Pierre-Yves de Müllenheim

It can be so frustrating to read STS research. Here are Bartscherer and Reinhart osf.io/rbyt6_v1/ Look at their flawed logic: Replication is *used as a career strategy*. The evidence? People were not known for earlier research. The clear confound? ECR’s drove the replication movement! 1/x

Measures of Central Tendency for an Asymmetric Distribution, and Confidence Intervals – Statistical Thinking There are three widely applicable measures of central tendency for general continuous distributions: the mean, median, and pseudomedian. Each measure has its own advantages and disadvantages, and the ...

New blog article in Statistical Thinking: www.fharrell.com/post/aci/ #Statistics #StatsSky #RStats

Cover page for the manuscript: Morey, R. D., & Davis-Stober, C. P. (2025). On the poor statistical properties of the P-curve meta-analytic procedure. Journal of the American Statistical Association, 1–19. https://doi.org/10.1080/01621459.2025.2544397

Abstract for the paper: The P-curve (Simonsohn, Nelson, & Simmons, 2014; Simonsohn, Simmons, & Nelson, 2015) is a widely-used suite of meta-analytic tests advertised for detecting problems in sets of studies. They are based on nonparametric combinations of p values (e.g., Marden, 1985) across significant (p < .05) studies and are variously claimed to detect “evidential value”, “lack of evidential value”, and “left skew” in p values. We show that these tests do not have the properties ascribed to them. Moreover, they fail basic desiderata for tests, including admissibility and monotonicity. In light of these serious problems, we recommend against the use of the P-curve tests.

Paper drop, for anyone interested in #metascience, #statistics, or #metaanalysis! @clintin.bsky.social and I show in a new paper in JASA that the P-curve, a popular forensic meta-analysis method, has deeply undesirable statistical properties. www.tandfonline.com/doi/full/10.... 1/?

Not Even Scientists Can Easily Explain P-values P-values have taken quite a beating lately. These widely used and commonly misapplied statistics have been blamed for giving a veneer of legitimacy to dodgy stu…

Can you make it easier than "the probability of getting results at least as extreme as the ones you observed, given that the null hypothesis is correct"?

Nope... this may look quite technical but I would say that every definition that would be more simple would be likely incorrect. Maybe some concepts like p values cannot really be understood without making some efforts to learn basics, here in statistics.

Finally got around to incorporating all Github feedback on my open access textbook, making around 20 minor improvements, and updating some references. 17 chapters of state of the art stats and methods education, freely available for any course you teach. lakens.github.io/statistical_...

Julia works quite well through Nix, but I need to do some further testing and make sure that the top, say, 100 packages build correctly for the list of dates I choose to support (rix::available_dates()). This will become my next milestone for rix.

Researchers should not make claims based on exploratory analyses because the probability that these claims are wrong can be unacceptably high.

doi.org/10.1080/0264...

Research Methods Module Videos - YouTube These videos are some of the online content that I prepared shortly before leaving Solent University for a 2nd year (level 5 in UK terminology) module called...

So here you go... the brief online video content for the research methods module I used to teach to undergraduate sports therapy students as a playlist:

We are delighted to now be part of the Free Journal Network, a community of free-to-read and free-to-publish journals like ours! freejournals.org/current-memb...

New pre-print out!

🔬In this article, we explain why our clinical trial to improve the automatic tendency to approach PA failed and offer recommendations to help future studies maximise patient acceptability.
doi.org/10.1101/2025...
@labosens.bsky.social @bcheval.bsky.social

In all my years using #RStats, I’ve seen this kind of posts often. I think that R has evolved past the "traditional" way of using it which was installing it system-wide, and then installing packages system-wide. That worked, because there were less packages, but also because the kind of projects 1/3

Air, an extremely fast R formatter We are thrilled to announce Air, a new R formatter.

@lionelhenry.bsky.social and I are so excited to finally announce Air - an extremely fast R code formatter! 🎉

With Air, you'll never need to worry about styling your #rstats code ever again. All you need to do is save, and Air takes care of the rest.

www.tidyverse.org/blog/2025/02...

Now published in PeerJ (@PeerJLife is a PCI friendly journal) without further peer-review and with a few minor adjustments (abstract, figure 5 correction, typos). peerj.com/articles/18885 #Cardiology #Rehabilitation #SportsMedicine

Defense Against Dishonest Charts This is a guide to protect ourselves and to preserve what is good about turning data into visual things.

Dishonest charts are coming from all directions, and they're only going to hit harder. Unless we push back. So, an interactive guide: Defense Against Dishonest Charts flowingdata.com/projects/dis...

PRISMA2020: Make Interactive 'PRISMA' Flow Diagrams Systematic reviews should be described in a high degree of methodological detail. The 'PRISMA' Statement calls for a high level of reporting detail in systematic reviews and meta-analyses. An integral...

The 'PRISMA2020' R package can used to make both static and interactive 'PRISMA' Flow Diagrams. Thanks to @heeminkang.bsky.social for the tip! cran.r-project.org/web/packages...

Our new #preprint led by @bcheval.bsky.social is out: Neural correlates of approach-avoidance tendencies toward physical activity & sedentary stimuli: an MRI study. doi.org/10.1101/2025...

Here are some 2024 highlights of PCI and Peer Community Journal.📊 Grateful to everyone who contributed to this success🦸. We're excited and prepared for an amazing 2025🤩!

Forgot to say, we'll be submitting this to @psych.peercommunityin.org once it launches, but we'd love any feedback form people before then if they have any 😊

New Sci Rep paper with odd-looking figures that don't look like any image conversion issue I've seen before. Has some AI vibes (fonts, line spacing of legend), but there are likely better explanations. Any thoughts?

@nature.com pls investigate (who copy-edited this)

www.nature.com/articles/s41...

How to move more and reduce sedentary behavior in daily life? 💡

Since September, during my teleworking days, I’ve covered 720 km (the equivalent of over 17 marathons), taken 1.2 million steps, and burned 50,000 Kcal — all during my working hours!

A list of 1,360 predatory publishers 😮
predatoryjournals.org/publisher-li...

Brave decision by JHE Editorial Board.
Hopefully there are transparent, open and even free initiatives out there for human evolution scientific publications.

#rstat note to myself:
If you want to convert a factor variable that contains numbers between quotes (eg, "1", "3", etc) to a numeric variable while keeping the original numbers:

DO:
as.numeric(as.character(my_var))

NOT DO:
as.numeric(my_var)

The 2nd option returns variable level numbers!

It was about time…🕰️

Sharing data along with an article is important for transparency/repro. A nice consequency is that it allows teachers to propose interesting examples to students for improving/assessing data analysis skills. So thanks to Communications in Kinesiology & @pci-hms.bsky.social for their policies!

Call for Help: As AE for JSS, I am looking for reviewers for a neat paper (and #Rstats package) with low/mid/high precision float support, something of interest for (very) large models and machine learning. If you,dear reader, could review this please get in touch via email. Thanks in advance.

There are more authors on this paper than cases of CTE in the manuscript.

We’re encouraging meaningful commentary with $1,000 rewards for selected PubPeer comments. bsky.app/profile/pubp...

Now imagine if @hhmi.bsky.social , NIH, NSF, etc also awarded outstanding public reviews.

These could boost careers and CVs while building a vital layer of scientific evaluation.