Talk:Two-proportion Z-test

Two-proportion Z-test was nominated as a Mathematics good article, but it did not meet the good article criteria at the time (October 8, 2025, reviewed version). There are suggestions on the review page for improving the article. If you can improve it, please do; it may then be renominated.

GA review

Failed. —David Eppstein (talk) 05:58, 8 October 2025 (UTC)[reply]

The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

This review is transcluded from Talk:Two-proportion Z-test/GA1. The edit link for this section can be used to add comments to the review.

Nominator: Talgalili (talk · contribs) 13:55, 7 October 2025 (UTC)[reply]

Reviewer: David Eppstein (talk · contribs) 05:18, 8 October 2025 (UTC)[reply]

Lead too technical; does not provide context for readers who do not already know what all these terms mean. The claim "It is used in various fields to compare success rates, response rates, or other proportions across different groups." appears to be unsourced and not a summary of later material; most later material is not summarized in the lead.

Reference

There are many unsourced sentences, bullet points, and whole paragraphs:

Two at the end of "Hypothesis test"
Two at the end of "Confidence interval"
One at the start and one at the end of the main "Minimum detectable effect" text
Almost all of the hidden "Minimum detectable effect" proof
One at the start and another at the end of "Assumptions and conditions"
One near the start and several bullets at the end of "Relation to other statistical methods"
All of "Example"
All of the code examples

All claims in all paragraphs should be supported by published reliable sources.

Several sources appear likely to be unreliable by Wikipedia standards

[1] stattrek: self-published?
[6] COOLSerdash is an open forum, stackexchange
[8] blog.statsig.com appears to be a blog
[9] blog.x.com appears to be a blog

Many other footnotes appear to be from reliable sources but are missing important metadata:

[2] Su and [7] Chow give books but not the page numbers cited from it
[3] Intro stat 2e should list authors
[4] nist.gov gives only a page title
[10] missing doi doi:10.20982/tqmp.03.2.p043
[11] what is this excerpted from?
[12] missing doi doi:10.1002/(SICI)1097-0258(19980430)17:8<873::AID-SIM779>3.0.CO;2-I

Reference [10] failed verification: Its suggestion in Table 3 for a number of samples "At least 20 overall, no cell smaller than 5" is for a different test. It is possible that this test comes under the table entry for "Measuring group differences" but that has different sample size suggestions. And there is no support for this suggestion being "Typically".

Overall this is very far from the sourcing criteria of the Good Article criteria, WP:GACR #2. It was not ready for a Good Article nomination. I think it is also far from the requirement in GACR #1 that the material be "understandable to an appropriately broad audience". These issues need to be addressed before this is renominated. —David Eppstein (talk) 05:58, 8 October 2025 (UTC)[reply]

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

Article name

I'm considering to rename this article into: "Two independent proportions tests", or "Comparing two independent proportions", similar to Binomial proportion confidence interval - so as to allow the article to include various variations of solutions. Tal Galili (talk) 16:34, 15 October 2025 (UTC)[reply]