{"id":29,"date":"2026-11-02T20:00:13","date_gmt":"2026-11-02T20:00:13","guid":{"rendered":"https:\/\/demo.galussothemes.com\/gt-ribosome\/qa-what-a-p-value-can-tell-you\/"},"modified":"2026-11-02T20:00:13","modified_gmt":"2026-11-02T20:00:13","slug":"qa-what-a-p-value-can-tell-you","status":"publish","type":"post","link":"https:\/\/demo.galussothemes.com\/gt-ribosome\/qa-what-a-p-value-can-tell-you\/","title":{"rendered":"Q&#038;A: a statistician on what a p-value can and can\u2019t tell you"},"content":{"rendered":"<p class=\"wp-block-paragraph\">Ruth Adeyemi is a statistician at Larkfield University, a fictional university in our example, where she helps biologists and doctors design experiments. We asked her about the number that appears in almost every scientific paper.<\/p>\n<h3 class=\"wp-block-heading\">What is a p-value, in one sentence?<\/h3>\n<p class=\"wp-block-paragraph\">It is the chance of getting results at least as striking as yours if the thing you are testing actually had no effect at all.<\/p>\n<h3 class=\"wp-block-heading\">So a p-value of 0.03 means a 3% chance the result is wrong?<\/h3>\n<p class=\"wp-block-paragraph\">No, and that is the most common mistake. It means: if the treatment did nothing, results this strong would turn up about 3% of the time by luck. That is not the same as the chance the treatment works. To get that, you need other information, such as how plausible the idea was in the first place.<\/p>\n<h3 class=\"wp-block-heading\">Why is 0.05 the magic line?<\/h3>\n<p class=\"wp-block-paragraph\">It is a convention, not a law of nature. A result at 0.049 is not meaningfully different from one at 0.051. Treating 0.05 as a pass mark has encouraged people to slice and re-slice their data until something crosses it.<\/p>\n<div class=\"wp-block-group gt-ribosome-box is-method is-layout-flow wp-block-group-is-layout-flow\">\n<h3 class=\"wp-block-heading gt-ribosome-box-title\">Three questions before believing a result<\/h3>\n\n\n\n<ol class=\"wp-block-list gt-ribosome-box-list\">\n<li>How big is the effect? A tiny effect can be statistically significant in a big study and still not matter.<\/li>\n\n<li>How sure are we? Look for a confidence interval: a range of plausible sizes for the effect.<\/li>\n\n<li>Has anyone repeated it? One study is a hint. Several independent studies are evidence.<\/li>\n<\/ol>\n<\/div>\n\n<h3 class=\"wp-block-heading\">What would you like readers to look for instead?<\/h3>\n<p class=\"wp-block-paragraph\">Size and range. \u201cThe new fertiliser increased yield by 4%, somewhere between 1% and 7%\u201d tells you far more than \u201cthe result was significant\u201d.<\/p>\n<h3 class=\"wp-block-heading\">Is the p-value useless, then?<\/h3>\n<p class=\"wp-block-paragraph\">Not at all. It is a useful alarm bell: it tells you whether a pattern is surprising enough to take seriously. It just cannot tell you, on its own, whether the pattern is real, large or important.<\/p><blockquote class=\"wp-block-quote is-layout-flow wp-block-quote-is-layout-flow\"><p class=\"wp-block-paragraph\">Significant is a word about surprise, not about importance.<\/p><\/blockquote>\n<h3 class=\"wp-block-heading\">What is the best thing a scientist can say?<\/h3>\n<p class=\"wp-block-paragraph\">\u201cWe are not sure yet, and here is how we will find out.\u201d That sentence has never embarrassed anyone.<\/p>","protected":false},"excerpt":{"rendered":"<p>Ruth Adeyemi explains the most misunderstood number in science, and the three questions she asks before believing a result.<\/p>\n","protected":false},"author":30,"featured_media":30,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_at_level":"Deep dive","_at_discipline":"Statistics","_at_source":"Interview","_at_reviewer":"Dr. Maren Holt, ecologist","footnotes":""},"categories":[7],"tags":[25,24,44],"class_list":["post-29","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-how-we-know","tag-interview","tag-methods","tag-statistics"],"_links":{"self":[{"href":"https:\/\/demo.galussothemes.com\/gt-ribosome\/wp-json\/wp\/v2\/posts\/29","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/demo.galussothemes.com\/gt-ribosome\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/demo.galussothemes.com\/gt-ribosome\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/demo.galussothemes.com\/gt-ribosome\/wp-json\/wp\/v2\/users\/30"}],"replies":[{"embeddable":true,"href":"https:\/\/demo.galussothemes.com\/gt-ribosome\/wp-json\/wp\/v2\/comments?post=29"}],"version-history":[{"count":0,"href":"https:\/\/demo.galussothemes.com\/gt-ribosome\/wp-json\/wp\/v2\/posts\/29\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/demo.galussothemes.com\/gt-ribosome\/wp-json\/wp\/v2\/media\/30"}],"wp:attachment":[{"href":"https:\/\/demo.galussothemes.com\/gt-ribosome\/wp-json\/wp\/v2\/media?parent=29"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/demo.galussothemes.com\/gt-ribosome\/wp-json\/wp\/v2\/categories?post=29"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/demo.galussothemes.com\/gt-ribosome\/wp-json\/wp\/v2\/tags?post=29"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}