Build a bias-interaction plot-data bundle (FACETS Table 13: ranked bias list)
Source:R/api-reports.R
bias_interaction_report.RdBundles the ranked flagged-cells view of a bias-interaction run for downstream printing and plotting. The three sibling reports in this family are intentionally distinct:
bias_interaction_report()(this one) = FACETS Table 13: a ranked list of interaction cells witht,bias size, and screening tail area – use when reviewing which(facet_a, facet_b)cells deserve follow-up.bias_iteration_report()= iteration history / convergence trace for the bias recalibration (FACETS Table 9 territory) – use when diagnosing whether the bias run itself stabilised.bias_pairwise_report()= pairwise contrast table for a target facet (FACETS Table 14 territory) – use when comparing levels within a facet while controlling for the other.
Usage
bias_interaction_report(
x,
diagnostics = NULL,
facet_a = NULL,
facet_b = NULL,
interaction_facets = NULL,
max_abs = 10,
omit_extreme = TRUE,
max_iter = 4,
tol = 0.001,
top_n = 50,
abs_t_warn = 2,
abs_bias_warn = 0.5,
p_max = 0.05,
sort_by = c("abs_t", "abs_bias", "prob")
)Arguments
- x
Output from
estimate_bias()orfit_mfrm().- diagnostics
Optional output from
diagnose_mfrm()(used whenxis fit).- facet_a
First facet name (required when
xis fit andinteraction_facetsis not supplied).- facet_b
Second facet name (required when
xis fit andinteraction_facetsis not supplied).- interaction_facets
Character vector of two or more facets.
- max_abs
Bound for absolute bias size when estimating from fit.
- omit_extreme
Omit extreme-only elements when estimating from fit.
- max_iter
Iteration cap for bias estimation when
xis fit.- tol
Convergence tolerance for bias estimation when
xis fit.- top_n
Maximum number of ranked rows to keep.
- abs_t_warn
Warning cutoff for absolute t statistics.
- abs_bias_warn
Warning cutoff for absolute bias size.
- p_max
Warning cutoff for p-values.
- sort_by
Ranking key:
"abs_t","abs_bias", or"prob".
Details
Preferred bundle API for interaction-bias diagnostics. The function can:
use a precomputed bias object from
estimate_bias(), orestimate internally from
mfrm_fit+ facet specification.
Interpreting output
Focus on ranked rows where multiple screening criteria converge:
large absolute t statistic
large absolute bias size
small screening tail area
The bundle is optimized for downstream summary() and
plot_bias_interaction() views.
Typical workflow
Run
estimate_bias()(or providemfrm_fithere).Build
bias_interaction_report(...).Review
summary(out)and visualize withplot_bias_interaction().
Examples
# \donttest{
toy <- load_mfrmr_data("example_bias")
fit <- fit_mfrm(toy, "Person", c("Rater", "Criterion"), "Score", method = "JML", maxit = 30)
#> Warning: Optimization convergence review did not produce an inference-ready numerical solution (code = 1, status = iteration_limit). Optimizer reached the iteration limit before the terminal gradient became small enough for review-only acceptance. Inspect the model specification, data support, and starting values. Do not interpret estimates until the review is resolved.
diag <- diagnose_mfrm(fit, residual_pca = "none")
bias <- estimate_bias(fit, diag, facet_a = "Rater", facet_b = "Criterion", max_iter = 2)
out <- bias_interaction_report(bias, top_n = 10)
summary(out)
#> mfrmr Bias Interaction Summary
#> Class: mfrm_bias_interaction
#> Components: 14
#>
#> Interaction summary
#> InteractionFacets InteractionOrder InteractionMode FacetA FacetB Cells
#> Rater x Criterion 2 pairwise Rater Criterion 16
#> Flagged FlaggedPercent MeanAbsT MeanAbsBias FlagStatus
#> 3 18.75 1.008 0.31 3 of 16 cell(s) flagged.
#>
#> Ranked interaction rows: ranked_table
#> InteractionFacets InteractionOrder InteractionMode FacetA FacetB Level1
#> Rater x Criterion 2 pairwise Rater Criterion R04
#> Rater x Criterion 2 pairwise Rater Criterion R01
#> Rater x Criterion 2 pairwise Rater Criterion R04
#> Rater x Criterion 2 pairwise Rater Criterion R01
#> Rater x Criterion 2 pairwise Rater Criterion R03
#> Rater x Criterion 2 pairwise Rater Criterion R01
#> Rater x Criterion 2 pairwise Rater Criterion R02
#> Rater x Criterion 2 pairwise Rater Criterion R04
#> Rater x Criterion 2 pairwise Rater Criterion R03
#> Rater x Criterion 2 pairwise Rater Criterion R02
#> Level2 ObsExpAverage BiasSize SE t Prob ObservedCount LRChiSq
#> Accuracy 0 -1.102 0.333 -3.310 0.003 24 NA
#> Accuracy 0 0.777 0.308 2.521 0.019 24 NA
#> Organization 0 0.683 0.299 2.281 0.032 24 NA
#> Organization 0 -0.362 0.293 -1.235 0.229 24 NA
#> Organization 0 -0.313 0.296 -1.059 0.301 24 NA
#> Content 0 -0.277 0.301 -0.921 0.367 24 NA
#> Accuracy 0 0.247 0.288 0.856 0.401 24 NA
#> Language 0 0.234 0.297 0.788 0.439 24 NA
#> Content 0 0.247 0.315 0.783 0.442 24 NA
#> Language 0 -0.208 0.294 -0.709 0.485 24 NA
#> LRDF LRProb ProfileCILower ProfileCIUpper ProfileCILevel ProfileCIStatus
#> NA NA NA NA NA <NA>
#> NA NA NA NA NA <NA>
#> NA NA NA NA NA <NA>
#> NA NA NA NA NA <NA>
#> NA NA NA NA NA <NA>
#> NA NA NA NA NA <NA>
#> NA NA NA NA NA <NA>
#> NA NA NA NA NA <NA>
#> NA NA NA NA NA <NA>
#> NA NA NA NA NA <NA>
#> LikelihoodBasis Pair AbsT AbsBias TFlag BiasFlag PFlag Flag
#> <NA> R04 | Accuracy 3.310 1.102 TRUE TRUE TRUE TRUE
#> <NA> R01 | Accuracy 2.521 0.777 TRUE TRUE TRUE TRUE
#> <NA> R04 | Organization 2.281 0.683 TRUE TRUE TRUE TRUE
#> <NA> R01 | Organization 1.235 0.362 FALSE FALSE FALSE FALSE
#> <NA> R03 | Organization 1.059 0.313 FALSE FALSE FALSE FALSE
#> <NA> R01 | Content 0.921 0.277 FALSE FALSE FALSE FALSE
#> <NA> R02 | Accuracy 0.856 0.247 FALSE FALSE FALSE FALSE
#> <NA> R04 | Language 0.788 0.234 FALSE FALSE FALSE FALSE
#> <NA> R03 | Content 0.783 0.247 FALSE FALSE FALSE FALSE
#> <NA> R02 | Language 0.709 0.208 FALSE FALSE FALSE FALSE
#> Facet1 Facet1_Level Facet2 Facet2_Level FacetA_Level FacetB_Level
#> Rater R04 Criterion Accuracy R04 Accuracy
#> Rater R01 Criterion Accuracy R01 Accuracy
#> Rater R04 Criterion Organization R04 Organization
#> Rater R01 Criterion Organization R01 Organization
#> Rater R03 Criterion Organization R03 Organization
#> Rater R01 Criterion Content R01 Content
#> Rater R02 Criterion Accuracy R02 Accuracy
#> Rater R04 Criterion Language R04 Language
#> Rater R03 Criterion Content R03 Content
#> Rater R02 Criterion Language R02 Language
#>
#> Settings
#> Setting Value
#> abs_t_warn 2
#> abs_bias_warn 0.5
#> p_max 0.05
#> sort_by abs_t
#> top_n 10
#>
#> Notes
#> - Bias interaction report with ranked cells and facet-level profiles.
p_bi <- plot(out, draw = FALSE)
p_bi$data$plot
#> [1] "scatter"
# }