 42 package_native_not_facets_equivalent external_match_not_established
#> 43 package_native_not_facets_equivalent external_match_not_established
#> 44 partial_package_native_contract external_match_not_established
#> 45 not_applicable not_applicable
#> 46 not_applicable not_applicable
#> 47 partial_package_native_contract external_match_not_established
#> 48 not_available not_applicable
#> 49 not_available not_applicable
#> 50 not_available not_applicable
#> 51 not_available not_applicable
#> 52 not_available not_applicable
#> 53 package_native_caveated external_match_not_established
#> 54 not_available not_applicable
#> 55 not_available not_applicable
#> 56 not_available not_applicable
#> OperationalStatus
#> 1 package_route_available
#> 2 package_route_available
#> 3 partial_workflow_only
#> 4 package_route_available
#> 5 partial_workflow_only
#> 6 package_route_available
#> 7 available_with_mandatory_caveat
#> 8 partial_workflow_only
#> 9 available_with_mandatory_caveat
#> 10 available_with_mandatory_caveat
#> 11 available_with_mandatory_caveat
#> 12 partial_workflow_only
#> 13 available_with_mandatory_caveat
#> 14 partial_workflow_only
#> 15 package_route_available
#> 16 package_route_available
#> 17 package_route_available
#> 18 partial_workflow_only
#> 19 partial_workflow_only
#> 20 package_route_available
#> 21 package_route_available
#> 22 partial_workflow_only
#> 23 partial_workflow_only
#> 24 partial_workflow_only
#> 25 external_only
#> 26 partial_workflow_only
#> 27 partial_workflow_only
#> 28 available_with_mandatory_caveat
#> 29 package_route_available
#> 30 partial_workflow_only
#> 31 package_route_available
#> 32 package_route_available
#> 33 partial_workflow_only
#> 34 package_route_available
#> 35 partial_workflow_only
#> 36 partial_workflow_only
#> 37 partial_workflow_only
#> 38 blocked
#> 39 package_route_available
#> 40 package_route_available
#> 41 package_route_available
#> 42 package_route_available
#> 43 package_route_available
#> 44 partial_workflow_only
#> 45 external_only
#> 46 external_only
#> 47 partial_workflow_only
#> 48 blocked
#> 49 blocked
#> 50 blocked
#> 51 blocked
#> 52 blocked
#> 53 available_with_mandatory_caveat
#> 54 blocked
#> 55 blocked
#> 56 blocked
#> Capability
#> 1 Structured run settings and reproducibility context.
#> 2 Rows, exclusions, missingness, score support, and response-pattern QC.
#> 3 Convergence and replayed iteration evidence.
#> 4 Case-level unexpected-response screening.
#> 5 Facet coverage, category counts, and subset/connectivity checks.
#> 6 Common-logit person/facet/threshold display with FACETS-style asterisk ruler or native SE/CI rendering.
#> 7 Co-observation graph components and facet coverage, with the fitted identification audit.
#> 8 Facet statistics and visual summaries.
#> 9 Measures, SEs, fit, anchoring status, and review flags.
#> 10 Rasch/FACETS-style separation, reliability, and chi-square summaries.
#> 11 Raw-category observed agreement and model-probability expected agreement, plus pairwise rater-network, rater-by-criterion halo network, and rater-agreement views.
#> 12 Two-category Rasch-category summaries are available.
#> 13 Core category diagnostics and thresholds for one RSM scale or one PCM step facet on the package's shared observed score scale.
#> 14 Category structure and transition summaries.
#> 15 Category probability and expected-score curve data.
#> 16 Bias recalibration path and final iteration status.
#> 17 Unexpected rows after the current bias-screening layer.
#> 18 Cell counts, logit shifts and separate baseline/remaining score residuals.
#> 19 Distributional and visual bias summaries are available.
#> 20 Ranked cell-level bias/interactions with screening statistics.
#> 21 Pairwise contrasts for two-way bias runs.
#> 22 R-native scatter, heatmap, and facet-profile bias displays.
#> 23 Reusable plot data supports custom R graphics.
#> 24 Users can build X-Y plots from returned data frames.
#> 25 No package-native Webpage plot generator.
#> 26 Selected 3D/surface-ready plot data are available.
#> 27 Several package outputs include histogram-like summaries.
#> 28 Observed-score univariate/multivariate component models for supported crossed/nested designs, fixed score composites and D-study plans, with component/covariance plots, curves and paired normal-theory plan comparisons. The separate main-effects route retains residual-scaling sensitivity and IdentificationStatus columns.
#> 29 Separate facet-level co-observation, rater score-relation, and rater-by-criterion halo networks with reusable node/edge tables.
#> 30 R-native reproducibility manifest and replay script.
#> 31 Candidate direct-anchor tables mechanically extracted from fitted estimates.
#> 32 Graphfile-style native output and separate import of reported category/expected-score curves.
#> 33 Structured appendix/report artifacts.
#> 34 Standalone observation-level residual CSV/TSV output, residual tables, and residual visualizations are available.
#> 35 Native observation-level score export and separate FACETS element-level import, including standalone source-measure and fit plots.
#> 36 Simulation data and explicit simulation specifications.
#> 37 Connected-subset summary/node files and group-anchor inputs/checks are available.
#> 38 No Winsteps control/data export route.
#> 39 Category probability curve data and plots.
#> 40 Expected-score curves over theta.
#> 41 Cumulative category-probability curve data, flipped direction data, and approximate .5 boundaries are available.
#> 42 Design-weighted test/scale information curves.
#> 43 Category-specific information contributions, total information curves, and facet/level contribution curves are available.
#> 44 Category probability curves conditional on theta are available.
#> 45 R function arguments are the package interface.
#> 46 Selected fixed-width handoff is available.
#> 47 Fit/score table import is supported.
#> 48 Native mfrmr portable artifacts support fixed-standard-normal RSM/PCM MML, qualified estimated-normal GPCM MML and qualified reference-prior RSM/PCM or shared-owner GPCM JML; consult mfrmr_output_guide("calibration_capabilities") for scope and source checks.
#> 49 No current public route accepts general threshold or step anchors or a threshold starting-value contract.
#> 50 Each fit uses one observed score scale and one homogeneous response-model family.
#> 51 The current RSM, PCM, and GPCM routes model ordered category probabilities only.
#> 52 Binary ordered scores are available as the two-category special case of the current ordered-response kernel.
#> 53 A positive numeric observation weight multiplies that row's conditional ordered-category likelihood contribution and can represent a defensible row-replication weight.
#> 54 The current public estimator and score routes are unidimensional.
#> 55 JML requires slope_facet to equal step_facet; MML permits separate owners with one slope family.
#> 56 FACETS PCM/JMLE can serve as the direct equal-discrimination comparison after the full estimation contract is aligned.
#> Limitation
#> 1 Not an exact FACETS line-printer layout.
#> 2 Structured QC replaces FACETS text layout.
#> 3 Does not reproduce the complete FACETS iteration trace.
#> 4 Structured table and plots, not printer-identical FACETS output.
#> 5 Does not reproduce FACETS response moments, raw/residual variance and global Pearson summaries; this is substantive partial coverage.
#> 6 Visual correspondence does not establish numerical equivalence; compare output from a documented FACETS version under aligned settings.
#> 7 These are not FACETS identification subsets; a single component can conceal confounded panels.
#> 8 FACETS M/S/Q/X printer-graph formatting is not reproduced exactly.
#> 9 Estimator and person-score basis, extreme handling, df/ZSTD conventions, and FACETS options can differ; external numerical equivalence is not established.
#> 10 Uses package-native structured output and documented df conventions; external numerical equivalence is not established.
#> 11 The package does not translate category positions across multiple independent scales, apply FACETS agreement-based SE inflation, or establish external Table 7 parity.
#> 12 FACETS binomial-trial and Poisson-specific reports are not implemented.
#> 13 Multiple independent scales and scale-specific anchors/starting values are not reproduced. Cumulative boundaries are grid approximations; category residual fit is descriptive, with no unit-baseline ZSTD test.
#> 14 FACETS line-printer artwork is not reproduced exactly.
#> 15 Uses R-native plot data rather than FACETS graph text.
#> 16 Conditional screening semantics are documented separately.
#> 17 Structured table replaces FACETS text layout.
#> 18 FACETS Table 11 response moments and variance decompositions are not supplied.
#> 19 FACETS vertical frequency bar-chart is not reproduced exactly.
#> 20 Reported as screening evidence, not final fairness inference.
#> 21 Higher-order runs omit pairwise sections by design.
#> 22 Excel-specific output is not implemented.
#> 23 FACETS arbitrary R/Web plotting menus are not mirrored.
#> 24 No dedicated FACETS-style arbitrary X-Y plot wrapper.
#> 25 Webpage menu output is a FACETS UI feature.
#> 26 No arbitrary FACETS X-Y-Z plot wrapper.
#> 27 No general FACETS histogram menu clone.
#> 28 Not a FACETS/gtheory UI clone or general multivariate/profile G-theory implementation. Incomplete source data do not authorize sparse future plans. Component SEs/intervals use balanced ANOVA and normal random effects for a score or fixed composite; two-score covariance intervals have explicit generalized and conservative choices with signed bounds. Explicit reml_profile component intervals also accept MINQUE(0), retaining original points and a separate REML center without supplying a sampling SE. Separately, vcov(g) supplies normal-model sampling covariance and SEs for original ANOVA/MINQUE components, including incomplete sources; product-moment correction retains negative or unidentified sampling variances and does not establish normal-reference intervals. Joint reml_profile covariance intervals also accept identifiable incomplete sources, with signed bounds, no sampling SE and unestablished finite-sample coverage. Balanced crossed G also accepts an explicit experimental score interval using extended nuisance parameters; empty numerical confidence sets and failed calculations have distinct diagnostics, with no Phi or paired-plan transfer. Explicit experimental reml_profile G/Phi and SEM intervals fit the used ratings under normal effects, preserving ANOVA/MINQUE points and reporting a separate REML IntervalCenter; finite-sample and boundary coverage are not guaranteed. With reference, the shared-profile transformation supplies G/Phi/SEM differences for one-facet plans or nested plans with a common future child count, including incomplete sources; the base approximation and numerical limitations remain. SEM intervals profile error variance before taking square roots and do not enumerate remote disconnected accepted regions. Plan-difference intervals have a separate normal-theory scope. Inspect negative components, metric-specific statuses and main-effects boundary/singularity flags.
#> 29 Co-observation connectedness does not establish full design rank; consult the retained identification audit. Response networks likewise do not establish a common scale. R-native igraph analysis and display rather than FACETS menu output.
#> 30 Does not write a FACETS command specification file.
#> 31 Export does not certify source readiness, cross-run identity, or invariance; uses R/CSV tables rather than FACETS fixed syntax.
#> 32 Import supports named CSV/TSV and validated ASCII fixed-width headings, preserves source coding and availability, and does not align Graphfile and element coordinates or reconstruct a calibration. Command-level graph options are not fully mirrored.
#> 33 Full FACETS report-file emulation is not implemented.
#> 34 Uses package-native residual columns rather than exact FACETS fixed-field residual syntax.
#> 35 Export and import have different units/formats. Fixed imports require validated ASCII all-field headings; use named delimited columns for other layouts or non-ASCII labels. Imported results are descriptive, without external-calibration scoring or inferred joint uncertainty. GPCM score-side equivalence is outside scope.
#> 36 Not a FACETS simulated-data file clone.
#> 37 The standalone subset writer exports connectivity review tables, not a full FACETS UI-compatible subset command file.
#> 38 Would require a separate Winsteps output contract.
#> 39 R-native plots replace FACETS graph menu output.
#> 40 Not labeled as FACETS ICC/IRF menu output.
#> 41 R-native plot data replace FACETS graph-menu output.
#> 42 R-native information definition and plot data.
#> 43 R-native plot data replace FACETS graph-menu output.
#> 44 FACETS conditional-probability menu semantics are not mirrored exactly.
#> 45 Parsing arbitrary FACETS command files is outside the package scope.
#> 46 Exact full report emulation is outside the package scope.
#> 47 General raw FACETS report parsing is not implemented.
#> 48 No current route converts a FACETS or another program's calibration file into an mfrmr portable artifact; native artifact scoring does not establish transportability or external-program equivalence.
#> 49 Element and group anchors do not make threshold ladders fixed or supply a general calibration-import schema.
#> 50 There is no per-observation ScaleId contract, scale-specific category map, or ragged scale-specific PCM threshold block.
#> 51 A category-probability vector that sums to one is not an unordered nominal-response or multinomial-logit model; category order enters every current likelihood.
#> 52 Grouped binomial trials, Poisson counts, negative-binomial counts, and other count likelihoods are not implemented; integer scores are interpreted as ordered category codes.
#> 53 It is not a general collapsed-person frequency table: for MML, powering responses inside one Person pattern is not equivalent to replicating a complete Person pattern after marginalization. It also does not create a count-response family, model within-cell dependence, or make non-unit-weight fits eligible for the common information-criterion panel.
#> 54 Residual PCA is exploratory dimensionality evidence, not native multidimensional estimation or dimension-specific scoring.
#> 55 GPCM support does not establish an unrestricted free-discrimination model family.
#> 56 FACETS Table 7 Estimated Discrimination is a post-fit diagnostic that does not update other Rasch estimates, so it is not the jointly estimated GPCM slope from mfrmr.
#> Alternative
#> 1 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 2 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 3 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 4 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 5 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 6 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 7 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 8 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 9 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 10 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 11 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 12 Use the closest documented mfrmr table or plot for analysis; use FACETS externally when the omitted statistic or format is required.
#> 13 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 14 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 15 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 16 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 17 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 18 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 19 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 20 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 21 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 22 Use the closest documented mfrmr table or plot for analysis; use FACETS externally when the omitted statistic or format is required.
#> 23 Build the display from package plot data or another R graphics system; use FACETS externally for its menu-specific view.
#> 24 Build the display from package plot data or another R graphics system; use FACETS externally for its menu-specific view.
#> 25 Use package-native R output where suitable, or run the relevant external program for its program-specific file, interface, or report.
#> 26 Build the display from package plot data or another R graphics system; use FACETS externally for its menu-specific view.
#> 27 Build the display from package plot data or another R graphics system; use FACETS externally for its menu-specific view.
#> 28 Start with the G-theory vignette and declare the score, random facets and assignments. Use the main-effects adapter only for its different decomposition; use FACETS externally for its UI or other external-only options.
#> 29 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 30 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 31 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 32 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 33 Use the closest documented mfrmr table or plot for analysis; use FACETS externally when the omitted statistic or format is required.
#> 34 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 35 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 36 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 37 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 38 Use package-native R output where suitable, or run the relevant external program for its program-specific file, interface, or report.
#> 39 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 40 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 41 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 42 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 43 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 44 Use the closest documented mfrmr table or plot for analysis; use FACETS externally when the omitted statistic or format is required.
#> 45 Use package-native R output where suitable, or run the relevant external program for its program-specific file, interface, or report.
#> 46 Use package-native R output where suitable, or run the relevant external program for its program-specific file, interface, or report.
#> 47 Use the closest documented mfrmr table or plot for analysis; use FACETS externally when the omitted statistic or format is required.
#> 48 Use the native mfrmr calibration workflow for eligible fits; use a separately validated external workflow when third-party calibration import is required.
#> 49 Use current element/group anchor routes only for their documented scope; retain threshold anchoring for a later release.
#> 50 Fit supported single-scale designs separately; retain multi-scale and mixed-family workflows for a later release.
#> 51 Use a nominal-response or multinomial-regression implementation externally when category order is not substantively defined.
#> 52 Use FACETS or another count-model implementation for an appropriate binomial-trial or Poisson estimand; do not relabel an ordered-category fit as a count model.
#> 53 Retain one distinguishable event per row when possible; preserve distinct Person response patterns, and report the exact likelihood-weight interpretation.
#> 54 Use exploratory dimensionality diagnostics and external multidimensional software when a multidimensional estimator is required.
#> 55 Use mfrmr_output_guide("gpcm_capabilities") and the documented GPCM route; retain unrestricted GPCM for a later release.
#> 56 Use FACETS for the matched PCM/JML comparison or as a deliberately misspecified equal-discrimination control; use a genuinely slope-estimating program only after the GPCM kernel and identification are matched.
facets_guide('coverage', "partial")
#> FACETSArea FACETSFeature
#> 3 Output table Table 3: main iteration report
#> 5 Output table Table 5: measurable data summary
#> 8 Output table Table 6.2: graphical facet statistics
#> 12 Output table Table 8.1: dichotomous/binomial/Poisson statistics
#> 14 Output table Table 8: scale-structure bar chart
#> 18 Output table Table 11: bias-calculation counts
#> 19 Output table Table 12: bias summary report
#> 22 Output table DIF/bias Excel plot
#> 23 R/Web plots Scatterplots and histograms from FACETS menus
#> 24 R/Web plots X-Y plot: R Statistics
#> 26 R/Web plots X-Y-Z plot: R Statistics
#> 27 R/Web plots Histogram: R Statistics
#> 30 Output file Specification settings file
#> 33 Output file Output report file
#> 35 Output file Score output file
#> 36 Output file Simulated data file
#> 37 Output file Subset group-anchor file
#> 44 Graph menu Conditional probability curves
#> 47 Specification/workflow Raw FACETS report-text import
#> FACETSReference
#> 3 table3.htm
#> 5 table5.htm
#> 8 table6_2.htm
#> 12 table8_1dichotomous.htm
#> 14 table8barchart.htm
#> 18 table11.htm
#> 19 table12.htm
#> 22 difbiasplot.htm
#> 23 outputtableindex.htm
#> 24 xyplotr.htm
#> 26 xyzplotr.htm
#> 27 histogramr.htm
#> 30 specificationfile.htm
#> 33 outputfile.htm
#> 35 scorefile.htm
#> 36 simulatedfile.htm
#> 37 subsetfile.htm
#> 44 graphs.htm
#> 47 outputtableindex.htm
#> mfrmrRoute
#> 3 estimation_iteration_report()
#> 5 measurable_summary_table(); describe_mfrm_data()
#> 8 facet_statistics_report(); plot(...)
#> 12 rating_scale_table() for two-category ordered scores
#> 14 category_structure_report()
#> 18 bias_count_table()
#> 19 summary(estimate_bias(...)); plot(x, type = "bias")
#> 22 plot(plot = ..., type = "bias")
#> 23 plot_data(); package plot helpers
#> 24 plot_data(); user-defined R plotting
#> 26 plot(fit, type = "ccc_surface"); plot_data()
#> 27 plot_data(); plot(fit, type = "wright"); plot(x, type = "qc")
#> 30 build_mfrm_manifest(); build_mfrm_replay_script()
#> 33 export_summary_appendix(); build_fixed_reports()
#> 35 facets_output_file_bundle(include = "score"); read_facets_results(); facets_fit_review(); plot()
#> 36 simulate_mfrm_data(); build_mfrm_sim_spec()
#> 37 write_mfrm_subset_file(); group_anchors; review_mfrm_anchors(); make_anchor_table()
#> 44 category_curves_report()
#> 47 read_facets_results() for delimited/fixed-field score extracts
#> Status SurfaceCoverage StatisticalContract
#> 3 partial partial partial_package_native_contract
#> 5 partial partial partial_package_native_contract
#> 8 partial partial partial_package_native_contract
#> 12 partial partial partial_package_native_contract
#> 14 partial partial partial_package_native_contract
#> 18 partial partial partial_package_native_contract
#> 19 partial partial partial_package_native_contract
#> 22 partial partial partial_package_native_contract
#> 23 partial partial partial_package_native_contract
#> 24 partial partial partial_package_native_contract
#> 26 partial partial partial_package_native_contract
#> 27 partial partial partial_package_native_contract
#> 30 partial partial partial_package_native_contract
#> 33 partial partial partial_package_native_contract
#> 35 partial partial partial_package_native_contract
#> 36 partial partial partial_package_native_contract
#> 37 partial partial partial_package_native_contract
#> 44 partial partial partial_package_native_contract
#> 47 partial partial partial_package_native_contract
#> ValidationEvidence OperationalStatus
#> 3 external_match_not_established partial_workflow_only
#> 5 external_match_not_established partial_workflow_only
#> 8 external_match_not_established partial_workflow_only
#> 12 external_match_not_established partial_workflow_only
#> 14 external_match_not_established partial_workflow_only
#> 18 external_match_not_established partial_workflow_only
#> 19 external_match_not_established partial_workflow_only
#> 22 external_match_not_established partial_workflow_only
#> 23 external_match_not_established partial_workflow_only
#> 24 external_match_not_established partial_workflow_only
#> 26 external_match_not_established partial_workflow_only
#> 27 external_match_not_established partial_workflow_only
#> 30 external_match_not_established partial_workflow_only
#> 33 external_match_not_established partial_workflow_only
#> 35 external_match_not_established partial_workflow_only
#> 36 external_match_not_established partial_workflow_only
#> 37 external_match_not_established partial_workflow_only
#> 44 external_match_not_established partial_workflow_only
#> 47 external_match_not_established partial_workflow_only
#> Capability
#> 3 Convergence and replayed iteration evidence.
#> 5 Facet coverage, category counts, and subset/connectivity checks.
#> 8 Facet statistics and visual summaries.
#> 12 Two-category Rasch-category summaries are available.
#> 14 Category structure and transition summaries.
#> 18 Cell counts, logit shifts and separate baseline/remaining score residuals.
#> 19 Distributional and visual bias summaries are available.
#> 22 R-native scatter, heatmap, and facet-profile bias displays.
#> 23 Reusable plot data supports custom R graphics.
#> 24 Users can build X-Y plots from returned data frames.
#> 26 Selected 3D/surface-ready plot data are available.
#> 27 Several package outputs include histogram-like summaries.
#> 30 R-native reproducibility manifest and replay script.
#> 33 Structured appendix/report artifacts.
#> 35 Native observation-level score export and separate FACETS element-level import, including standalone source-measure and fit plots.
#> 36 Simulation data and explicit simulation specifications.
#> 37 Connected-subset summary/node files and group-anchor inputs/checks are available.
#> 44 Category probability curves conditional on theta are available.
#> 47 Fit/score table import is supported.
#> Limitation
#> 3 Does not reproduce the complete FACETS iteration trace.
#> 5 Does not reproduce FACETS response moments, raw/residual variance and global Pearson summaries; this is substantive partial coverage.
#> 8 FACETS M/S/Q/X printer-graph formatting is not reproduced exactly.
#> 12 FACETS binomial-trial and Poisson-specific reports are not implemented.
#> 14 FACETS line-printer artwork is not reproduced exactly.
#> 18 FACETS Table 11 response moments and variance decompositions are not supplied.
#> 19 FACETS vertical frequency bar-chart is not reproduced exactly.
#> 22 Excel-specific output is not implemented.
#> 23 FACETS arbitrary R/Web plotting menus are not mirrored.
#> 24 No dedicated FACETS-style arbitrary X-Y plot wrapper.
#> 26 No arbitrary FACETS X-Y-Z plot wrapper.
#> 27 No general FACETS histogram menu clone.
#> 30 Does not write a FACETS command specification file.
#> 33 Full FACETS report-file emulation is not implemented.
#> 35 Export and import have different units/formats. Fixed imports require validated ASCII all-field headings; use named delimited columns for other layouts or non-ASCII labels. Imported results are descriptive, without external-calibration scoring or inferred joint uncertainty. GPCM score-side equivalence is outside scope.
#> 36 Not a FACETS simulated-data file clone.
#> 37 The standalone subset writer exports connectivity review tables, not a full FACETS UI-compatible subset command file.
#> 44 FACETS conditional-probability menu semantics are not mirrored exactly.
#> 47 General raw FACETS report parsing is not implemented.
#> Alternative
#> 3 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 5 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 8 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 12 Use the closest documented mfrmr table or plot for analysis; use FACETS externally when the omitted statistic or format is required.
#> 14 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 18 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 19 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 22 Use the closest documented mfrmr table or plot for analysis; use FACETS externally when the omitted statistic or format is required.
#> 23 Build the display from package plot data or another R graphics system; use FACETS externally for its menu-specific view.
#> 24 Build the display from package plot data or another R graphics system; use FACETS externally for its menu-specific view.
#> 26 Build the display from package plot data or another R graphics system; use FACETS externally for its menu-specific view.
#> 27 Build the display from package plot data or another R graphics system; use FACETS externally for its menu-specific view.
#> 30 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 33 Use the closest documented mfrmr table or plot for analysis; use FACETS externally when the omitted statistic or format is required.
#> 35 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 36 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 37 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 44 Use the closest documented mfrmr table or plot for analysis; use FACETS externally when the omitted statistic or format is required.
#> 47 Use the closest documented mfrmr table or plot for analysis; use FACETS externally when the omitted statistic or format is required.
facets_guide('coverage', "supported_with_caveat")
#> FACETSArea
#> 7 Output table
#> 9 Output table
#> 10 Output table
#> 11 Output table
#> 13 Output table
#> 28 R/Web plots
#> 53 Observation weighting
#> FACETSFeature
#> 7 Table 6.0.0: disjoint element listing
#> 9 Table 7: facet measurement report
#> 10 Table 7: reliability and chi-square
#> 11 Table 7: agreement statistics
#> 13 Table 8.1: polytomous rating-scale/partial-credit statistics
#> 28 Generalizability Theory via R package gtheory
#> 53 Row-frequency weights for ordered ratings
#> FACETSReference
#> 7 table6_0_0.htm
#> 9 table7.htm
#> 10 table7summarystatistics.htm
#> 11 table7agreementstatistics.htm
#> 13 table8_1ratingscale.htm
#> 28 gtheory.htm
#> 53 current mfrmr weight contract
#> mfrmrRoute
#> 7 subset_connectivity_report()
#> 9 fit_measures_table(); diagnose_mfrm(); summary(fit)
#> 10 facets_chisq_table(); diagnose_mfrm()$reliability
#> 11 interrater_agreement_table(); rater_network_analysis(); rater_halo_network_analysis(); plot(x, type = "agreement")
#> 13 rating_scale_table(); category_structure_report()
#> 28 mfrm_g_study(); mfrm_d_study(); mfrm_multivariate_d_compare(); mfrm_generalizability(); mfrm_d_study()
#> 53 fit_mfrm(weight = ...)
#> Status SurfaceCoverage StatisticalContract
#> 7 supported_with_caveat available_with_caveat package_native_caveated
#> 9 supported_with_caveat available_with_caveat package_native_caveated
#> 10 supported_with_caveat available_with_caveat package_native_caveated
#> 11 supported_with_caveat available_with_caveat package_native_caveated
#> 13 supported_with_caveat available_with_caveat package_native_caveated
#> 28 supported_with_caveat available_with_caveat package_native_caveated
#> 53 supported_with_caveat available_with_caveat package_native_caveated
#> ValidationEvidence OperationalStatus
#> 7 external_match_not_established available_with_mandatory_caveat
#> 9 external_match_not_established available_with_mandatory_caveat
#> 10 external_match_not_established available_with_mandatory_caveat
#> 11 external_match_not_established available_with_mandatory_caveat
#> 13 external_match_not_established available_with_mandatory_caveat
#> 28 external_match_not_established available_with_mandatory_caveat
#> 53 external_match_not_established available_with_mandatory_caveat
#> Capability
#> 7 Co-observation graph components and facet coverage, with the fitted identification audit.
#> 9 Measures, SEs, fit, anchoring status, and review flags.
#> 10 Rasch/FACETS-style separation, reliability, and chi-square summaries.
#> 11 Raw-category observed agreement and model-probability expected agreement, plus pairwise rater-network, rater-by-criterion halo network, and rater-agreement views.
#> 13 Core category diagnostics and thresholds for one RSM scale or one PCM step facet on the package's shared observed score scale.
#> 28 Observed-score univariate/multivariate component models for supported crossed/nested designs, fixed score composites and D-study plans, with component/covariance plots, curves and paired normal-theory plan comparisons. The separate main-effects route retains residual-scaling sensitivity and IdentificationStatus columns.
#> 53 A positive numeric observation weight multiplies that row's conditional ordered-category likelihood contribution and can represent a defensible row-replication weight.
#> Limitation
#> 7 These are not FACETS identification subsets; a single component can conceal confounded panels.
#> 9 Estimator and person-score basis, extreme handling, df/ZSTD conventions, and FACETS options can differ; external numerical equivalence is not established.
#> 10 Uses package-native structured output and documented df conventions; external numerical equivalence is not established.
#> 11 The package does not translate category positions across multiple independent scales, apply FACETS agreement-based SE inflation, or establish external Table 7 parity.
#> 13 Multiple independent scales and scale-specific anchors/starting values are not reproduced. Cumulative boundaries are grid approximations; category residual fit is descriptive, with no unit-baseline ZSTD test.
#> 28 Not a FACETS/gtheory UI clone or general multivariate/profile G-theory implementation. Incomplete source data do not authorize sparse future plans. Component SEs/intervals use balanced ANOVA and normal random effects for a score or fixed composite; two-score covariance intervals have explicit generalized and conservative choices with signed bounds. Explicit reml_profile component intervals also accept MINQUE(0), retaining original points and a separate REML center without supplying a sampling SE. Separately, vcov(g) supplies normal-model sampling covariance and SEs for original ANOVA/MINQUE components, including incomplete sources; product-moment correction retains negative or unidentified sampling variances and does not establish normal-reference intervals. Joint reml_profile covariance intervals also accept identifiable incomplete sources, with signed bounds, no sampling SE and unestablished finite-sample coverage. Balanced crossed G also accepts an explicit experimental score interval using extended nuisance parameters; empty numerical confidence sets and failed calculations have distinct diagnostics, with no Phi or paired-plan transfer. Explicit experimental reml_profile G/Phi and SEM intervals fit the used ratings under normal effects, preserving ANOVA/MINQUE points and reporting a separate REML IntervalCenter; finite-sample and boundary coverage are not guaranteed. With reference, the shared-profile transformation supplies G/Phi/SEM differences for one-facet plans or nested plans with a common future child count, including incomplete sources; the base approximation and numerical limitations remain. SEM intervals profile error variance before taking square roots and do not enumerate remote disconnected accepted regions. Plan-difference intervals have a separate normal-theory scope. Inspect negative components, metric-specific statuses and main-effects boundary/singularity flags.
#> 53 It is not a general collapsed-person frequency table: for MML, powering responses inside one Person pattern is not equivalent to replicating a complete Person pattern after marginalization. It also does not create a count-response family, model within-cell dependence, or make non-unit-weight fits eligible for the common information-criterion panel.
#> Alternative
#> 7 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 9 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 10 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 11 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 13 Use the documented mfrmr route; use FACETS externally only when its exact layout or option surface is required.
#> 28 Start with the G-theory vignette and declare the score, random facets and assignments. Use the main-effects adapter only for its different decomposition; use FACETS externally for its UI or other external-only options.
#> 53 Retain one distinguishable event per row when possible; preserve distinct Person response patterns, and report the exact likelihood-weight interpretation.
facets_guide('coverage', "not_implemented")
#> FACETSArea
#> 38 Output file
#> 48 Current scope boundary
#> 49 Current scope boundary
#> 50 Current scope boundary
#> 51 Current scope boundary
#> 52 Current scope boundary
#> 54 Current scope boundary
#> 55 Current scope boundary
#> 56 Current scope boundary
#> FACETSFeature
#> 38 Winsteps control and data file
#> 48 FACETS or third-party frozen-calibration import
#> 49 General threshold or step anchors and starting-value import
#> 50 Multiple observed scales and scale-specific PCM
#> 51 Nominal/multinomial response models
#> 52 Binomial-trial and Poisson/count response models
#> 54 Native multidimensional estimation and dimension-specific scores
#> 55 Unrestricted GPCM
#> 56 FACETS free-slope polytomous GPCM comparison
#> FACETSReference
#> 38 winstepsfile.htm
#> 48 current mfrmr public contract
#> 49 current mfrmr public contract
#> 50 current mfrmr public contract
#> 51 current mfrmr public contract
#> 52 models.htm
#> 54 current mfrmr public contract
#> 55 current mfrmr public contract
#> 56 t7menu.htm
#> mfrmrRoute
#> 38 none
#> 48 no external import; native route: mfrmr_output_guide("calibration_capabilities"); load_mfrm_calibration(); score_mfrm_calibration()
#> 49 not available in the current package
#> 50 not available in the current package
#> 51 not available in the current package
#> 52 not available in the current package
#> 54 not available in the current package
#> 55 not available for unrestricted GPCM
#> 56 none as a direct common-estimand comparison
#> Status SurfaceCoverage StatisticalContract ValidationEvidence
#> 38 not_implemented unavailable not_available not_applicable
#> 48 not_implemented unavailable not_available not_applicable
#> 49 not_implemented unavailable not_available not_applicable
#> 50 not_implemented unavailable not_available not_applicable
#> 51 not_implemented unavailable not_available not_applicable
#> 52 not_implemented unavailable not_available not_applicable
#> 54 not_implemented unavailable not_available not_applicable
#> 55 not_implemented unavailable not_available not_applicable
#> 56 not_implemented unavailable not_available not_applicable
#> OperationalStatus
#> 38 blocked
#> 48 blocked
#> 49 blocked
#> 50 blocked
#> 51 blocked
#> 52 blocked
#> 54 blocked
#> 55 blocked
#> 56 blocked
#> Capability
#> 38 No Winsteps control/data export route.
#> 48 Native mfrmr portable artifacts support fixed-standard-normal RSM/PCM MML, qualified estimated-normal GPCM MML and qualified reference-prior RSM/PCM or shared-owner GPCM JML; consult mfrmr_output_guide("calibration_capabilities") for scope and source checks.
#> 49 No current public route accepts general threshold or step anchors or a threshold starting-value contract.
#> 50 Each fit uses one observed score scale and one homogeneous response-model family.
#> 51 The current RSM, PCM, and GPCM routes model ordered category probabilities only.
#> 52 Binary ordered scores are available as the two-category special case of the current ordered-response kernel.
#> 54 The current public estimator and score routes are unidimensional.
#> 55 JML requires slope_facet to equal step_facet; MML permits separate owners with one slope family.
#> 56 FACETS PCM/JMLE can serve as the direct equal-discrimination comparison after the full estimation contract is aligned.
#> Limitation
#> 38 Would require a separate Winsteps output contract.
#> 48 No current route converts a FACETS or another program's calibration file into an mfrmr portable artifact; native artifact scoring does not establish transportability or external-program equivalence.
#> 49 Element and group anchors do not make threshold ladders fixed or supply a general calibration-import schema.
#> 50 There is no per-observation ScaleId contract, scale-specific category map, or ragged scale-specific PCM threshold block.
#> 51 A category-probability vector that sums to one is not an unordered nominal-response or multinomial-logit model; category order enters every current likelihood.
#> 52 Grouped binomial trials, Poisson counts, negative-binomial counts, and other count likelihoods are not implemented; integer scores are interpreted as ordered category codes.
#> 54 Residual PCA is exploratory dimensionality evidence, not native multidimensional estimation or dimension-specific scoring.
#> 55 GPCM support does not establish an unrestricted free-discrimination model family.
#> 56 FACETS Table 7 Estimated Discrimination is a post-fit diagnostic that does not update other Rasch estimates, so it is not the jointly estimated GPCM slope from mfrmr.
#> Alternative
#> 38 Use package-native R output where suitable, or run the relevant external program for its program-specific file, interface, or report.
#> 48 Use the native mfrmr calibration workflow for eligible fits; use a separately validated external workflow when third-party calibration import is required.
#> 49 Use current element/group anchor routes only for their documented scope; retain threshold anchoring for a later release.
#> 50 Fit supported single-scale designs separately; retain multi-scale and mixed-family workflows for a later release.
#> 51 Use a nominal-response or multinomial-regression implementation externally when category order is not substantively defined.
#> 52 Use FACETS or another count-model implementation for an appropriate binomial-trial or Poisson estimand; do not relabel an ordered-category fit as a count model.
#> 54 Use exploratory dimensionality diagnostics and external multidimensional software when a multidimensional estimator is required.
#> 55 Use mfrmr_output_guide("gpcm_capabilities") and the documented GPCM route; retain unrestricted GPCM for a later release.
#> 56 Use FACETS for the matched PCM/JML comparison or as a deliberately misspecified equal-discrimination control; use a genuinely slope-estimating program only after the GPCM kernel and identification are matched.