Build an MFRM network review

Usage

build_mfrm_network_review(
  fit = NULL,
  diagnostics = NULL,
  sparse_design = NULL,
  peer_review_design = NULL,
  top_n_subsets = NULL,
  min_observations = 0,
  top_n = 10,
  include_graph = FALSE,
  data = NULL,
  person = NULL,
  facets = NULL,
  score = NULL,
  basis = c("observed", "planned"),
  weight = NULL
)

# S3 method for class 'mfrm_network_review'
summary(object, digits = 3, top_n = 10, ...)

Arguments

fit

Output from fit_mfrm().

diagnostics

Optional output from diagnose_mfrm().

sparse_design

Optional sparse-design metadata. Supply either the generated data frame that carries the mfrm_sparse_design attribute, the attribute itself, or a data frame with sparse design columns such as SparseDesignActive, DesignDensity, MinCommonPersonsPerRaterPair, and ZeroCommonRaterPairs.

peer_review_design

Optional peer-review design metadata. Supply either the generated data frame that carries the mfrm_peer_review_design attribute, the attribute itself, or its overview data frame.

top_n_subsets

Optional maximum number of connected-subset rows to retain before constructing the graph; passed to mfrm_network_analysis().

min_observations

Minimum observations required to keep a subset row; passed to mfrm_network_analysis().

top_n

For build_mfrm_network_review(): Number of central/cut/bridge rows to retain in the review.

For summary(): Number of central/cut/bridge rows to keep in the compact summary.

include_graph

Logical; if TRUE, keep the underlying igraph object in the nested source_network bundle.

data

Optional raw assignment/rating data, supplied instead of fit.

person, facets

Names of the person column and one or more facet columns in data. Include only Rater in facets to inspect person-rater links without a common Criterion node obscuring separate panels.

score

Numeric score column for basis = "observed". Recode study-specific missing markers to NA first; no category recoding or estimation is done.

basis

"observed" includes rows with complete identifiers and nonmissing finite scores; "planned" includes complete assignment rows without scores. Planned networks require score = NULL and weight = NULL.

weight

Optional numeric weight column for observed raw data. Missing and zero-weight rows do not create edges. Edge weights still count rows, not sampling weights. Negative or infinite weights are refused.

object

Output from build_mfrm_network_review().

digits

Number of digits for printed numeric values.

...

Reserved for generic compatibility.

Value

A bundle of class mfrm_network_review containing:

overview: connectedness and front-door review status
network_summary: graph-level metrics from mfrm_network_analysis()
facet_summary: facet-level vulnerability summaries
top_central_nodes, top_cut_nodes, top_bridge_edges: follow-up rows
sparse_review: optional sparse-design linking review
peer_review: optional peer-review assignment and linkage diagnostics
reporting_map: boundary between MFRM, design network, sparse design, peer-review design, and rater-effect network routes

For summary(): An object of class summary.mfrm_network_review.

Details

build_mfrm_network_review() is a synthesis layer over mfrm_network_analysis(). It keeps the measurement model and graph view in separate sections: MFRM estimates remain the measurement results, while the network review summarizes co-observation connectedness and linking vulnerability in the observed design. This is especially useful for sparse or incomplete rater-mediated designs, where common-person links, connected subsets, articulation points, and bridge edges can explain why an otherwise estimable model depends on fragile design links.

The review status is deliberately conservative and descriptive. It is not a literature-derived adequacy cut point for fit, separation, recovery, or rater quality. Use it to decide which design links, anchors, or additional observations need inspection before making common-scale claims.

summary() method: Summarize an MFRM network review

References

Wind, S. A., & Jones, E. (2018). The stabilizing influences of linking set size and model-data fit in sparse rater-mediated assessment networks. Educational and Psychological Measurement. doi:10.1177/0013164417703733.
Wind, S. A., Jones, E., & Grajeda, S. (2023). Does sparseness matter? Examining the use of generalizability theory and many-facet Rasch measurement in sparse rating designs. Applied Psychological Measurement, 47(5-6), 351-364. doi:10.1177/01466216231182148.
DeMars, C. E., Shapovalov, Y. A., & Hathcoat, J. D. (2023). Many-Facet Rasch Designs: How Should Raters be Assigned to Examinees? NCME presentation.

Before fitting

Supply data, person, facets and score instead of fit. Known levels from incomplete/unscored rows remain as nodes, including isolates. A planned assignment roster can be examined with basis = "planned"; absent assignments cannot be reconstructed from an observed-only file. row_usage records each input row's inclusion and first exclusion reason. With a fit, only its retained rows are available here; review original missing assignments before fitting. No prefit result supplies an identification audit or model-readiness decision. settings$is_disjoint describes the source graph, even when subset filters omit components; compare source/retained subset counts. Filters do not establish linking. See online networks and reporting guide for a panel example.

Examples

# Supplementary workflow: run explicitly to keep routine checks brief.
# example("build_mfrm_network_review", run.dontrun = TRUE)
if (FALSE) { # \dontrun{
# Saved synthetic JML/RSM analysis of example_core.
# For your own data, fit a model first; see ?fit_mfrm.
example_fit <- readRDS(system.file("examples", "core-jml.rds", package = "mfrmr"))
fit <- example_fit$fit
if (requireNamespace("igraph", quietly = TRUE)) {
  review <- build_mfrm_network_review(fit)
  summary(review)
  build_summary_table_bundle(review)
}
} # }