toad.social

Dr Mircea Zloteanu ☀️ 🌊🌴<a href="https://mastodon.social/tags/statstab" class="mention hashtag" rel="nofollow noopener" target="_blank">#statstab</a> #363 p-checker The one-for-all p-value analyzerThoughts: Easy way to check for publication bias using some current tools.<a href="https://mastodon.social/tags/shiny" class="mention hashtag" rel="nofollow noopener" target="_blank">#shiny</a> <a href="https://mastodon.social/tags/pvalue" class="mention hashtag" rel="nofollow noopener" target="_blank">#pvalue</a> <a href="https://mastodon.social/tags/phacking" class="mention hashtag" rel="nofollow noopener" target="_blank">#phacking</a> <a href="https://mastodon.social/tags/QRPs" class="mention hashtag" rel="nofollow noopener" target="_blank">#QRPs</a> <a href="https://mastodon.social/tags/zcurve" class="mention hashtag" rel="nofollow noopener" target="_blank">#zcurve</a> <a href="https://mastodon.social/tags/bias" class="mention hashtag" rel="nofollow noopener" target="_blank">#bias</a> <a href="https://mastodon.social/tags/pcurve" class="mention hashtag" rel="nofollow noopener" target="_blank">#pcurve</a> <a href="https://mastodon.social/tags/rindex" class="mention hashtag" rel="nofollow noopener" target="_blank">#rindex</a><a href="https://shinyapps.org/apps/p-checker/" rel="nofollow noopener" translate="no" target="_blank">https://shinyapps.org/apps/p-checker/</a>

Dr Mircea Zloteanu ☀️ 🌊🌴<a href="https://mastodon.social/tags/statstab" class="mention hashtag" rel="nofollow noopener" target="_blank">#statstab</a> #358 What are some of the problems with stepwise regression?Thoughts: Model selection is not an easy task, but maybe don't naively try step wise reg.<a href="https://mastodon.social/tags/stepwise" class="mention hashtag" rel="nofollow noopener" target="_blank">#stepwise</a> <a href="https://mastodon.social/tags/regression" class="mention hashtag" rel="nofollow noopener" target="_blank">#regression</a> <a href="https://mastodon.social/tags/QRPs" class="mention hashtag" rel="nofollow noopener" target="_blank">#QRPs</a> <a href="https://mastodon.social/tags/issues" class="mention hashtag" rel="nofollow noopener" target="_blank">#issues</a> <a href="https://mastodon.social/tags/phacking" class="mention hashtag" rel="nofollow noopener" target="_blank">#phacking</a> <a href="https://mastodon.social/tags/modelselection" class="mention hashtag" rel="nofollow noopener" target="_blank">#modelselection</a> <a href="https://mastodon.social/tags/bias" class="mention hashtag" rel="nofollow noopener" target="_blank">#bias</a><a href="https://www.stata.com/support/faqs/statistics/stepwise-regression-problems/" rel="nofollow noopener" translate="no" target="_blank">https://www.stata.com/support/faqs/statistics/stepwise-regression-problems/</a>

Gert :debian: :gnu: :linux:Il P-hacking è la pratica di modificare l'analisi o i dati per ottenere un risultato statisticamente significativo. Si cerca di ottenere un risultato desiderabile e si riportano solo i risultati ottenuti, ignorando tutte le volte in cui non si è ottenuto nulla. Potrebbe portare a una pubblicazione a breve termine, ma il P-hacking contribuisce alla crisi di riproducibilità e replicabilità nella scienza, riempiendo la letteratura scientifica di conclusioni dubbie o infondate. <a href="https://qoto.org/tags/science" class="mention hashtag" rel="nofollow noopener" target="_blank">#science</a> <a href="https://qoto.org/tags/statistic" class="mention hashtag" rel="nofollow noopener" target="_blank">#statistic</a> <a href="https://qoto.org/tags/phacking" class="mention hashtag" rel="nofollow noopener" target="_blank">#phacking</a> <a href="https://www.nature.com/articles/d41586-025-01246-1" rel="nofollow noopener" target="_blank">https://www.nature.com/articles/d41586-025-01246-1</a>

Felix Schönbrodt<a href="https://scicomm.xyz/tags/phacking" class="mention hashtag" rel="nofollow noopener" target="_blank">#phacking</a> <a href="https://scicomm.xyz/tags/LLM" class="mention hashtag" rel="nofollow noopener" target="_blank">#LLM</a> performance:"The researchers found that big players like Meta, Google, OpenAI, and Amazon are given special privileges to privately test multiple versions of their models and only publish the best results. This hidden practice allows them to inflate their rankings by cherry picking data, making their models appear stronger than they actually are."<a href="https://arxiv.org/abs/2504.20879?utm_source=beehiiv&utm_medium=newsletter&utm_campaign=mediamobilize&_bhlid=045dede5ad6eedcb96ec953f2fbac11a166a2243" rel="nofollow noopener" translate="no" target="_blank">https://arxiv.org/abs/2504.20879?utm_source=beehiiv&utm_medium=newsletter&utm_campaign=mediamobilize&_bhlid=045dede5ad6eedcb96ec953f2fbac11a166a2243</a>via Sabine Hossenfelder's newsletter

Dr Mircea Zloteanu ☀️ 🌊🌴<a href="https://mastodon.social/tags/statstab" class="mention hashtag" rel="nofollow noopener" target="_blank">#statstab</a> #330 Encourage Playing with Data and Discourage Questionable Reporting PracticesThoughts: What are and aren't "Questionable Research Practices"? Where is the "grey area"? Interesting opinion piece.<a href="https://mastodon.social/tags/QRPs" class="mention hashtag" rel="nofollow noopener" target="_blank">#QRPs</a> <a href="https://mastodon.social/tags/exploratory" class="mention hashtag" rel="nofollow noopener" target="_blank">#exploratory</a> <a href="https://mastodon.social/tags/EDA" class="mention hashtag" rel="nofollow noopener" target="_blank">#EDA</a> <a href="https://mastodon.social/tags/posthoc" class="mention hashtag" rel="nofollow noopener" target="_blank">#posthoc</a> <a href="https://mastodon.social/tags/phacking" class="mention hashtag" rel="nofollow noopener" target="_blank">#phacking</a><a href="https://link.springer.com/article/10.1007/s11336-015-9445-1" rel="nofollow noopener" translate="no" target="_blank">https://link.springer.com/article/10.1007/s11336-015-9445-1</a>

Dr Mircea Zloteanu ☀️ 🌊🌴<a href="https://mastodon.social/tags/statstab" class="mention hashtag" rel="nofollow noopener" target="_blank">#statstab</a> #168 Large P-Values Cannot be Explained by Power AnalysisThoughts: "Researchers cannot “aim” for p = .05, not even with a careful, perfectly accurate, power analysis."<a href="https://mastodon.social/tags/research" class="mention hashtag" rel="nofollow noopener" target="_blank">#research</a> <a href="https://mastodon.social/tags/nhst" class="mention hashtag" rel="nofollow noopener" target="_blank">#nhst</a> <a href="https://mastodon.social/tags/pvalues" class="mention hashtag" rel="nofollow noopener" target="_blank">#pvalues</a> <a href="https://mastodon.social/tags/power" class="mention hashtag" rel="nofollow noopener" target="_blank">#power</a> <a href="https://mastodon.social/tags/QRPs" class="mention hashtag" rel="nofollow noopener" target="_blank">#QRPs</a> <a href="https://mastodon.social/tags/phacking" class="mention hashtag" rel="nofollow noopener" target="_blank">#phacking</a><a href="https://quentinandre.net/post/large-pvalues-and-power-analysis/" rel="nofollow noopener" translate="no" target="_blank">https://quentinandre.net/post/large-pvalues-and-power-analysis/</a>

Dr Mircea Zloteanu ☀️ 🌊🌴<a href="https://mastodon.social/tags/statstab" class="mention hashtag" rel="nofollow noopener" target="_blank">#statstab</a> #135 {p-checker} The one-for-all p-value analyzerThoughts: Why not try your hand at some tools for detecting publication bias (mileage may vary). Useful teaching demo tools.<a href="https://mastodon.social/tags/shinyapp" class="mention hashtag" rel="nofollow noopener" target="_blank">#shinyapp</a> <a href="https://mastodon.social/tags/r" class="mention hashtag" rel="nofollow noopener" target="_blank">#r</a> <a href="https://mastodon.social/tags/rstats" class="mention hashtag" rel="nofollow noopener" target="_blank">#rstats</a> <a href="https://mastodon.social/tags/phacking" class="mention hashtag" rel="nofollow noopener" target="_blank">#phacking</a> <a href="https://mastodon.social/tags/pvalues" class="mention hashtag" rel="nofollow noopener" target="_blank">#pvalues</a> <a href="https://mastodon.social/tags/NHST" class="mention hashtag" rel="nofollow noopener" target="_blank">#NHST</a> <a href="https://mastodon.social/tags/education" class="mention hashtag" rel="nofollow noopener" target="_blank">#education</a> <a href="https://mastodon.social/tags/edu" class="mention hashtag" rel="nofollow noopener" target="_blank">#edu</a><a href="https://shinyapps.org/apps/p-checker/" rel="nofollow noopener" translate="no" target="_blank">https://shinyapps.org/apps/p-checker/</a>

Bjørn SætrevikTable 1 of the "False-positive psychology" paper (Simmons, Nelson & Simonsohn 2011, <a href="https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2916240" rel="nofollow noopener" target="_blank">https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2916240</a>) estimate the false-positive rate of some questionable <a href="https://fediscience.org/tags/ResearchPractices" class="mention hashtag" rel="nofollow noopener" target="_blank">#ResearchPractices</a>, both alone and in combination (see attached figure). I remember reading somewhere that the authors later stated that the numbers were inaccurate and should have been somewhat higher. Does anyone have a reference for this? <a href="https://fediscience.org/tags/FalsePositivePsychology" class="mention hashtag" rel="nofollow noopener" target="_blank">#FalsePositivePsychology</a> <a href="https://fediscience.org/tags/QRP" class="mention hashtag" rel="nofollow noopener" target="_blank">#QRP</a> <a href="https://fediscience.org/tags/phacking" class="mention hashtag" rel="nofollow noopener" target="_blank">#phacking</a> <a href="https://fediscience.org/tags/HARKing" class="mention hashtag" rel="nofollow noopener" target="_blank">#HARKing</a> <a href="https://fediscience.org/tags/ReproducibiliTea" class="mention hashtag" rel="nofollow noopener" target="_blank">#ReproducibiliTea</a>

Karthik SrinivasanA nice hatchet job from the New Yorker! <a href="https://www.newyorker.com/magazine/2023/10/09/they-studied-dishonesty-was-their-work-a-lie" rel="nofollow noopener" translate="no" target="_blank">https://www.newyorker.com/magazine/2023/10/09/they-studied-dishonesty-was-their-work-a-lie</a>The story of morally and intellectually bankrupt superstar researchers who elevated what is/was/has always been a somewhat semi-decent, hardly scientific enterprise called behavioral economics (within social psychology) into a juggernaut of "just-so stories" that swayed nations, industries and academias to form absolutely horrendous worldviews and make decisions likewise. The field: The poor need to be nudged (read, extra inconvenience) to make rational decisions with marginally any value to them. Meanwhile, dear corporations, here are some tax write-offs and subsidies; you can also engage in open thievery. That's our way to nudge you to do the right thing, i.e., fudging with the accounts. This is how we social engineer (All this, based on absolutely moronic and morally dubious "studies". I am reminded of the evo-psych psychopaths). As if economics as it stands isn't dismal and unscientific enough, whether you legitimize with a couple Nobels or not, this is like "polishing a turd".Of course not much is likely to change in academia across fields as well. It will continue to seek and hire more such people who will bring the money, fame, and storytelling at the expense of hard work and reality. Knowledge building and intellectual work... what's that? <a href="https://neuromatch.social/tags/Dishonesty" class="mention hashtag" rel="nofollow noopener" target="_blank">#Dishonesty</a> <a href="https://neuromatch.social/tags/Academia" class="mention hashtag" rel="nofollow noopener" target="_blank">#Academia</a> <a href="https://neuromatch.social/tags/BehavioralEconomics" class="mention hashtag" rel="nofollow noopener" target="_blank">#BehavioralEconomics</a> <a href="https://neuromatch.social/tags/Management" class="mention hashtag" rel="nofollow noopener" target="_blank">#Management</a> <a href="https://neuromatch.social/tags/BusinessSchools" class="mention hashtag" rel="nofollow noopener" target="_blank">#BusinessSchools</a> <a href="https://neuromatch.social/tags/Economics" class="mention hashtag" rel="nofollow noopener" target="_blank">#Economics</a> <a href="https://neuromatch.social/tags/Nudge" class="mention hashtag" rel="nofollow noopener" target="_blank">#Nudge</a> <a href="https://neuromatch.social/tags/pHacking" class="mention hashtag" rel="nofollow noopener" target="_blank">#pHacking</a> <a href="https://neuromatch.social/tags/SocialPsychology" class="mention hashtag" rel="nofollow noopener" target="_blank">#SocialPsychology</a>

Research Network Digi-Oek.ch[en] Cheating in Science: Harvard "Honesty Scholar" May Have Been Caught in Dishonesty"... dishonesty can lead to creativity" - an interesting and somewhat amusing read. The New York Times: "Questions about a widely cited paper are the latest to be raised about methods used in <a href="https://social.tchncs.de/tags/behavioral" class="mention hashtag" rel="nofollow noopener" target="_blank">#behavioral</a> research."<a href="https://datacolada.org/111" rel="nofollow noopener" target="_blank">https://datacolada.org/111</a><a href="https://social.tchncs.de/tags/ResearchHighlights" class="mention hashtag" rel="nofollow noopener" target="_blank">#ResearchHighlights</a> <a href="https://social.tchncs.de/tags/honesty" class="mention hashtag" rel="nofollow noopener" target="_blank">#honesty</a> <a href="https://social.tchncs.de/tags/dishonesty" class="mention hashtag" rel="nofollow noopener" target="_blank">#dishonesty</a> <a href="https://social.tchncs.de/tags/phacking" class="mention hashtag" rel="nofollow noopener" target="_blank">#phacking</a> <a href="https://social.tchncs.de/tags/harking" class="mention hashtag" rel="nofollow noopener" target="_blank">#harking</a> <a href="https://social.tchncs.de/tags/dredging" class="mention hashtag" rel="nofollow noopener" target="_blank">#dredging</a> <a href="https://social.tchncs.de/tags/gino" class="mention hashtag" rel="nofollow noopener" target="_blank">#gino</a> <a href="https://social.tchncs.de/tags/harvard" class="mention hashtag" rel="nofollow noopener" target="_blank">#harvard</a> <a href="https://social.tchncs.de/tags/fraud" class="mention hashtag" rel="nofollow noopener" target="_blank">#fraud</a> <a href="https://social.tchncs.de/tags/cheating" class="mention hashtag" rel="nofollow noopener" target="_blank">#cheating</a> <a href="https://social.tchncs.de/tags/academic" class="mention hashtag" rel="nofollow noopener" target="_blank">#academic</a> <a href="https://social.tchncs.de/tags/datacolada" class="mention hashtag" rel="nofollow noopener" target="_blank">#datacolada</a>

Nicola Righetti“It’s a bit like seeing a rabbit shape in the clouds and then testing whether all clouds look like rabbits… using the same cloud. I hope you appreciate that you’re going to need some new clouds to test your theory.Any datapoint you use to inspire a theory or question can’t be used to test that same theory.”<a href="https://sciences.social/tags/phacking" class="mention hashtag" rel="nofollow noopener" target="_blank">#phacking</a> <a href="https://sciences.social/tags/statistics" class="mention hashtag" rel="nofollow noopener" target="_blank">#statistics</a> <a href="https://sciences.social/tags/rstats" class="mention hashtag" rel="nofollow noopener" target="_blank">#rstats</a><a href="https://towardsdatascience.com/the-most-powerful-idea-in-data-science-78b9cd451e72" rel="nofollow noopener" target="_blank">https://towardsdatascience.com/the-most-powerful-idea-in-data-science-78b9cd451e72</a><a href="https://a.gup.pe/u/sociology" class="u-url mention" rel="nofollow noopener" target="_blank">@sociology</a> <a href="https://a.gup.pe/u/communicationscholars" class="u-url mention" rel="nofollow noopener" target="_blank">@communicationscholars</a> <a href="https://a.gup.pe/u/rstats" class="u-url mention" rel="nofollow noopener" target="_blank">@rstats</a>

Anita Graser 🇪🇺🇺🇦🇬🇪"Do not chase better metrics. Chase better models."Excellent lighting talk on how spatial sampling affects the quality of data driven models by Luísa Lucchese<a href="https://youtu.be/q9dy2dQuTak" rel="nofollow noopener" target="_blank">https://youtu.be/q9dy2dQuTak</a>Chasing better <a href="https://fosstodon.org/tags/metrics" class="mention hashtag" rel="nofollow noopener" target="_blank">#metrics</a> is like <a href="https://fosstodon.org/tags/phacking" class="mention hashtag" rel="nofollow noopener" target="_blank">#phacking</a> <a href="https://fosstodon.org/tags/GeoAI" class="mention hashtag" rel="nofollow noopener" target="_blank">#GeoAI</a> <a href="https://fosstodon.org/tags/gischat" class="mention hashtag" rel="nofollow noopener" target="_blank">#gischat</a> <a href="https://fosstodon.org/tags/datascience" class="mention hashtag" rel="nofollow noopener" target="_blank">#datascience</a> <a href="https://fosstodon.org/tags/machinelearning" class="mention hashtag" rel="nofollow noopener" target="_blank">#machinelearning</a>

Carl T. BergstromIf you’re interested in <a href="https://fediscience.org/tags/statistics" class="mention hashtag" rel="nofollow noopener" target="_blank">#statistics</a>, <a href="https://fediscience.org/tags/Metascience" class="mention hashtag" rel="nofollow noopener" target="_blank">#Metascience</a>, <a href="https://fediscience.org/tags/openScience" class="mention hashtag" rel="nofollow noopener" target="_blank">#openScience</a>, <a href="https://fediscience.org/tags/pHacking" class="mention hashtag" rel="nofollow noopener" target="_blank">#pHacking</a>, etc., this is quite a good read. <a href="https://joebakcoleman.com/blog/2023/pcurve/" rel="nofollow noopener" target="_blank">https://joebakcoleman.com/blog/2023/pcurve/</a>

Isi Schneiderjust a little engineering mishap I guess, low probability to happen, but HUGE effect size 🤣🤣🤣 <a href="https://mstdn.science/tags/Science" class="mention hashtag" rel="nofollow noopener" target="_blank">#Science</a> <a href="https://mstdn.science/tags/pHacking" class="mention hashtag" rel="nofollow noopener" target="_blank">#pHacking</a>

Recent searches

Search options

Administered by:

Server stats:

#phacking