https://www.explainxkcd.com/wiki/api.php?action=feedcontributions&feedformat=atom&user=Pescamillamexplain xkcd - User contributions [en]2026-06-15T07:55:31ZUser contributionsMediaWiki 1.30.0https://www.explainxkcd.com/wiki/index.php?title=1574:_Trouble_for_Science&diff=1012861574: Trouble for Science2015-09-07T13:19:35Z<p>Pescamillam: fixed wrong links</p>
<hr />
<div>{{comic<br />
| number = 1574<br />
| date = September 7, 2015<br />
| title = Trouble for Science<br />
| image = trouble_for_science.png<br />
| titletext = Careful mathematical analysis demonstrates small-scale irregularities in Gaussian distribution<br />
}}<br />
<br />
==Explanation==<br />
Five newspaper headlines are shown:<br />
<br />
;Many commercial antibody-based immunoassays are unreliable<br />
This sentence is true. See Kebaneilwe Lebani, [http://espace.library.uq.edu.au/view/UQ:352531 Antibody Discovery for Development of a Serotyping Dengue Virus NS1 Capture Assay], 2014. In this PhD thesis, 11 references are given.<br />
<br />
;Problems with the p-value as an indicator of significance<br />
p-value is the probability that an event is observed just by chance. If p-value is under a treshold level (''α'', usually <5%, or <1% for being more conservative) one can assume that the event observed "exists".<br />
The value used for ''α'' has been proposed by [http://web.lru.dk/sites/lru.dk/files/lru/docs/kap9/kapitel_9_126_On_the_origins.pdf Fisher] and is completely arbitrary.<br />
<br />
The use of p-values as a measure of statistical significance is frequently criticized, for example in [http://wiki.bio.dtu.dk/~agpe/papers/pval_notuseful.pdf Hubbard and Lindsay].<br />
;Overfeeding of laboratory rodents compromises animal models<br />
<br />
[http://tpx.sagepub.com/content/24/6/757.full.pdf Keenan et al.] make this case.<br />
<br />
;Replication study fails to reproduce many published results<br />
A [https://explorable.com/replication-study Replication Study] is a study designed to replicate the results of a previous study by using the same methods for a different set of subjects and experimenters. It aims to recreate the results to gain confidence in the results of the previous study as well as ensuring that the findings of the previous study are transferable to other similar areas of study.<br />
<br />
Randall is probably referring to this recent study: http://www.nature.com/news/over-half-of-psychology-studies-fail-reproducibility-test-1.18248<br />
<br />
;Controlled trials show Bunsen burners make things colder<br />
This is a joke. There is probably some methodological error if putting something under the Bunsen burner flame (which is between 1000K and 2000K) makes it colder. Unless of course that thing is already much hotter than the flame (more than 2000 degrees Kelvin). <br />
<br />
Title text: <br />
<br />
;Careful mathematical analysis demonstrates small-scale irregularities in Gaussian distribution<br />
<br />
This is another joke, as gaussian samples are only approximated by the gaussian distribution. The quality of the approximation grows with the size of the sample, but will never be exact, meaning it will show "irregularities" on a "small scale".<br />
<br />
==Transcript==<br />
[A simple comic consisting of five newspaper headlines]<br />
<br />
Many Commercial Antibody-Based Immunoassays Are Unreliable<br />
<br />
Problems With the p-Value as an Indicator of Significance<br />
<br />
Overfeeding of Laboratory Rodents Compromises Animal Models<br />
<br />
Replication Study Fails to Reproduce Many Published Results<br />
<br />
Controlled Trials Show Bunsen Burners Make Things Colder<br />
<br />
<br />
{{comic discussion}}<br />
<br />
[[Category:Science]]<br />
[[Category:Biology]]<br />
[[Category:Chemistry]]<br />
[[Category:Math]]</div>Pescamillam