iiinova — where your digital self keeps evolving

Reposted by Mehmet Necip Tunc

moin syed @syeducation.bsky.social · 25/06/2026

New post! "Making Sense of Replications (and Some Nonsense about the Replication Crisis)," in which I suggest a journal club reading list of new papers on replication, and push back on claims that the replication crisis is fake. getsyeducated.substack.com/p/making-sen...

getsyeducated.substack.com

Making Sense of Replications (and Some Nonsense about the Replication Crisis)

It’s hard to overstate just how much the culture around replication in science has changed over the past 15 years.

34720

Reposted by Mehmet Necip Tunc

Daniel Lakens @lakens.bsky.social · 21/06/2026

New preprint: There is only one correct analysis. osf.io/preprints/ps... with @sajedehra.bsky.social and @mntunc.bsky.social We take a critical look at recent proposals for multiverse or many analyst procedures, and strongly argue that epistemic uncertainty should be reduced, not “embraced”. >

osf.io

OSF

55823

Reposted by Mehmet Necip Tunc

Daniel Lakens @lakens.bsky.social · 25/05/2026

New blog post: Evaluating Dr. Cuddy’s Claim that the Debunking of Power Posing is a Myth. daniellakens.blogspot.com/2026/05/eval... On an AI generated description of a non-existent study, incorrectly citing findings from studies, and the importance of scientific criticism.

daniellakens.blogspot.com

Evaluating Dr. Cuddy’s Claim that the Debunking of Power Posing is a Myth

In this blog post I will analyse the arguments that Dr. Amy Cuddy provided in a blog post “The "Power Posing Was Debunked" Myth: What the Re...

1517870

Reposted by Mehmet Necip Tunc

Lisa DeBruine @debruine.bsky.social · 25/05/2026

Debunking the debunking of the original debunking…. Debunkception!

2152

Reposted by Mehmet Necip Tunc

Sajedeh Rasti @sajedehra.bsky.social · 24/02/2026

"The benefits of and motivations behind large-team coordination in psychology" is finally out as preprint. In this paper, @lakens.bsky.social, Krist Vaesen, and I discussed the possible rewards of large-team collaborations that are common in coordinated research.

1125

Mehmet Necip Tunc @mntunc.bsky.social · 29/06/2025

In Philosophy of Nature, Feyerabend says that his position can be seen as exploring the implications of Levi-Strauss' ideas on myths for the phil of sci. I think it's a fascinating connection, especially given his indirect but significant influence on STS & the strong programme.

020

Mehmet Necip Tunc @mntunc.bsky.social · 28/05/2025

The paper you shared seems to be telling a different story, or am I missing something here?

110

Reposted by Mehmet Necip Tunc

Darren Dahly @statsepi.bsky.social · 24/05/2025

Im sorry for empowering trump to attack science by my asking people to use better data management and statistical practices. I take full responsibility for my actions and apologize to those who could so obviously see how my efforts would be responsible for ending American science.

5719

Reposted by Mehmet Necip Tunc

Edouard Machery @edouardmachery.bsky.social · 25/05/2025

The motto of some anti trumpers in science. these days: let 1000 wansink and staple bloom!

292

Mehmet Necip Tunc @mntunc.bsky.social · 22/05/2025

But look what Nagel says in that very book about standpoints and objectivity:

030

Mehmet Necip Tunc @mntunc.bsky.social · 22/05/2025

The View From Nowhere is the name of a book written by T. Nagel, often quoted to demonstrate the absurdity of the "positivist" position. The position attributed to Nagel is criticized as impossible and mythical, especially by those who emphasize the inevitability of different standpoints in science.

100

Mehmet Necip Tunc @mntunc.bsky.social · 04/05/2025

I follow up on this here:https://bsky.app/profile/mntunc.bsky.social/post/3lockwudb372r

000

Mehmet Necip Tunc @mntunc.bsky.social · 04/05/2025

12/ It should be emphasized that a scientific community committing to a specific alpha is exercising a form of discretion, since it can never be known with certainty how close these values are to the true optimum for long term error control. But discretion ≠ arbitrariness.

000

Mehmet Necip Tunc @mntunc.bsky.social · 04/05/2025

11/ Not really. As long as the specific value that these thresholds are supposed to take is defended in an epistemically principled way, there is rational disagreement, not arbitrariness. And rational disagreement in science is a feature not a bug.

100

Mehmet Necip Tunc @mntunc.bsky.social · 04/05/2025

10/ We admit that conventional evidential thresholds are **imperfect** solutions (or rather approximations) to an optimization problem. So, doesn't that mean the specific values are always open to debate and thus "arbitrary"?

100

Mehmet Necip Tunc @mntunc.bsky.social · 04/05/2025

9/ Widely shared evidential standards rooted in epistemic considerations are indispensable for collective pursuit of truth as without them there is no way to create a collectively accepted set of reference (evidential base).

100

Mehmet Necip Tunc @mntunc.bsky.social · 04/05/2025

..They aren’t perfect, but are deemed to be close enough to serve the long-run aim of controlling error and so converging on truth. This is also what makes it possible to learn from experiment in a piecemeal but socially organized fashion.

100

Mehmet Necip Tunc @mntunc.bsky.social · 04/05/2025

8/ Field conventions (like 0.05) approximate the epistemic optimum under (sometimes) conflicting epistemic aims such as discovery and justification...

100

Mehmet Necip Tunc @mntunc.bsky.social · 04/05/2025

7/ Alpha levels reflect an **epistemic optimization** problem. Scientists seek thresholds that maximize true positives while minimizing false ones, given sample sizes, measurement noise, and prior odds. That’s not arbitrary—that’s calibration.

100

Mehmet Necip Tunc @mntunc.bsky.social · 04/05/2025

6/ In fallibilist epistemology, justified belief doesn’t require certainty. So why should scientific inference require absolute thresholds? All thresholds are approximations—but that doesn’t make them unjustifiable or value-driven.

110

Mehmet Necip Tunc @mntunc.bsky.social · 04/05/2025

5/ There is a lesson to be learned from sorites paradox: vagueness ≠ meaninglessness. Concepts like heap are vague but still usable. “Statistical significance” is likewise vague at the boundary, but functionally essential. Fuzziness at the margins doesn’t nullify the category.

100

Mehmet Necip Tunc @mntunc.bsky.social · 04/05/2025

...then no amount of sand added individually, no matter how large N is, will form a heap. Similarly, no single increase in the third decimal of p-values can by itself indicate signal rather than noise.

100

Mehmet Necip Tunc @mntunc.bsky.social · 04/05/2025

4/ Critics claim 0.049 ≠ 0.051 is meaningless. This leads us into the **Sorites Paradox**: One grain of sand is not a heap. If we add one more sand to it, it still isn't - so if N sand is not a heap, and N+1 sand is not a heap…

100

Mehmet Necip Tunc @mntunc.bsky.social · 04/05/2025

3/ Yes, different fields use different thresholds (e.g., 5σ in physics, p < .05 in psych), but this isn't relativism. It's responsive adaptation to domain constraints. What’s shared is the logic of error control—not value judgments, but probability theory.

110

Mehmet Necip Tunc @mntunc.bsky.social · 04/05/2025

..as they are usually pre-specified, field-wide, and rooted in epistemic considerations like sample size, base rates, & discovery/accuracy trade-offs (albeit loosely or as an approximation).

100

Mehmet Necip Tunc @mntunc.bsky.social · 04/05/2025

2 / The meaning of arbitrariness here is ambiguous. Does it mean unfixed, inconsistent, unjustified? Standard α-levels cannot be described by any of these...

100

Mehmet Necip Tunc @mntunc.bsky.social · 04/05/2025

1/ But what about the counterargument that conventional evidential thresholds (like p < 0.05) are arbitrary? Doesn’t “God love .06 as much as .05”?

100

Mehmet Necip Tunc @mntunc.bsky.social · 03/05/2025

Thank you for your kind words. We would be glad to hear your takes on this.

010

Reposted by Mehmet Necip Tunc

Mehmet Necip Tunc @mntunc.bsky.social · 01/05/2025

1/ In our recent paper with @uygun_tunc (philsci-archive.pitt.edu/25196/), we defend the use of conventional alpha levels (e.g., 0.05, 0.01, or 5 sigma) in scientific inference. We challenge the claim that these thresholds should be set in a value-laden or context-dependent way. 🧵👇

philsci-archive.pitt.edu

297

Mehmet Necip Tunc @mntunc.bsky.social · 01/05/2025

20/ Final word: Let scientists decide what counts as evidence. Let society decide what to do with it. Don’t confuse acceptance with action. Neyman’s behaviorism in science is about inquiry, not policy.

100

Mehmet Necip Tunc @mntunc.bsky.social · 01/05/2025

19/ Epistemic decisions must be guided by internal standards—replication, robustness, predictive power—not external stakes. Mixing these contexts leads to strategic science, not trustworthy science.

100

Mehmet Necip Tunc @mntunc.bsky.social · 01/05/2025

18/ The right model: scientists manage epistemic risks (false positives, negatives); policymakers manage practical risks (health, safety, equity). Confusing the two collapses responsible governance.

100

Mehmet Necip Tunc @mntunc.bsky.social · 01/05/2025

17/ That creates a feedback loop where science no longer disciplines belief—it validates pre-existing preferences. This undermines the function of evidence entirely.

100

Mehmet Necip Tunc @mntunc.bsky.social · 01/05/2025

16/ Importantly, if scientists tailor evidence thresholds to social outcomes, we introduce circularity: values guide evidence which justifies action that was value-driven to begin with.

100

Mehmet Necip Tunc @mntunc.bsky.social · 01/05/2025

15/ Saying “we found p < 0.05” is not the same as saying “this is true,” or “you should act on it.” It’s saying: this finding meets our pre-defined criteria for signal over noise - so that we can use it in our next study as we control long term error rates.

100

Mehmet Necip Tunc @mntunc.bsky.social · 01/05/2025

14/ Also, scientific communication already encodes uncertainty. Terms like “suggests,” “might indicate,” “limited evidence” are epistemic hedges. Scientists rarely make strong categorical claims.

100

Mehmet Necip Tunc @mntunc.bsky.social · 01/05/2025

13/ Evidence doesn’t become stronger or weaker because consequences are serious. Raising the bar for “significance” due to social stakes isn’t cautious science—it’s epistemic distortion.

100

Mehmet Necip Tunc @mntunc.bsky.social · 01/05/2025

12/ Scientists are responsible for error control—not for managing risks in the world. Those risks must be evaluated by policymakers who apply the science, not by those who produce it.

100

Mehmet Necip Tunc @mntunc.bsky.social · 01/05/2025

11/ That being said, many, including inductive risk theorists, collapse this distinction. They argue that scientists bear moral responsibility for what happens if their accepted claims turn out false. But this confuses roles.

100

Mehmet Necip Tunc @mntunc.bsky.social · 01/05/2025

10/ That’s why we say scientific inference is a speech act, not a conduct. It expresses an epistemic commitment, not a decision to act. The moral or political responsibilities we assign to scientists should reflect this distinction.

100

Mehmet Necip Tunc @mntunc.bsky.social · 01/05/2025

9/ Therefore long term error rates control the overall reliability of published findings in science. To use the technical jargon, it regulates the pursuit-worthiness. So, the aim is to keep the type 1 errors low while not hurting the rate of discovery.

100

Mehmet Necip Tunc @mntunc.bsky.social · 01/05/2025

8/ Scientific “behavior” = integrating findings into the ongoing research process. A result becomes part of background knowledge, a premise in the next experiment—not a basis for action in the world.

100

Mehmet Necip Tunc @mntunc.bsky.social · 01/05/2025

7/ So what does “behavior” mean in Neyman’s framework when it comes to science? Not deploying a drug or regulating a chemical. It means: treat a result as a stepping stone in inquiry—use it in future studies.

100

Mehmet Necip Tunc @mntunc.bsky.social · 01/05/2025

6/ However, in science rejecting a null hypothesis (H₀) isn’t endorsing policy or a specific conduct. It’s saying: “Given our data and pre-specified error rates, this outcome counts as significant.” That’s all.

100

Mehmet Necip Tunc @mntunc.bsky.social · 01/05/2025

5/ Yet, the same duality causes a lot of confusion when applied to scientific inference. If alpha levels guide action, why don’t we adjust them based on the (social, political) consequences of these actions?

100

Mehmet Necip Tunc @mntunc.bsky.social · 01/05/2025

4/ In applied context, such as industrial quality control, this duality creates no specific tension. The objective is dictated by the actual constraints, such as profitability, so the significance thresholds unproblematically reflect those in a case to case basis.

100

Mehmet Necip Tunc @mntunc.bsky.social · 01/05/2025

3/ Statistical inference, especially the Neyman-Pearson framework, talks about "inductive behavior" or "decision rules"—these have interesting double quality: behavioral and epistemic. They regulate error through long term action.

100

Mehmet Necip Tunc @mntunc.bsky.social · 01/05/2025

2/ Context-dependent evidential threshold arguments often ignore a crucial distinction: epistemic vs practical decisions. In science, accepting a hypothesis ≠ physically acting on it. Confusing the two leads to serious category mistakes in reasoning.

120

Mehmet Necip Tunc @mntunc.bsky.social · 01/05/2025

philsci-archive.pitt.edu

297