Skip to contents

Bootstrap IPW-Based Bias-Corrected Estimates for Ranking Data

Source: R/imprr_weights_boot.R
imprr_weights_boot.Rd

This function repeatedly resamples respondents, reruns imprr_weights(), and summarizes downstream quantities of interest such as average ranks, pairwise probabilities, top-k probabilities, and marginal rank probabilities. It provides bootstrap uncertainty estimates for the IPW workflow in a format parallel to imprr_direct().

Usage

imprr_weights_boot(
  data,
  J = NULL,
  main_q,
  anc_correct = NULL,
  population = "non-random",
  assumption = "contaminated",
  n_bootstrap = 200,
  seed = 123456,
  weight = NULL,
  verbose = FALSE,
  p_random = NULL
)

Arguments

data

The input dataset with ranking data.

J

The number of items in the ranking question. Defaults to NULL, in which case it will be inferred from the data.

main_q

Ranking question to be analyzed. When `main_q` is a single column name or unquoted symbol such as `app_identity`, the function looks for `app_identity_1`, `app_identity_2`, `app_identity_3`, and so on. You may also supply `main_q` directly as a character vector or unquoted `c(...)` expression of ranking columns such as `c(party, gender, race, religion)`.

anc_correct

Optional indicator for passing the anchor question. If `NULL`, `p_random` is used when supplied; otherwise the function defaults to `p_random = 0` and applies no correction.

population

Choice of the target population out of non-random respondents (default) or all respondents.

assumption

Choice of identifying assumption when `population = "all"`: `uniform` assumes random respondents would have uniform counterfactual preferences, while `contaminated` assumes their counterfactual preferences match those of non-random respondents.

n_bootstrap

Number of bootstrap resamples. Defaults to 200.

seed

Seed for set.seed for reproducibility.

weight

The name of the weight column in `data`. Defaults to `NULL`, which uses equal weights. This can also be supplied as a numeric vector or as an unquoted column name.

verbose

Indicator for verbose output. Defaults to FALSE.

p_random

Optional fixed proportion of random/inattentive respondents. When supplied, this overrides `anc_correct` and a message is shown if both are provided.

Value

A list with two elements:

est_p_random

A data frame with summary statistics for the estimated proportion of random respondents, including columns mean, lower, and upper (95% confidence interval).

results

A data frame with bootstrap summaries for the IPW-based bias-corrected quantities of interest, grouped by item, qoi, and outcome, with columns mean, lower, and upper.

Examples

out <- imprr_weights_boot(
  identity,
  main_q = "app_identity",
  anc_correct = "anc_correct_identity",
  n_bootstrap = 2,
  seed = 123
)
#> No weight column supplied; using equal weights for all observations.
out$est_p_random
#>        mean     lower     upper
#> 1 0.3172868 0.3108736 0.3237001
head(out$results)
#> # A tibble: 6 × 6
#>   item           qoi              outcome               mean  lower  upper
#>   <chr>          <chr>            <chr>                <dbl>  <dbl>  <dbl>
#> 1 app_identity_1 average rank     Avg: app_identity_1 3.27   3.27   3.27  
#> 2 app_identity_1 marginal ranking Ranked 1            0.0590 0.0545 0.0635
#> 3 app_identity_1 marginal ranking Ranked 2            0.133  0.124  0.142 
#> 4 app_identity_1 marginal ranking Ranked 3            0.287  0.281  0.293 
#> 5 app_identity_1 marginal ranking Ranked 4            0.521  0.519  0.523 
#> 6 app_identity_1 pairwise ranking v. app_identity_2   0.351  0.348  0.355