Package index • rasch

Estimation

Pairwise conditional estimation of the Rasch model family, with Godambe sandwich standard errors and Warm person measures.

rasch(): Fit and diagnose a Rasch model by pairwise conditional estimation
rasch_mfrm(): Fit a many-facet Rasch model
rasch_efrm(): Fit the extended frame of reference model
pcml(): Estimate Rasch thresholds by pairwise conditional maximum likelihood
pcml_pc(): Estimate Rasch thresholds via the Andrich principal-components reparameterisation
threshold_index(): Enumerate item-category thresholds
item_moments(): Category-score moments for a polytomous item

Test of fit and comparison

The fit-residual and chi-square apparatus, targeting and reliability, and model comparison.

fit_summary_table(): Test-of-fit summary as a table
targeting_table(): Targeting and reliability summary as a table
chisq_detail(): Class-interval detail for one item's chi-square test of fit
test_information(): Test information function
lr_test(): Likelihood-ratio test of the partial credit against the rating scale model
compare_fits(): Compare fitted Rasch models
guttman_table(): Guttman-ordered response matrix and reproducibility

Persons

person_wle(): Warm's weighted likelihood estimates by raw score
person_extrapolated(): Person measures with extrapolated extreme scores
score_table(): Raw score to measure conversion table

Invariance and DIF

Differential item functioning over any number of person factors, DIF magnitudes in logits, resolution by item splitting, and equating.

dif_anova(): Differential item functioning by residual analysis of variance
dif_contrasts(): Planned DIF contrasts derived from the factor structure
dif_size(): DIF magnitude in logits with pairwise comparisons
split_items(): Split items by a person factor to resolve DIF
resolve_dif(): Resolve differential item functioning by iterative item splitting
equate_tests(): Equate two test calibrations through their common items
tailored_analysis(): Tailored analysis for guessing

Independence and dimensionality

Residual principal components, the Smith t-test with magnitude estimation, Q3 local dependence, and the structural remedies.

residual_correlations(): Residual correlations for local dependence (Yen's Q3)
residual_pca(): Principal components of the residual correlations
dimensionality_test(): Residual-component test of unidimensionality
dimensionality_magnitude(): Magnitude of multidimensionality from a subtest analysis
dependence_magnitude(): Estimate the magnitude of response dependence between two items
spread_test(): Spread-parameter test for dependence within subtests
combine_items(): Combine items into subtests and re-analyse
rack_data() stack_data(): Reshape repeated measurements for racked or stacked analysis

Multiple choice and traditional statistics

distractor_analysis(): Distractor analysis for multiple-choice items
distractor_rescore(): Propose polytomous option scores from the distractor evidence
ctt_table(): Traditional (classical test theory) statistics

Paired comparisons

The Bradley-Terry-Luce model as the conditional form of the dichotomous Rasch model, with judge diagnostics, within-judge dependence, judge-group DIF, and the pair-structure analogues of the independence diagnostics (transitivity and the residual bimension decomposition).

btl(): Fit the Bradley-Terry-Luce model to paired comparisons
btl_dif(): DIF analysis for paired comparisons
btl_efrm(): Fit the extended frame of reference model for paired comparisons
btl_transitivity(): Transitivity of paired comparisons
btl_dimensionality(): Residual dimensionality of paired comparisons
btl_equate(): Equate two paired-comparison calibrations through their common objects
btl_information(): Information and targeting of a paired-comparison design
btl_next_pairs(): Recommend the next informative comparisons (adaptive step)
judge_surprise(): Unexpected judgements of one judge
judge_pair_surprise(): Unexpected judgements of one judge, pair by pair
plot_btl(): Plot Bradley-Terry-Luce object locations
plot_btl_categories(): Plot graded-comparison category curves
plot_btl_icc(): Plot an object characteristic curve
plot_btl_dependence(): Plot a within-judge dependence effect
plot_btl_transitivity(): Consistency plot for paired-comparison transitivity
plot_btl_scree(): Scree of paired-comparison residual bimensions
plot_btl_dim_map(): Residual map of the leading paired-comparison bimension
plot_btl_judge_map(): Unexpected-judgement map for one judge (pair level)
plot_btl_units(): Plot the frame units of a paired-comparison EFRM fit
plot_btl_equate(): Plot a paired-comparison equating comparison
plot_btl_targeting(): Targeting plot for a paired-comparison design

Plots

One plotting function per display, all base graphics.

plot_catfreq(): Plot category frequencies
plot_ccc(): Plot category probability curves
plot_distractors(): Plot multiple-choice option curves
plot_equate(): Plot a test-equating comparison
plot_facets(): Plot facet severities
plot_frames(): Plot frame units
plot_guttman(): Plot the Guttman scalogram
plot_icc(): Plot an item characteristic curve
plot_icc_frames(): Plot an item's characteristic curves across frames
plot_item_map(): Plot the item map (location against fit residual)
plot_kidmap(): Plot a kidmap
plot_pca(): Plot residual principal-component loadings
plot_pca_biplot(): Biplot of the first two residual components
plot_pcc(): Plot a person characteristic curve
plot_person_fit(): Plot person fit
plot_pimap(): Plot the person-item threshold distribution
plot_recovery(): Recovery scatter of planted against recovered parameters
plot_resid_cor(): Plot the residual-correlation heatmap
plot_resid_dist(): Plot the fit residual distribution
plot_scree(): Scree plot of the residual components with parallel analysis
plot_tcc(): Plot the test characteristic curve
plot_threshold_map(): Plot the threshold map
plot_threshold_prob(): Plot threshold probability curves
plot_tif(): Plot the test information function
plot_wright(): Plot a Wright map

Data simulation

Generate data from each model family with dial-in departures from it, so a known pathology can be planted and the matching diagnostic watched as it fires. The true parameters are attached to every result.

simulate_rasch(): Simulate person-by-item Rasch data with dial-in misfit
simulate_btl(): Simulate paired-comparison (BTL) data with dial-in misfit
simulate_mfrm(): Simulate many-facet (rated) data with dial-in misfit
simulate_efrm(): Simulate extended frame-of-reference data with differing units
simulate_btl_efrm(): Simulate paired-comparison EFRM data with differing frame units
sim_replicate(): Replicate a simulation for Monte Carlo studies
sim_recovery(): Parameter recovery of a fit against the simulation truth
plot_recovery(): Recovery scatter of planted against recovered parameters

Export and interface

save_outputs(): Save every output of a Rasch analysis to a folder
save_item_plots(): Save a plot for every item
save_person_plots(): Save a kidmap for every person
report_html(): Write a self-contained HTML report of a Rasch analysis
run_app(): Launch the rasch graphical interface