Gun to your head, what would you consider a sensible default value for the smallest effect size of interest? I.e. what is the largest effect you would consider ridiculously small, regardless of justification?
I'll go first 🙂🔫 Pearson's r = 0.01 is ridiculously small

No single practice is perfect. There *are* limitations to everything, and preregs are not immune to that.

But we don't debate those. Instead there is this cyclical thing where people write the most repetitive boring misconceptions and others have to correct them again and again

Thanks for citing my 2022 Sample Size Justification 1000 times. online.ucpress.edu/collabra/artic…

This is a good moment to tell you that you (in general) suck at citing my paper. I made a whole app to help you write a solid justification. Please use it.

shiny.ieis.tue.nl/sample_size_ju….

As recruiting season proceeds, I find myself pondering this paper that discusses (and models) conceptions of researcher contribution. Some researchers are not seen as 'the rockest of rock stars' but contribute by make everyone around them better.

journals.sagepub.com/doi/full/10.11…

For interested folks, here is preprint (w/ Donald Lynam Nate Phillips Josh Miller) that provides a response to the recent article by David Klonsky (which I've included below in the thread) where he shares concerns over preregistration becoming a norm in the field.

Colin Vize just posted a preprint on preregistration by himself, me, Nate Phillips, and Josh Miller that is a reply to a recent critique of preregistration by David Klonsky. Colin is a man of few words, but I am not.
1/15
osf.io/preprints/psya…

Late Breaking News! A student project from my course Human-AI Interaction turned into a LBW paper at CHI24. Our user study shows that AI literacy level influences people's preferences for different XAI methods, including a novel Visual Map method. dl.acm.org/doi/10.1145/36…

Hi peeps!

Moritz Heene, Ben Domingue, Marcos Jiménez, and I wrote an uncomplicated introduction to representational measurement theory (RMT).

You can read the preprint here: osf.io/preprints/psya…

1/3

Accepted for publication at IRSP, automatically published with PCIRR Stage 2 endorsement:

'Licensing via Credentials: Replication Registered Report of Monin and Miller (2001) with Extensions'

Failed replication.

Read:
researchgate.net/publication/37…

Chapter 5 of my book 'A Student's Guide to Open Science: Using the Replication Crisis to Reform Psychology' is now open access via PsyArXiv!

This chapter is arguably the most important: a hands-on guide to implementing open science in your workflow!

osf.io/preprints/psya…

Thanks for citing my 2022 Sample Size Justification 1000 times. online.ucpress.edu/collabra/artic…

This is a good moment to tell you that you (in general) suck at citing my paper. I made a whole app to help you write a solid justification. Please use it.

shiny.ieis.tue.nl/sample_size_ju….

Tomorrow we will record a Nullius In Verba podcast on 'eponymous laws that vaguely hint at a quantitative relationship but are actually more like a proverb'. We have a decent list, but what is your favorite?

Slides and a talk with a discussion about inference on the multiverse ;)

It is time to once again recommend Glenn Schafer's book A Mathematical Theory of Evidence.

jstor.org/stable/j.ctv10…

'An adequate summary of the impact of the evidence on a particular proposition A must include at least two items of information: a report on how well A is supported and a report on how well its negation ¬A is supported.'
Shafer, 1976, A Mathematical Theory of Evidence

Another day, another poll showing half of the respondents are not trained well in statistics. Well played, Carlisle Rainey!

Also, here is my paper about why p-values are not measures of evidence osf.io/7ng4w/download/

What does it mean to say ‘preregistration is a plan, not a prison’? In my recent paper on When and How to Deviate From a Preregistration online.ucpress.edu/collabra/artic… I explain the value of the plan, but also when to deviate from it. > 🧵

Software Engineering cluster at TU Eindhoven is looking for a new Assistant Professor jobs.tue.nl/en/vacancy/ass… #academics #facultyjobs RT = ♥️

Have been buying some new records out of nostalgia. But I also hope this counts as a classic :)

I should also say Daniël Lakens's book:
lakens.github.io/statistical_in…
(sorry I forgot that one)