dif_report: Generate report

In knickodem/WBdif: Investigate DIF and generate reports

Generate report

Description

Produces a report summarizing DIF analysis for a given measure and grouping variable, the robustness of treatment effect estimates to DIF, or both

Usage

dif_report(
  dif.analysis,
  file.name,
  report.format = "html_document",
  report.title = file.name,
  measure.name = "measure",
  dataset.name = "dataset",
  biased.items = "IRT"
)

effect_report(
  dif.models,
  effect.robustness,
  file.name,
  report.format = "html_document",
  report.title = file.name,
  measure.name = "measure",
  dataset.name = "dataset"
)

dif_effect_report(
  dif.analysis,
  dif.models,
  effect.robustness,
  file.name,
  report.format = "html_document",
  report.title = file.name,
  measure.name = "measure",
  dataset.name = "dataset",
  biased.items = "IRT"
)

Arguments

dif.analysis	An object returned from dif_analysis
file.name	File name to create on disk. The file path can also be specified here. If the path is omitted, the file is saved to the working directory.
report.format	File format of the report. Default is HTML ("html_document"). See render for other options.
report.title	An optional character string indicating the report title, which is printed in the report.
measure.name	An optional character string naming the measure being evaluated, which is printed in the report.
dataset.name	An optional character string naming the dataset used, which is printed in the report.
biased.items	Οne of c("MH", "logistic", "IRT"). Determines which DIF method should be used to identify biased items. Default is "IRT".
dif.models	An object returned from dif_models
effect.robustness	An object returned from effect_robustness

Details

If uniform DIF is detected in dif.analysis with biased.items = "IRT", the direction of uniform DIF is determined using the b parameter for dichotomous items and the location of the first threshold for polytomous items.

Value

a summary report of:

• dif_report - DIF analysis results conducted in dif.analysis

• effect_report - robustness of treatment effects to DIF as estimated from dif_models and effect_robustness

• dif_effect_report - DIF analysis results and treatment effect robustness checks

Examples

data("mdat")

# prep data
dif.data <- dif_data_prep(item.data = mdat`[`5:ncol(mdat)],
                             tx.group.id = mdat$treated,
                             dif.group.id = mdat$gender,
                             cluster.id = mdat$clusterid,
                             na.to.0 = TRUE)

# DIF analysis by dif.group.id
# using rest scores and binning match.scores by deciles to avoid empty cells in MH analysis
dif.analysis <- dif_analysis(dif.data = dif.data,
                           methods =  c("MH", "IRT"),
                           match.type = "Rest",
                           match.bins = seq(0, 1, by = .1))

# Report only DIF analysis results
dif_report(dif.analysis = dif.analysis,
           file.name = "DIF-Gender-MDAT-Language",
           report.format = "html_document",
           report.title = "MDAT Language: Gender DIF",
           measure.name = "MDAT Language",
           biased.items = "IRT")

# Report DIF analysis and treatment effect robustness check
dif.models <- dif_models(dif.analysis = dif.analysis, biased.items = "MH")
effect.robustness <- effect_robustness(dif.models = dif.models, irt.scoring = "WLE")
dif_report(dif.analysis = dif.analysis,
           dif.models = dif.models,
           effect.robustness = effect.robustness,
           file.name = "DIF-Effect-Gender-MDAT-Language",
           biased.items = "IRT",
           report.format = "html_document",
           report.title = "MDAT Language: Gender DIF and Tx Effects",
           measure.name = "MDAT Language")

knickodem/WBdif documentation built on Feb. 3, 2024, 2:20 a.m.