Ingo Rohlfing
@ingorohlfing.bsky.social
I am here for all interesting and funny posts on the social sciences, broadly understood and including open science and meta science, academia, teaching and research. https://linktr.ee/ingorohlfing
Reposted by Ingo Rohlfing
For some research studies the optimal sample size should be estimated at 0
November 25, 2025 at 10:51 AM
For some research studies the optimal sample size should be estimated at 0
Reposted by Ingo Rohlfing
🚨 LSE Assistant Professor in Political Science 🚨
We’re hiring a tenure-track assistant professor - any area of empirical political science - to join our wonderful Government Dept @lsegovernment.bsky.social
Any questions, please reach out to me
📣 Please share! 📣
jobs.lse.ac.uk/Vacancies/W/...
We’re hiring a tenure-track assistant professor - any area of empirical political science - to join our wonderful Government Dept @lsegovernment.bsky.social
Any questions, please reach out to me
📣 Please share! 📣
jobs.lse.ac.uk/Vacancies/W/...
November 25, 2025 at 8:38 AM
🚨 LSE Assistant Professor in Political Science 🚨
We’re hiring a tenure-track assistant professor - any area of empirical political science - to join our wonderful Government Dept @lsegovernment.bsky.social
Any questions, please reach out to me
📣 Please share! 📣
jobs.lse.ac.uk/Vacancies/W/...
We’re hiring a tenure-track assistant professor - any area of empirical political science - to join our wonderful Government Dept @lsegovernment.bsky.social
Any questions, please reach out to me
📣 Please share! 📣
jobs.lse.ac.uk/Vacancies/W/...
Reposted by Ingo Rohlfing
A new article by Chloe Patton in #ResearchEvaluation shows how debates about #OpenScience often slip into absurdity – like demanding #replication from the #Humanities. You can’t replicate history, culture, or interpretation the way you replicate a physics experiment: doi.org/10.1093/rese...
November 23, 2025 at 8:43 AM
A new article by Chloe Patton in #ResearchEvaluation shows how debates about #OpenScience often slip into absurdity – like demanding #replication from the #Humanities. You can’t replicate history, culture, or interpretation the way you replicate a physics experiment: doi.org/10.1093/rese...
Reposted by Ingo Rohlfing
lads i found ANOTHER Table of Impossible Summary Statistics in a "green economics" paper 🚩
Can you see the problem?
Can you see the problem?
November 22, 2025 at 2:29 PM
lads i found ANOTHER Table of Impossible Summary Statistics in a "green economics" paper 🚩
Can you see the problem?
Can you see the problem?
Reposted by Ingo Rohlfing
"If we are unable to educate clinicians then merely persuading them to use CI's rather than p-values is to replace
the unthinking use of one technique with that of another."
A.P.Grieve (1992) Royal Statistical Society News and Notes, 18(7), 3-4.
the unthinking use of one technique with that of another."
A.P.Grieve (1992) Royal Statistical Society News and Notes, 18(7), 3-4.
November 18, 2025 at 6:29 PM
"If we are unable to educate clinicians then merely persuading them to use CI's rather than p-values is to replace
the unthinking use of one technique with that of another."
A.P.Grieve (1992) Royal Statistical Society News and Notes, 18(7), 3-4.
the unthinking use of one technique with that of another."
A.P.Grieve (1992) Royal Statistical Society News and Notes, 18(7), 3-4.
Reposted by Ingo Rohlfing
💫 Beginner-friendly #qualitative courses this Feb ⤵️
🔹Applications of Focus Groups @karenlumsden.bsky.social
🔹Case Study Research: Method & Practice @ingorohlfing.bsky.social
🔹Intro to Qualitative Comparative Analysis
📆 16–27 Feb 2026, Online
🐦 Early Bird until 5 Jan buff.ly/qoXc0pJ
#ecprms
🔹Applications of Focus Groups @karenlumsden.bsky.social
🔹Case Study Research: Method & Practice @ingorohlfing.bsky.social
🔹Intro to Qualitative Comparative Analysis
📆 16–27 Feb 2026, Online
🐦 Early Bird until 5 Jan buff.ly/qoXc0pJ
#ecprms
November 18, 2025 at 1:00 PM
💫 Beginner-friendly #qualitative courses this Feb ⤵️
🔹Applications of Focus Groups @karenlumsden.bsky.social
🔹Case Study Research: Method & Practice @ingorohlfing.bsky.social
🔹Intro to Qualitative Comparative Analysis
📆 16–27 Feb 2026, Online
🐦 Early Bird until 5 Jan buff.ly/qoXc0pJ
#ecprms
🔹Applications of Focus Groups @karenlumsden.bsky.social
🔹Case Study Research: Method & Practice @ingorohlfing.bsky.social
🔹Intro to Qualitative Comparative Analysis
📆 16–27 Feb 2026, Online
🐦 Early Bird until 5 Jan buff.ly/qoXc0pJ
#ecprms
Reposted by Ingo Rohlfing
Luckily, you don't need to choose b/c you can use them both! The framing of "p value bad"/"CI good" means that some people never really get that they're based on the same information & the same basic stat. philosophy (some of these points are made here: richarddmorey.medium.com/power-and-pr...)
Power and precision
Why the push for replacing “power” with “precision” is misguided
richarddmorey.medium.com
November 18, 2025 at 1:42 PM
Luckily, you don't need to choose b/c you can use them both! The framing of "p value bad"/"CI good" means that some people never really get that they're based on the same information & the same basic stat. philosophy (some of these points are made here: richarddmorey.medium.com/power-and-pr...)
Reposted by Ingo Rohlfing
#dataviz Very cool interactive - 7 sets Venn Diagram
128 color combinations from mixing 7 colors
moebio.com/research/sev...
128 color combinations from mixing 7 colors
moebio.com/research/sev...
7 sets Venn Diagram
moebio.com
November 18, 2025 at 2:34 PM
#dataviz Very cool interactive - 7 sets Venn Diagram
128 color combinations from mixing 7 colors
moebio.com/research/sev...
128 color combinations from mixing 7 colors
moebio.com/research/sev...
Reposted by Ingo Rohlfing
"academic publishing is dominated by profit-oriented, multinational companies for whom scientific knowledge is a commodity to be sold back to the academic community who created it... The dominant four collectively generated... $12 billion in profits between 2019 and 2024."
November 18, 2025 at 6:48 AM
"academic publishing is dominated by profit-oriented, multinational companies for whom scientific knowledge is a commodity to be sold back to the academic community who created it... The dominant four collectively generated... $12 billion in profits between 2019 and 2024."
Reposted by Ingo Rohlfing
This is an education problem, not a tool problem; and we don't want people simply moving from thinking p-values are magic to thinking confidence intervals are.
Next: Geoff Cumming @thenewstats.bsky.social with 'Statistical significance and p values: The researcher’s heroin'
* p values are highly unrealiable - don't trust them, don't use them!
www.thenewstatistics.com
tiny.cc/osfsigroulette
#IRICSydney
* p values are highly unrealiable - don't trust them, don't use them!
www.thenewstatistics.com
tiny.cc/osfsigroulette
#IRICSydney
November 18, 2025 at 8:45 AM
This is an education problem, not a tool problem; and we don't want people simply moving from thinking p-values are magic to thinking confidence intervals are.
Nature's 5 best science book picks are behind a paywall. www.nature.com/articles/d41... You can see the first three books completely, the fourth book has the review capped, the fifth is completely hidden. I am sure one can do a nice quasi-experimental study with this.
November 17, 2025 at 3:42 PM
Nature's 5 best science book picks are behind a paywall. www.nature.com/articles/d41... You can see the first three books completely, the fourth book has the review capped, the fifth is completely hidden. I am sure one can do a nice quasi-experimental study with this.
Just one level below sacrificing half a thumb mpra.ub.uni-muenchen.de/36335/1/Righ...
"Author changes name, publishes 10 papers in journals that banned him" https://retractionwatch.com/2025/11/10/author-changes-name-publishes-10-papers-in-journals-that-banned-him/
Of course...
Of course...
Author changes name, publishes 10 papers in journals that banned him
How to render a publishing ban moot? Change your surname and just keep submitting. That’s what happened in the case of Hashem Babaei, aka Hashem Gharababaei. In 2010, the Institution of Mechanical Engineers (IMechE), a professional society based in the U.K., banned the mechanical engineering researcher from the University of Guilan from submitting his work … Continue reading Author changes name, publishes 10 papers in journals that banned him
retractionwatch.com
November 16, 2025 at 1:48 PM
Just one level below sacrificing half a thumb mpra.ub.uni-muenchen.de/36335/1/Righ...
Reposted by Ingo Rohlfing
Synchronous Robustness Reports could explore implications of different analytical choices – but they could still suffer from bias. Hardwicke argues that preregistration is crucial to prevent it.
@tomhardwicke.bsky.social
@tomhardwicke.bsky.social
Risk of bias in robustness reports: https://osf.io/wj26e
November 14, 2025 at 2:54 PM
Synchronous Robustness Reports could explore implications of different analytical choices – but they could still suffer from bias. Hardwicke argues that preregistration is crucial to prevent it.
@tomhardwicke.bsky.social
@tomhardwicke.bsky.social
Statistical Intuition without Coding (or Teachers) [and w/o LLMs]
www.cambridge.org/core/journal...
I think this is a very useful approach when one does not want to teach coding in parallel. Simulating data and quantities of interest are insightful features, though at the expense of not using 1/
www.cambridge.org/core/journal...
I think this is a very useful approach when one does not want to teach coding in parallel. Simulating data and quantities of interest are insightful features, though at the expense of not using 1/
November 14, 2025 at 12:55 PM
Statistical Intuition without Coding (or Teachers) [and w/o LLMs]
www.cambridge.org/core/journal...
I think this is a very useful approach when one does not want to teach coding in parallel. Simulating data and quantities of interest are insightful features, though at the expense of not using 1/
www.cambridge.org/core/journal...
I think this is a very useful approach when one does not want to teach coding in parallel. Simulating data and quantities of interest are insightful features, though at the expense of not using 1/
Okay, quote tweets do not work when replies are restricted too (makes sense).
I was referring to a post where someone wrote he "heard numbers" that a larger share of papers is never read and an even larger share never cited. There was no source. One should just not dish out such number w/o sources.
I was referring to a post where someone wrote he "heard numbers" that a larger share of papers is never read and an even larger share never cited. There was no source. One should just not dish out such number w/o sources.
As replies are restricted, a quote-post it has to be: What are the sources for these numbers? For political science, we have ongoing work in progress showing it is less than 50% of papers (for about 100 journals). Still high, but not that high.
November 12, 2025 at 2:50 PM
Okay, quote tweets do not work when replies are restricted too (makes sense).
I was referring to a post where someone wrote he "heard numbers" that a larger share of papers is never read and an even larger share never cited. There was no source. One should just not dish out such number w/o sources.
I was referring to a post where someone wrote he "heard numbers" that a larger share of papers is never read and an even larger share never cited. There was no source. One should just not dish out such number w/o sources.
The use of LLMs for qualitative research is new in itself and worth exploring. Determining the value of AI for this purpose requires human evaluation, IMO. Delegating paper writing to an LLM and the review of papers is also a new element of science, worth exploring too, I guess. There is work on 2/
November 12, 2025 at 1:51 PM
The use of LLMs for qualitative research is new in itself and worth exploring. Determining the value of AI for this purpose requires human evaluation, IMO. Delegating paper writing to an LLM and the review of papers is also a new element of science, worth exploring too, I guess. There is work on 2/
Open Conference of AI Agents for Qualitative Research 2026 www.aiagents4qual.org "The 1st open conference where AI serves as both primary authors and reviewers of research papers" that use AI for qualitative research.
Maybe I am missing something here: this seems to overdo it with exploring AI 1/
Maybe I am missing something here: this seems to overdo it with exploring AI 1/
November 12, 2025 at 1:48 PM
Open Conference of AI Agents for Qualitative Research 2026 www.aiagents4qual.org "The 1st open conference where AI serves as both primary authors and reviewers of research papers" that use AI for qualitative research.
Maybe I am missing something here: this seems to overdo it with exploring AI 1/
Maybe I am missing something here: this seems to overdo it with exploring AI 1/
As replies are restricted, a quote-post it has to be: What are the sources for these numbers? For political science, we have ongoing work in progress showing it is less than 50% of papers (for about 100 journals). Still high, but not that high.
November 11, 2025 at 9:17 PM
As replies are restricted, a quote-post it has to be: What are the sources for these numbers? For political science, we have ongoing work in progress showing it is less than 50% of papers (for about 100 journals). Still high, but not that high.
The best time to slow down with publishing was yesterday. Unfortunately, publishing is likely to get accelerated by LLMs.
The race to churn out papers is a systemic problem.
Early career scholars are desperate to get more papers to compete in the academic job market. This can make it hard for faculty mentors hard to reduce their output unless they shrink their lab (which removes opportunities from next generation).
Early career scholars are desperate to get more papers to compete in the academic job market. This can make it hard for faculty mentors hard to reduce their output unless they shrink their lab (which removes opportunities from next generation).
November 11, 2025 at 9:14 PM
The best time to slow down with publishing was yesterday. Unfortunately, publishing is likely to get accelerated by LLMs.
Reposted by Ingo Rohlfing
Join our CSS department @gesis.org! Postdoc/senior researcher position, tenure track! All info at: www.gesis.org/institut/kar...
Details
GESIS Leibniz Institut für Sozialwissenschaften
www.gesis.org
November 11, 2025 at 2:48 PM
Join our CSS department @gesis.org! Postdoc/senior researcher position, tenure track! All info at: www.gesis.org/institut/kar...
Reposted by Ingo Rohlfing
Doing non-causal inference (and being explicit about it), yet using a causal word as second word in the title.
If you pay Nature € 10.690, they will publish this in Nature Ageing.
I can tell you what I think of that for free.
www.nature.com/articles/s43...
If you pay Nature € 10.690, they will publish this in Nature Ageing.
I can tell you what I think of that for free.
www.nature.com/articles/s43...
November 11, 2025 at 7:58 AM
Doing non-causal inference (and being explicit about it), yet using a causal word as second word in the title.
If you pay Nature € 10.690, they will publish this in Nature Ageing.
I can tell you what I think of that for free.
www.nature.com/articles/s43...
If you pay Nature € 10.690, they will publish this in Nature Ageing.
I can tell you what I think of that for free.
www.nature.com/articles/s43...
Harnessing generative artificial intelligence for teaching statistics in medical research: Strategies for accurate hypothesis testing
onlinelibrary.wiley.com/doi/full/10....
Using julius.ai, the article describes how to use LLMs in a stats class. I don't want to sound harsh, but it is not clear 1/
onlinelibrary.wiley.com/doi/full/10....
Using julius.ai, the article describes how to use LLMs in a stats class. I don't want to sound harsh, but it is not clear 1/
November 10, 2025 at 4:44 PM
Harnessing generative artificial intelligence for teaching statistics in medical research: Strategies for accurate hypothesis testing
onlinelibrary.wiley.com/doi/full/10....
Using julius.ai, the article describes how to use LLMs in a stats class. I don't want to sound harsh, but it is not clear 1/
onlinelibrary.wiley.com/doi/full/10....
Using julius.ai, the article describes how to use LLMs in a stats class. I don't want to sound harsh, but it is not clear 1/
Reposted by Ingo Rohlfing
This. First, it suggests reviews = meta-analyses, whereas 'reviews' actually encompasses loads of types, which may or may not involve meta-analysis. Second, it's not about needing fewer of them - we need fewer bad ones and more good ones
Also not sure about #2 - at least in public health and other policy relevant fields. Often what policymakers need most is a really good review article. If we want evidence to inform policy specifically reducing that type of study wouldn't be where I'd start.
I am slow to react to this recent Stockholm Declaration on scientific publishing. A lot of it sounds good, but I don't see how we get from here to there. I worry nothing substantial will happen until the cost disease kills the host.
November 7, 2025 at 1:08 PM
This. First, it suggests reviews = meta-analyses, whereas 'reviews' actually encompasses loads of types, which may or may not involve meta-analysis. Second, it's not about needing fewer of them - we need fewer bad ones and more good ones
Reposted by Ingo Rohlfing
We might ask whether we still need data visualization training when we have powerful LLMs to help us. Certainly, LLMs can help to optimize our code. But without a profound understanding of the produced code, we run the risk of creating figures that may look nice but that misrepresent our data. (6/n)
November 6, 2025 at 1:58 PM
We might ask whether we still need data visualization training when we have powerful LLMs to help us. Certainly, LLMs can help to optimize our code. But without a profound understanding of the produced code, we run the risk of creating figures that may look nice but that misrepresent our data. (6/n)