Paweł Lenartowicz
@freestylerscientist.pl
Aspiring statistician and meta-researcher, sailor and activists.
#PhilosophyofScience
#OpenScience
#Uncertanity
#PhilosophyofScience
#OpenScience
#Uncertanity
Ericsson & Simon (1980) Verbal reports as data.
8671 citations, and its impact and 'citation coverage' could be higher.
8671 citations, and its impact and 'citation coverage' could be higher.
November 10, 2025 at 7:52 PM
Ericsson & Simon (1980) Verbal reports as data.
8671 citations, and its impact and 'citation coverage' could be higher.
8671 citations, and its impact and 'citation coverage' could be higher.
I guess he did. Or at least he wrote a positive review of *Betrayers of the Truth*, a 1982 book about the ubiquity of fraud and manipulation in science.
October 19, 2025 at 8:24 PM
I guess he did. Or at least he wrote a positive review of *Betrayers of the Truth*, a 1982 book about the ubiquity of fraud and manipulation in science.
In case on detecting p-hacking, you might look at this: bsky.app/profile/free...
I’ve developed a new method for detecting publication bias in heterogeneous effects, with initial tests showing it outperforms existing approaches.
If you’d like to support a young independent scientist in advancing meta-science, please read, share, or review—your help makes a difference!
#MetaSci
If you’d like to support a young independent scientist in advancing meta-science, please read, share, or review—your help makes a difference!
#MetaSci
Likelihood Ratio Test for Publication Bias: Proof of Concept
Publication bias is a serious challenge to the integrity of scientific research and meta-analyses. And this article proposes a new method for estimating it.
freestylerscientist.pl
October 19, 2025 at 2:44 PM
In case on detecting p-hacking, you might look at this: bsky.app/profile/free...
This is especially true considering that the "Dunning-Kruger effect" was merely an analytic error on the part of the authors, not a real effect. www.sciencedirect.com/science/arti...
The Dunning-Kruger effect is (mostly) a statistical artefact: Valid approaches to testing the hypothesis with individual differences data
The Dunning-Kruger hypothesis states that the degree to which people can estimate their ability accurately depends, in part, upon possessing the abili…
www.sciencedirect.com
September 28, 2025 at 8:46 PM
This is especially true considering that the "Dunning-Kruger effect" was merely an analytic error on the part of the authors, not a real effect. www.sciencedirect.com/science/arti...
Thanks! I did not read this paper carefully, but have similar feeling, that this paper is published because of UCL in affiliation, not value of analysis itself.
August 22, 2025 at 8:40 PM
Thanks! I did not read this paper carefully, but have similar feeling, that this paper is published because of UCL in affiliation, not value of analysis itself.
How do you know it's p-hacked?
August 22, 2025 at 7:18 PM
How do you know it's p-hacked?
Reposted by Paweł Lenartowicz
(2) My 2024 article with Chris Donkin, in which we consider the advantages and disadvantages of confirmatory and exploratory hypothesis tests and conclude that exploratory tests can be more compelling than confirmatory tests.
Open Access: doi.org/10.1080/0951...
Open Access: doi.org/10.1080/0951...
August 22, 2025 at 6:43 AM
(2) My 2024 article with Chris Donkin, in which we consider the advantages and disadvantages of confirmatory and exploratory hypothesis tests and conclude that exploratory tests can be more compelling than confirmatory tests.
Open Access: doi.org/10.1080/0951...
Open Access: doi.org/10.1080/0951...
What about revealing reviewers identities, when paper is not published?
August 13, 2025 at 5:42 PM
What about revealing reviewers identities, when paper is not published?
Why USA, but not just Florida?
August 11, 2025 at 5:59 PM
Why USA, but not just Florida?
Last lobotomy was performed at least in 80', not in 1967.
August 11, 2025 at 3:32 PM
Last lobotomy was performed at least in 80', not in 1967.
I tried prove that you are not correct, but Chat GPT went wrong immediately (lol).
August 11, 2025 at 3:02 PM
I tried prove that you are not correct, but Chat GPT went wrong immediately (lol).
Are you really sure about that?
August 11, 2025 at 11:33 AM
Are you really sure about that?
If you correct an AI with faked date...
August 11, 2025 at 9:20 AM
If you correct an AI with faked date...
In 2 month, there should be update and final draft :)
August 10, 2025 at 10:04 PM
In 2 month, there should be update and final draft :)
Look also at freestylerscientist.pl/projects/lik... :)
Likelihood Ratio Test for Publication Bias - Paweł Lenartowicz
Likelihood Ratio Test for Publication Bias a Proof of Concept Abstract Publication bias poses a serious challenge to the integrity of scientific research and...
freestylerscientist.pl
August 10, 2025 at 9:36 PM
Look also at freestylerscientist.pl/projects/lik... :)
Yes, it's good for a significance test if there is a bias, but it's pretty bad at estimating it.
August 10, 2025 at 1:36 PM
Yes, it's good for a significance test if there is a bias, but it's pretty bad at estimating it.
Exactly!
I do not compare yet! I had to take a break from my academic work to earn a living.
I do not compare yet! I had to take a break from my academic work to earn a living.
August 10, 2025 at 1:34 PM
Exactly!
I do not compare yet! I had to take a break from my academic work to earn a living.
I do not compare yet! I had to take a break from my academic work to earn a living.
And Caliper Test www.nowpublishers.com/article/Deta...
now publishers - Do Statistical Reporting Standards Affect What Is Published? Publication Bias in Two Leading Political Science Journals
Publishers of Foundations and Trends, making research accessible
www.nowpublishers.com
August 10, 2025 at 11:58 AM
And Caliper Test www.nowpublishers.com/article/Deta...
Z-Curve, by @ulrichschimmack.bsky.social and @fbartos.bsky.social replicationindex.com/2025/07/08/z...
Z-Curve.3.0 Tutorial: Introduction - Replicability-Index
This post is Chapter 1. The R-code for this chapter can be found on my github: zcurve3.0/Tutorial.R.Script.Chapter1.R at main · UlrichSchimmack/zcurve3.0(the picture for this post shows a "finger-plot...
replicationindex.com
August 10, 2025 at 11:58 AM
Here is the LRBT. It is a method that I developed, and it has not been published yet. The final draft will probably be ready in two months. freestylerscientist.pl/projects/lik...
Likelihood Ratio Test for Publication Bias - Paweł Lenartowicz
Likelihood Ratio Test for Publication Bias a Proof of Concept Abstract Publication bias poses a serious challenge to the integrity of scientific research and...
freestylerscientist.pl
August 10, 2025 at 11:57 AM
Here is the LRBT. It is a method that I developed, and it has not been published yet. The final draft will probably be ready in two months. freestylerscientist.pl/projects/lik...
Both z-curve and LRBT (second link) do exactly that, the third method would be caliper test
The problem with p-curve is also that "implausible patterns arising from bias", is not as unplausible (it relies heavily on assumption that original data in analysed studies are normally distributed)
The problem with p-curve is also that "implausible patterns arising from bias", is not as unplausible (it relies heavily on assumption that original data in analysed studies are normally distributed)
August 10, 2025 at 11:54 AM
Both z-curve and LRBT (second link) do exactly that, the third method would be caliper test
The problem with p-curve is also that "implausible patterns arising from bias", is not as unplausible (it relies heavily on assumption that original data in analysed studies are normally distributed)
The problem with p-curve is also that "implausible patterns arising from bias", is not as unplausible (it relies heavily on assumption that original data in analysed studies are normally distributed)
OSF
osf.io
August 10, 2025 at 7:19 AM