FACETS_manual_mapping.md
In mfrmr: Estimation and Diagnostics for Many-Facet Measurement Models

FACETS Manual Mapping

This package mirrors FACETS-style workflows while keeping a native R implementation.

Output table index: https://www.winsteps.com/facetman/outputtableindex.htm
Table 5 measurable data summary: https://www.winsteps.com/facetman/table5.htm
Table 7 measurement report: https://www.winsteps.com/facetman/table7.htm
Table 7 reliability/chi-square notes: https://www.winsteps.com/facetman/table7summarystatistics.htm
Table 7 agreement notes: https://www.winsteps.com/facetman/table7agreementstatistics.htm
Table 8.1 rating scale report: https://www.winsteps.com/facetman/table8_1ratingscale.htm
Table 8 bar-chart report: https://www.winsteps.com/facetman/table8barchart.htm
Table 8 probability-curves report: https://www.winsteps.com/facetman/table8curves.htm
Graph output (Graphfile=): https://www.winsteps.com/facetman/graphoutputfile.htm
Table 9 bias-iteration report: https://www.winsteps.com/facetman/table9.htm
Table 10 unexpected-after-bias report: https://www.winsteps.com/facetman/table10.htm
Table 11 bias-calculation counts report: https://www.winsteps.com/facetman/table11.htm
Table 12 DIF/bias summary report: https://www.winsteps.com/facetman/table12.htm
Table 13 DIF/bias detail report: https://www.winsteps.com/facetman/table13.htm
Table 14 pairwise bias report: https://www.winsteps.com/facetman/table14.htm

Core multifacet estimation (RSM/PCM, MML/JML): fit_mfrm() / mfrm_estimate()
Diagnostics core bundle (obs, fit, reliability, interactions, subsets): diagnose_mfrm()
Table 1-style specification summary: specifications_report()
Table 2-style data summary report: data_quality_report()
Table 3-style iteration report (replayed): estimation_iteration_report()
Table 4-style unexpected responses: unexpected_response_table(), plot_unexpected()
Table 5-style measurable summary bundle: measurable_summary_table()
Table 6.0.0-style subset/disjoint listing: subset_connectivity_report()
Table 6.2-style facet-statistics graphic summary: facet_statistics_report()
Table 7-style facet/person measures and fit summary: diagnose_mfrm() + summary.mfrm_fit()
Table 7 reliability + facet chi-square style summaries: diagnose_mfrm()$reliability, diagnose_mfrm()$facets_chisq, facets_chisq_table(), plot_facets_chisq()
Table 7 agreement style summaries: diagnose_mfrm()$interrater, interrater_agreement_table(), plot_interrater_agreement()
Table 8.1-style rating scale bundle: rating_scale_table()
Table 8-style bar-chart and curves exporters: category_structure_report(), category_curves_report()
Output-file emulation (GRAPH= / SCORE= style): facets_output_file_bundle()
Table 12 fair-average style output bundle: fair_average_table(), plot_fair_average()
Displacement diagnostics (FACETS-style anchor drift check): displacement_table(), plot_displacement()
Bias re-estimation iteration (Table 9 workflow): estimate_bias() / estimate_bias_interaction()
Table 10-style unexpected-after-bias output: unexpected_after_bias_table()
Table 11-style bias-count report: bias_count_table()
Table 12/13/14 style bias outputs:
summary/detail: estimate_bias() result (summary, table, chi_sq)
higher-order mode (3+ facets): estimate_bias(..., interaction_facets = c(...))
pairwise: build_fixed_reports() (pairwise_table, pairwise_fixed) for 2-way runs
Table 13 plot export: bias_interaction_report(), plot_bias_interaction() (including facet_profile mode)
Fixed-width report generation: build_fixed_reports()
APA narrative/table helpers: build_apa_outputs(), apa_table()
Residual PCA checks (overall + by facet): analyze_residual_pca(), plot_residual_pca()
QC dashboard (base graphics): plot_qc_dashboard()
Anchor workflow: audit_mfrm_anchors(), make_anchor_table()
Data packaging/loading helpers: list_mfrmr_data(), load_mfrmr_data()
Automated FACETS parity checks (columns + core metrics): facets_parity_report(), tests/testthat/test-facets-column-contract.R, tests/testthat/test-facets-metric-contract.R, inst/references/facets_column_contract.csv

Design policy:
structured tables and visualization APIs are primary deliverables
fixed-width / line-printer text is optional and secondary (audit/log use)
exact FACETS line-printer emulation is intentionally out of scope
legacy numbered table* names are internal and not exported
Table 1/2/3 reports:
current: specifications_report(), data_quality_report(), estimation_iteration_report() with structured output and optional fixed-width text
gap: FACETS fixed-width text layout and exact optimizer-internal iteration path are not yet 1:1
Table 5 measurable data summary:
current: measurable_summary_table() and describe_mfrm_data() (including observed inter-rater agreement bundle)
gap: FACETS column-by-column textual layout parity is not exact
Table 8.1 rating-scale report:
current: rating_scale_table() plus CCC/pathway visualization (plot.mfrm_fit, QC category panel)
gap: FACETS text layout and all legacy columns/order are not yet 1:1
Table 8 bar-chart / probability-curves exporters:
current: category_structure_report() and category_curves_report() including Graphfile-style wide output and optional fixed-width text mirrors
gap: exact FACETS line-printer artwork/fixed-column parity is intentionally not targeted
Table 6.2 graphical facet-statistics report:
current: facet_statistics_report() with fixed-width M/S/Q/X rulers
gap: FACETS native table layout and printer-graph formatting are not yet 1:1
Output-file emulation (GRAPH= / SCORE=):
current: facets_output_file_bundle() with graph coordinates, observation-level modeled score export, optional fixed-width mirrors, and optional file writing
gap: FACETS command-level options and fixed-column file-writing parity are not yet 1:1
Table 14 pairwise contrast report:
current: available for 2-way bias runs via build_fixed_reports()
gap: FACETS native layout and options are broader; higher-order runs intentionally omit pairwise section

No high-priority items in the current Table 5-14 scope; remaining gaps are mostly formatting/options parity.

Direct anchors (Facet, Level, Anchor) are fixed.
Group anchors (Facet, Level, Group, GroupValue) constrain group means.
If a level appears in both tables, direct anchor takes precedence.
Missing GroupValue is treated as 0.
Default recommendation thresholds:
common anchors per linking facet: >= 5
observations per element: >= 30
observations per score category: >= 10

Core estimation and diagnostics: ready
New inter-rater and facet-chi-square APIs: implemented and tested
Remaining work for closer FACETS parity is mostly report-format completeness, not model-estimation correctness