Plot report/table bundles with ggplot2

Usage

# S3 method for class 'mfrm_bundle'
plot(x, y = NULL, type = NULL, ...)

Arguments

x: A bundle object returned by mfrmr table/report helpers.
y: Reserved for generic compatibility.
type: Optional plot type. Available values depend on bundle class.
...: Additional arguments forwarded to class-specific plotters.

Value

A ggplot object (invisibly), carrying the underlying mfrm_plot_data payload accessible via plot_data().

Details

plot() dispatches by bundle class:

mfrm_unexpected -> plot(x, type = "unexpected")
mfrm_fair_average -> plot(x, type = "fair_average")
mfrm_displacement -> plot(x, type = "displacement")
mfrm_interrater -> plot(x, type = "agreement")
mfrm_facets_chisq -> plot(x, type = "facets_chisq")
mfrm_bias_interaction -> plot(x, type = "bias")
mfrm_bias_count -> bias-count plots (cell counts / low-count rates)
mfrm_fixed_reports -> pairwise-contrast diagnostics
mfrm_visual_summaries -> warning/summary message count plots
mfrm_category_structure -> category count and boundary plots
mfrm_category_curves -> overview (default), ogive, CCC / category probability / conditional probability, cumulative, total-information, and category-specific-information plots
mfrm_rating_scale -> category-counts/threshold plots
mfrm_measurable -> measurable-data coverage/count plots
mfrm_unexpected_after_bias -> post-bias unexpected-response plots
mfrm_output_bundle -> graph/score output-file diagnostics, including type = "score_se" when scorefile SE columns are available
mfrm_residual_pca -> residual PCA scree, parallel-analysis, or loadings views via plot(x, type = "residual_pca")
mfrm_specifications -> facet/anchor/convergence plots
mfrm_data_quality -> dashboard, quality-flag, score-map, facet-pattern, and row/category/missing-row plots
mfrm_facets_fit_review -> FACETS-style df-sensitivity plot
mfrm_fit_measures -> fit-status counts, Infit/Outfit scatter, measure normal bands, and FACETS-style df-sensitivity plots. For type = "measure_ci", the caption retains the source interval interpretation (including exploratory JML bands); fixed values use open diamonds and finite estimates without intervals use crosses. main = "" omits the title and show_notes = FALSE hides the caption without removing it from saved data. Changing ci_level does not change inferential eligibility. See fit_measures_table().
mfrm_iteration_report -> replayed-iteration trajectories
mfrm_subset_connectivity -> subset-observation/connectivity plots
mfrm_facet_statistics -> facet statistic profile plots
mfrm_export_bundle / mfrm_summary_appendix_export -> export handoff plots (formats, artifact_groups, selection_tables, selection_handoff, selection_handoff_bundles, selection_handoff_roles, selection_handoff_role_sections, selection_bundles, selection_roles, selection_sections)

If a class is outside these families, use dedicated plotting helpers or custom ggplot2 graphics on component tables.

For mfrm_category_curves, pass preset = "monochrome" for grayscale/line-type output. Cumulative .5 boundary lines are shown only for interpretable in-range boundaries by default; use boundary_status = "all" to show every finite boundary estimate or boundary_status = "none" / show_cumulative_boundaries = FALSE to suppress those vertical boundary lines. Use plot_data(x, component = "plot_long") on a category-curve bundle when you want one ggplot2/plotly-friendly table across all curve families. Keep curve families and groups separate in custom figures. All category-curve, category-structure, rating-scale, and network bundle views render natively via ggplot2. Step displays use point estimates and do not turn stored step SEs into crossing, gap or curve intervals.

Interpreting output

The returned object is a ggplot object whose underlying plot payload and notes remain accessible via plot_data() and plot_data_components(); set draw = TRUE for immediate display.

Typical workflow

Create bundle output (e.g., unexpected_response_table()).
Inspect routing with summary(bundle) if needed.
Call p <- plot(bundle, type = ..., draw = FALSE) and inspect plot_data(p) when needed.

Session plot defaults

See mfrmr_visual_diagnostics for options(mfrmr.plot_preset), argument precedence, and appearance when saving or reusing a plot.

Network display controls

For subset, design, rater and halo bundles with type = "network", use layout (a two-column coordinate matrix, optionally row-named by all node IDs, or a function of the displayed igraph), display_min_weight (absolute edge weight for the picture only), and graphical (a named list of plot.igraph arguments, such as vertex.size, vertex.label, vertex.color, edge.curved). Unnamed vectors and coordinates follow the returned nodes order. All nodes, including isolates, remain. Edge vectors in graphical follow the displayed edges order, after any display filter. The default is a deterministic circle; positions do not estimate anything. Labels are hidden by default above 80 nodes; set vertex.label explicitly to override. Supply the same named layout to compare views. Constructor filters such as min_weight change calculated metrics; display_min_weight does not. Returned source_edges retain the calculated graph, edges the displayed graph, and coordinates the actual positions. Titles are metadata by default; add a title explicitly via graphical. See online networks and reporting guide for bipartite assignment, directional disagreement and rater-by-criterion examples inspired by Lamprianou. With rater_network_analysis() and weight_metric = "DirectionalMagnitude", type = "fit", metric = "SWS" (or "PUS") shows the published network inconsistency index. The returned table retains unavailable rows/statuses; there are no calibrated reference cutoffs. An entirely unavailable index draws an unavailable message and retains the original rows/statuses in the returned table, rather than substituting zero. For signed alpha, type = "matrix" returns all eligible signed pair coefficients before graph filtering, retaining zero and undefined values separately. Network drawings use magnitude for edge width and dashed lines for negative agreement. Use plot.mfrm_network_ergm() for ERGM coefficient intervals and simulated GOF.

Examples

# Long-running workflow; run explicitly when needed.
# example("plot.mfrm_bundle", run.dontrun = TRUE)
if (FALSE) { # \dontrun{
toy_full <- load_mfrmr_data("example_core")
toy_people <- unique(toy_full$Person)[1:12]
toy <- toy_full[toy_full$Person %in% toy_people, , drop = FALSE]
fit <- suppressWarnings(
  fit_mfrm(toy, "Person", c("Rater", "Criterion"), "Score", method = "JML", maxit = 30)
)
t4 <- unexpected_response_table(fit, abs_z_min = 1.5, prob_max = 0.4, top_n = 5)
p <- plot(t4, draw = FALSE)
vis <- build_visual_summaries(fit, diagnose_mfrm(fit, residual_pca = "none"))
p_vis <- plot(vis, type = "comparison", draw = FALSE)
spec <- specifications_report(fit)
p_spec <- plot(spec, type = "facet_elements", draw = FALSE)
if (interactive()) {
  plot(
    t4,
    type = "severity",
    draw = TRUE,
    main = "Unexpected Response Severity (Customized)",
    palette = c(higher = "#d95f02", lower = "#1b9e77", bar = "#2b8cbe"),
    label_angle = 45
  )
  plot(
    vis,
    type = "comparison",
    draw = TRUE,
    main = "Warning vs Summary Counts (Customized)",
    palette = c(warning = "#cb181d", summary = "#3182bd"),
    label_angle = 45
  )
}
} # }