Talk:New study analyzes the varying levels of protection offered by COVID-19 vaccines

Thank you
Hey @SVTCobra, thanks for catching that title case issue. I got too focused on the article and getting the image to work that I forgot to check the headline. I'll be better about that next time. Dr vulpes (talk) 10:26, 11 May 2023 (UTC)


 * Speaking of the image. At a glance it looks like it is copyrighted. SVTCobra 12:49, 11 May 2023 (UTC)
 * Hey @SVTCobra, the paper and it's figures were published under the Creative Commons Attribution 4.0 license. I thought that license was fine to use but if I'm wrong just let me know and I'll remove it. Here's a link to it's license notice https://www.nature.com/articles/s41586-023-05896-x#rightslink Dr vulpes (talk) 18:23, 11 May 2023 (UTC)
 * OK, cool. When I clicked on the source link from Commons, I only saw the "© 2023 Springer Nature Limited" at the bottom of the page. All good! Thanks. SVTCobra 18:32, 11 May 2023 (UTC)
 * I admit I did this late last night and was kind of tired you scared me there for a second! Glad it's ok, I couldn't really find any other images that would work for this article. Normally if the data was publicly available I would just use that to make the figures in R or something. Dr vulpes (talk) 18:40, 11 May 2023 (UTC)
 * I was going to this link which is your source on Commons. It doesn't show the CC-BY-4.0. Sorry if I scared you. I will see if I can find a way to link the rights on Commons. I always fear Commons will delete an image we use and then Wikinews ends up with missing image in our archives. That's why I brought it up. Cheers, SVTCobra 21:40, 11 May 2023 (UTC)

Math
"Smokers showed the strongest reduction with measured antibodies being 240% lower than for nonsmokers." So they have negative 140% as many antibodies? It would be meaningful to say that non-smokers had 240% as many antibodies as smokers, but .... IAmNitpicking (talk) 12:39, 13 May 2023 (UTC)


 * A 240% decreases doesn’t make -140%, you can’t have negative antibodies the bottom of the range can only be zero. Both 240% decrease in smokers and 240% greater in non-smokers work. The text from the article for this fact was “Similarly, non-smokers had 2.4-fold higher nAB compared to smokers (mean difference = − 0.37, CI − 0.64 to − 0.10; p = 0.007; Fig. 4B).”. Dr vulpes (talk) 16:48, 13 May 2023 (UTC)
 * Quite correct. The idea of 240% less is just wrong.
 * While it is possible to restate this as 60% lower, it is conceptually challenging to get the quantitative picture even then. Thus saying something like "less than half the antibodies" would be easy to grasp.
 * (As an aside, upon reading the original paper, I found the article pretty shaky. The statistics are all over the place and were apparently run through a statistical analysis program without much critical thinking. General trends without p-values is about all you can report that is believable.) Mlee (talk) 01:16, 16 May 2023 (UTC)
 * I puzzled over this for a while and then just realized I was reading it all incorrectly and this is a simple factual error: smokers had -240% the number of antibodies...which is obviously very different from 240% less antibodies. So don't we need a correction, SVTCobra, as this is substantive and well over the 24-hour deadline? Heavy Water (talk) 02:27, 16 May 2023 (UTC)
 * I haven't got the time. It's not archived, so do what you think is best. SVTCobra 12:27, 16 May 2023 (UTC)
 * @Mlee: Did I get this correction right? Heavy Water (talk) 15:00, 16 May 2023 (UTC)