Skip to contents

Build a bias pairwise-contrast report

Source: R/api-reports.R
bias_pairwise_report.Rd

Build a bias pairwise-contrast report

Usage

bias_pairwise_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,
  target_facet = NULL,
  context_facet = NULL,
  top_n = 50,
  p_max = 0.05,
  sort_by = c("abs_t", "abs_contrast", "prob")
)

Arguments

x

Output from estimate_bias() or fit_mfrm().

diagnostics

Optional output from diagnose_mfrm() (used when x is fit).

facet_a

First facet name (required when x is fit and interaction_facets is not supplied).

facet_b

Second facet name (required when x is fit and interaction_facets is 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 x is fit.

tol

Convergence tolerance for bias estimation when x is fit.

target_facet

Facet whose local contrasts should be compared across the paired context facet. Defaults to the first interaction facet.

context_facet

Optional facet to condition on. Defaults to the other facet in a 2-way interaction.

top_n

Maximum number of ranked rows to keep.

p_max

Flagging cutoff for pairwise p-values.

sort_by

Ranking key: "abs_t", "abs_bias", or "prob".

Value

A named list with:

  • table: pairwise contrast rows

  • summary: one-row contrast summary

  • orientation_audit: interaction-facet sign audit

  • settings: resolved reporting options

Details

This helper exposes the pairwise contrast table that was previously only reachable through fixed-width output generation. It is available only for 2-way interactions. The pairwise contrast statistic uses a Welch/Satterthwaite approximation and is labeled as a Rasch-Welch comparison in the output metadata.

See also

estimate_bias(), bias_interaction_report(), build_fixed_reports()

Examples

toy <- load_mfrmr_data("example_bias")
fit <- fit_mfrm(toy, "Person", c("Rater", "Criterion"), "Score", method = "JML", maxit = 25)
diag <- diagnose_mfrm(fit, residual_pca = "none")
out <- bias_pairwise_report(fit, diagnostics = diag, facet_a = "Rater", facet_b = "Criterion")
summary(out)
#> mfrmr Bias Pairwise Summary 
#>   Class: mfrm_bias_pairwise
#>   Components (6): table, summary, orientation_audit, settings, direction_note, recommended_action
#> 
#> Pairwise summary
#>  TargetFacet ContextFacet Contrasts Flagged MeanAbsContrast MeanAbsT MixedSign
#>        Rater    Criterion        24       0           0.512      NaN     FALSE
#> 
#> Contrast rows: table
#>  Target Target N Target Measure Target S.E. Context1 Context1 N Local Measure1
#>     R01        1         -0.866       0.152 Accuracy          1         -0.090
#>     R01        1         -0.866       0.152 Accuracy          1         -0.090
#>     R01        1         -0.866       0.152 Accuracy          1         -0.090
#>     R01        1         -0.866       0.152  Content          2         -1.144
#>     R01        1         -0.866       0.152  Content          2         -1.144
#>     R01        1         -0.866       0.152 Language          3         -1.050
#>     R02        2          0.004       0.147 Accuracy          1          0.250
#>     R02        2          0.004       0.147 Accuracy          1          0.250
#>     R02        2          0.004       0.147 Accuracy          1          0.250
#>     R02        2          0.004       0.147  Content          2         -0.026
#>  SE1 Obs-Exp Avg1 Count1 ObsN1     Context2 Context2 N Local Measure2 SE2
#>   NA           NA     24    24      Content          2         -1.144  NA
#>   NA           NA     24    24     Language          3         -1.050  NA
#>   NA           NA     24    24 Organization          4         -1.229  NA
#>   NA           NA     24    24     Language          3         -1.050  NA
#>   NA           NA     24    24 Organization          4         -1.229  NA
#>   NA           NA     24    24 Organization          4         -1.229  NA
#>   NA           NA     24    24      Content          2         -0.026  NA
#>   NA           NA     24    24     Language          3         -0.205  NA
#>   NA           NA     24    24 Organization          4         -0.019  NA
#>   NA           NA     24    24     Language          3         -0.205  NA
#>  Obs-Exp Avg2 Count2 ObsN2 Contrast SE  t d.f. Prob. InferenceTier
#>            NA     24    24    1.054 NA NA   NA    NA     screening
#>            NA     24    24    0.960 NA NA   NA    NA     screening
#>            NA     24    24    1.139 NA NA   NA    NA     screening
#>            NA     24    24   -0.094 NA NA   NA    NA     screening
#>            NA     24    24    0.085 NA NA   NA    NA     screening
#>            NA     24    24    0.179 NA NA   NA    NA     screening
#>            NA     24    24    0.276 NA NA   NA    NA     screening
#>            NA     24    24    0.454 NA NA   NA    NA     screening
#>            NA     24    24    0.269 NA NA   NA    NA     screening
#>            NA     24    24    0.178 NA NA   NA    NA     screening
#>  SupportsFormalInference FormalInferenceEligible PrimaryReportingEligible
#>                    FALSE                   FALSE                    FALSE
#>                    FALSE                   FALSE                    FALSE
#>                    FALSE                   FALSE                    FALSE
#>                    FALSE                   FALSE                    FALSE
#>                    FALSE                   FALSE                    FALSE
#>                    FALSE                   FALSE                    FALSE
#>                    FALSE                   FALSE                    FALSE
#>                    FALSE                   FALSE                    FALSE
#>                    FALSE                   FALSE                    FALSE
#>                    FALSE                   FALSE                    FALSE
#>    ReportingUse
#>  screening_only
#>  screening_only
#>  screening_only
#>  screening_only
#>  screening_only
#>  screening_only
#>  screening_only
#>  screening_only
#>  screening_only
#>  screening_only
#>                                                                                      ContrastBasis
#>  difference between local target measures across contexts (target term cancels to a bias contrast)
#>  difference between local target measures across contexts (target term cancels to a bias contrast)
#>  difference between local target measures across contexts (target term cancels to a bias contrast)
#>  difference between local target measures across contexts (target term cancels to a bias contrast)
#>  difference between local target measures across contexts (target term cancels to a bias contrast)
#>  difference between local target measures across contexts (target term cancels to a bias contrast)
#>  difference between local target measures across contexts (target term cancels to a bias contrast)
#>  difference between local target measures across contexts (target term cancels to a bias contrast)
#>  difference between local target measures across contexts (target term cancels to a bias contrast)
#>  difference between local target measures across contexts (target term cancels to a bias contrast)
#>                                         SEBasis                  StatisticLabel
#>  combined context-specific bias standard errors Bias-contrast Welch screening t
#>  combined context-specific bias standard errors Bias-contrast Welch screening t
#>  combined context-specific bias standard errors Bias-contrast Welch screening t
#>  combined context-specific bias standard errors Bias-contrast Welch screening t
#>  combined context-specific bias standard errors Bias-contrast Welch screening t
#>  combined context-specific bias standard errors Bias-contrast Welch screening t
#>  combined context-specific bias standard errors Bias-contrast Welch screening t
#>  combined context-specific bias standard errors Bias-contrast Welch screening t
#>  combined context-specific bias standard errors Bias-contrast Welch screening t
#>  combined context-specific bias standard errors Bias-contrast Welch screening t
#>    ProbabilityMetric                           DFBasis AbsT AbsContrast  Flag
#>  screening tail area Welch-Satterthwaite approximation   NA       1.054 FALSE
#>  screening tail area Welch-Satterthwaite approximation   NA       0.960 FALSE
#>  screening tail area Welch-Satterthwaite approximation   NA       1.139 FALSE
#>  screening tail area Welch-Satterthwaite approximation   NA       0.094 FALSE
#>  screening tail area Welch-Satterthwaite approximation   NA       0.085 FALSE
#>  screening tail area Welch-Satterthwaite approximation   NA       0.179 FALSE
#>  screening tail area Welch-Satterthwaite approximation   NA       0.276 FALSE
#>  screening tail area Welch-Satterthwaite approximation   NA       0.454 FALSE
#>  screening tail area Welch-Satterthwaite approximation   NA       0.269 FALSE
#>  screening tail area Welch-Satterthwaite approximation   NA       0.178 FALSE
#> 
#> Settings
#>        Setting     Value
#>   target_facet     Rater
#>  context_facet Criterion
#>          top_n        50
#>          p_max      0.05
#>        sort_by     abs_t
#> 
#> Notes
#>  - Summary table and preview rows were extracted.