McNemar's statistical test addition: ard_mcnemartest

[Melkiades]

What changes are proposed in this pull request?

• Added ard_mcnemartest() statistical test function.

Closes #40

• I am wondering if we need to have ccorrect = TRUE as default, following the default in stats, and if we should make this option more apparent in the case we change this default.

• Also, I do not know if it would be relevant but I used an utility function to find dichotomies in a data-set that could result useful as a tool if there is not one already doing this:

# Assuming your data.frame is named df
df <- data.frame(
  A = c(1, 0, 1, 0), # dichotomous
  B = c(1, 2, 3, 4), # not dichotomous
  C = c("Yes", "No", "Yes", "No"), # dichotomous
  D = c(TRUE, FALSE, TRUE, FALSE) # dichotomous
)

# Function to find dichotomous columns
find_dichotomous_columns <- function(df) {
  dichotomous_columns <- c()
  for (col_name in names(df)) {
    if (length(unique(df[[col_name]])) == 2) {
      dichotomous_columns <- c(dichotomous_columns, col_name)
    }
  }
  return(dichotomous_columns)
}

# Find and display dichotomous columns
dichotomous_columns <- find_dichotomous_columns(df)
print(dichotomous_columns)
#> [1] "A" "C" "D"
di_cols <- find_dichotomous_columns(cards::ADSL)
print(di_cols)
#>  [1] "SEX"      "ETHNIC"   "EFFFL"    "COMP8FL"  "COMP16FL" "COMP24FL"
#>  [7] "DISCONFL" "DSRAEFL"  "DTHFL"    "DURDSGR1"

^{Created on 2024-02-14 with reprex v2.1.0}

• .paired_data_pivot_wider is in the Wilcoxon test. Should we have it as a tool in a dedicated file? It looks general to me.

---

Pre-review Checklist (if item does not apply, mark is as complete)

• All GitHub Action workflows pass with a ✅
• PR branch has pulled the most recent updates from master branch: usethis::pr_merge_main()
• If a bug was fixed, a unit test was added.
• Code coverage is suitable for any new functions/features (generally, 100% coverage for new code): devtools::test_coverage()
• Request a reviewer

Reviewer Checklist (if item does not apply, mark is as complete)

• If a bug was fixed, a unit test was added.
• Run pkgdown::build_site(). Check the R console for errors, and review the rendered website.
• Code coverage is suitable for any new functions/features: devtools::test_coverage()

When the branch is ready to be merged:

• Update NEWS.md with the changes from this pull request under the heading "# cards (development version)". If there is an issue associated with the pull request, reference it in parentheses at the end update (see NEWS.md for examples).
• All GitHub Action workflows pass with a ✅
• Approve Pull Request
• Merge the PR. Please use "Squash and merge" or "Rebase and merge".

[github-actions]

Unit Tests Summary

 1 files  14 suites   2s ⏱️
14 tests 10 ✅ 4 💤 0 ❌
30 runs  22 ✅ 8 💤 0 ❌

Results for commit 5781b8d.

♻️ This comment has been updated with latest results.

[github-actions]

Code Coverage Summary

Filename                 Stmts    Miss  Cover    Missing
---------------------  -------  ------  -------  ------------------------------------
R/ard_chisqtest.R           31       0  100.00%
R/ard_fishertest.R          33       0  100.00%
R/ard_kruskaltest.R         27       0  100.00%
R/ard_mcnemartest.R         41       0  100.00%
R/ard_proportion_ci.R       40       5  87.50%   63-67
R/ard_regression.R          49       0  100.00%
R/ard_ttest.R               78       0  100.00%
R/ard_wilcoxtest.R          83       0  100.00%
R/proportion_ci.R          182      28  84.62%   241, 244, 253-258, 266, 281, 381-404
TOTAL                      564      33  94.15%

Diff against main

Filename               Stmts    Miss  Cover
-------------------  -------  ------  --------
R/ard_mcnemartest.R      +41       0  +100.00%
TOTAL                    +41       0  +0.46%

Results for commit: 5781b8d

Minimum allowed coverage is 80%

♻️ This comment has been updated with latest results

[github-actions]

Unit Test Performance Difference

Test Suite	$Status$	Time on main	$±Time$	$±Tests$	$±Skipped$	$±Failures$	$±Errors$
ard_mcnemartest	👶		$+0.17$	$+4$	$0$	$0$	$0$

Additional test case details

Test Suite	$Status$	Time on main	$±Time$	Test Case
ard_mcnemartest	👶		$+0.17$	ard_mcnemartest_works

Results for commit 462ffb7

♻️ This comment has been updated with latest results.

[ddsjoberg]

thanks @Melkiades !

I am wondering if we need to have correct = TRUE as default, following the default in stats, and if we should make this option more apparent in the case we change this default.

Let's keep the same defaults as the the R function. To keep the API consistent with the other ard stat functions, we can still use ... with the explanation that these results will be passed to mcnemar.test(...). We can use the @inheritDotParams tag to make it very clear which options the user has.

.paired_data_pivot_wider is in the Wilcoxon test. Should we have it as a tool in a dedicated file? It looks general to me.

Very good point. A separate utils.R file would be great place for this function. Or do you think we make it an exported utility , and if we do that, should it live in cards instead?

[Melkiades]

.paired_data_pivot_wider is in the Wilcoxon test. Should we have it as a tool in a dedicated file? It looks general to me.

Very good point. A separate utils.R file would be great place for this function. Or do you think we make it an exported utility , and if we do that, should it live in cards instead?

To me, it seems useful, but I do not know if it is necessarily a missing piece in {cards}' toolbox. I would add it to a group of related utils and export it if you think it would be used. For users, I would make by1 and by2 customizable and the paired values as defaults (also could it make sense to extend it beyond paired data?).

[ddsjoberg]

This looks great! Just a few comments to address.

[codecov-commenter]

Welcome to Codecov 🎉

Once merged to your default branch, Codecov will compare your coverage reports and display the results in this comment.

Thanks for integrating Codecov - We've got you covered ☂️

[Melkiades]

@ddsjoberg thank you for your review! Could you take another look at the changes? ;)

[ddsjoberg]

thanks @Melkiades !! looks great!!