John V. Kane
@uptonorwell.bsky.social
Political Scientist. Professor at NYU’s Center for Global Affairs. Experiments, data analysis, guitar, drums, fan of comedy. Make guides for @statacorp.bsky.social users. Two boys and exhausted all the time. More at www.johnvkane.com
Stata users: if you aren't already, definitely start using bookmarks to create headings and sub-headings in your .do files. Such a big help! **# in a .do file creates a heading; **## = subheading; **### = sub-subheading. Then double-click on them to jump right to that section 👍
November 27, 2025 at 2:57 AM
Stata users: if you aren't already, definitely start using bookmarks to create headings and sub-headings in your .do files. Such a big help! **# in a .do file creates a heading; **## = subheading; **### = sub-subheading. Then double-click on them to jump right to that section 👍
Wow, super helpful--thx, David! I don't usually run models like this in my own work, so there's a lot more I need to learn (looks like this book will help!).
In that graph I posted, it's clear that residuals are positively correlated, which is why I thought to cluster. But this article says no 🤯
In that graph I posted, it's clear that residuals are positively correlated, which is why I thought to cluster. But this article says no 🤯
November 20, 2025 at 5:48 PM
Wow, super helpful--thx, David! I don't usually run models like this in my own work, so there's a lot more I need to learn (looks like this book will help!).
In that graph I posted, it's clear that residuals are positively correlated, which is why I thought to cluster. But this article says no 🤯
In that graph I posted, it's clear that residuals are positively correlated, which is why I thought to cluster. But this article says no 🤯
Always amazed how, each semester, teaching data analysis inevitably teaches me something new.
This week I learned that spatially correlated residuals in FE models (w/ data from Bailey's great book) probably doesn't require clustering SEs. Thx to @nickchk.com's AMAZING website--what a resource! 🙏
This week I learned that spatially correlated residuals in FE models (w/ data from Bailey's great book) probably doesn't require clustering SEs. Thx to @nickchk.com's AMAZING website--what a resource! 🙏
November 20, 2025 at 2:25 PM
Always amazed how, each semester, teaching data analysis inevitably teaches me something new.
This week I learned that spatially correlated residuals in FE models (w/ data from Bailey's great book) probably doesn't require clustering SEs. Thx to @nickchk.com's AMAZING website--what a resource! 🙏
This week I learned that spatially correlated residuals in FE models (w/ data from Bailey's great book) probably doesn't require clustering SEs. Thx to @nickchk.com's AMAZING website--what a resource! 🙏
Amen to this, Andrew!
November 2, 2025 at 11:35 PM
Amen to this, Andrew!
Halloween fun fact 🎃: "Skeleton" has its own IMDB page. 💀
In fact, it stars in one of my favorite horror movies as a kid: House on Haunted Hill (1959 version). Not necessarily the best performance in the film, but also not bad given that it lacks a central nervous system.
In fact, it stars in one of my favorite horror movies as a kid: House on Haunted Hill (1959 version). Not necessarily the best performance in the film, but also not bad given that it lacks a central nervous system.
October 31, 2025 at 1:11 PM
Halloween fun fact 🎃: "Skeleton" has its own IMDB page. 💀
In fact, it stars in one of my favorite horror movies as a kid: House on Haunted Hill (1959 version). Not necessarily the best performance in the film, but also not bad given that it lacks a central nervous system.
In fact, it stars in one of my favorite horror movies as a kid: House on Haunted Hill (1959 version). Not necessarily the best performance in the film, but also not bad given that it lacks a central nervous system.
🚨 I created a new guide on how to make nice dot charts. These can be really informative, yet remain weirdly underutilized. The examples are with @statacorp.bsky.social but the logic applies to all
software. Link in thread. 👇
Made it mostly for my students but hope you find it useful, too! 😁
software. Link in thread. 👇
Made it mostly for my students but hope you find it useful, too! 😁
October 15, 2025 at 9:36 PM
🚨 I created a new guide on how to make nice dot charts. These can be really informative, yet remain weirdly underutilized. The examples are with @statacorp.bsky.social but the logic applies to all
software. Link in thread. 👇
Made it mostly for my students but hope you find it useful, too! 😁
software. Link in thread. 👇
Made it mostly for my students but hope you find it useful, too! 😁
A short-and-sweet guide to developing research questions published in @nature.com.
Lots of good advice here 👍
Lots of good advice here 👍
October 8, 2025 at 3:35 PM
A short-and-sweet guide to developing research questions published in @nature.com.
Lots of good advice here 👍
Lots of good advice here 👍
Good question! It's looking like those with post-graduate degrees are the only ones with a (albeit narrow) plurality answering correctly.
But even still, it's a surprisingly small % 😕
But even still, it's a surprisingly small % 😕
July 29, 2025 at 3:16 AM
Good question! It's looking like those with post-graduate degrees are the only ones with a (albeit narrow) plurality answering correctly.
But even still, it's a surprisingly small % 😕
But even still, it's a surprisingly small % 😕
Inflation was a big issue--maybe even *the* issue--in 2024.
But do citizens understand how to interpret inflation rates?
Using some new data from @verasight.bsky.social, the answer seems to be largely: no.
On top of that, Republicans show significantly less understanding than Democrats.
But do citizens understand how to interpret inflation rates?
Using some new data from @verasight.bsky.social, the answer seems to be largely: no.
On top of that, Republicans show significantly less understanding than Democrats.
July 28, 2025 at 4:03 PM
Inflation was a big issue--maybe even *the* issue--in 2024.
But do citizens understand how to interpret inflation rates?
Using some new data from @verasight.bsky.social, the answer seems to be largely: no.
On top of that, Republicans show significantly less understanding than Democrats.
But do citizens understand how to interpret inflation rates?
Using some new data from @verasight.bsky.social, the answer seems to be largely: no.
On top of that, Republicans show significantly less understanding than Democrats.
Truly wild to see that this just hit 300 citations on GS 🤯
It's a good lesson for PhD students--this started as a short homework assignment for a class on causal inference. At first I just wanted to know if MC placement matters. But over time, it turned into much more & led to several more papers.
It's a good lesson for PhD students--this started as a short homework assignment for a class on causal inference. At first I just wanted to know if MC placement matters. But over time, it turned into much more & led to several more papers.
July 21, 2025 at 5:16 PM
Truly wild to see that this just hit 300 citations on GS 🤯
It's a good lesson for PhD students--this started as a short homework assignment for a class on causal inference. At first I just wanted to know if MC placement matters. But over time, it turned into much more & led to several more papers.
It's a good lesson for PhD students--this started as a short homework assignment for a class on causal inference. At first I just wanted to know if MC placement matters. But over time, it turned into much more & led to several more papers.
Thanks so much, Hans!
Yep, empirically, a candidate whose party has already served two terms may have a slight disadvantage—but it doesn’t seem to be inherent.
That disadvantage, though, pales in comparison to the advantage held by 1-term incumbents (historically speaking) 👍
Yep, empirically, a candidate whose party has already served two terms may have a slight disadvantage—but it doesn’t seem to be inherent.
That disadvantage, though, pales in comparison to the advantage held by 1-term incumbents (historically speaking) 👍
July 15, 2025 at 2:11 AM
Thanks so much, Hans!
Yep, empirically, a candidate whose party has already served two terms may have a slight disadvantage—but it doesn’t seem to be inherent.
That disadvantage, though, pales in comparison to the advantage held by 1-term incumbents (historically speaking) 👍
Yep, empirically, a candidate whose party has already served two terms may have a slight disadvantage—but it doesn’t seem to be inherent.
That disadvantage, though, pales in comparison to the advantage held by 1-term incumbents (historically speaking) 👍
My article on incumbency advantage in presidential elections is now out w/ issue and page numbers 🥳
Key finding: Being an incumbent president comes w/ inherent electoral advantages after 1 term. After 2 terms, a party is not inherently *dis*advantaged--just no more incumbency advantage.
Key finding: Being an incumbent president comes w/ inherent electoral advantages after 1 term. After 2 terms, a party is not inherently *dis*advantaged--just no more incumbency advantage.
July 14, 2025 at 6:25 PM
My article on incumbency advantage in presidential elections is now out w/ issue and page numbers 🥳
Key finding: Being an incumbent president comes w/ inherent electoral advantages after 1 term. After 2 terms, a party is not inherently *dis*advantaged--just no more incumbency advantage.
Key finding: Being an incumbent president comes w/ inherent electoral advantages after 1 term. After 2 terms, a party is not inherently *dis*advantaged--just no more incumbency advantage.
A look at partisanship in the U.S. right now--what partisans will tolerate, what they won't, and where the parties differ.
(From an ongoing project with @iganson.bsky.social; n = ~1,100; graph made in @statacorp.bsky.social 😊)
(From an ongoing project with @iganson.bsky.social; n = ~1,100; graph made in @statacorp.bsky.social 😊)
June 18, 2025 at 4:35 PM
A look at partisanship in the U.S. right now--what partisans will tolerate, what they won't, and where the parties differ.
(From an ongoing project with @iganson.bsky.social; n = ~1,100; graph made in @statacorp.bsky.social 😊)
(From an ongoing project with @iganson.bsky.social; n = ~1,100; graph made in @statacorp.bsky.social 😊)
Hard to express what a relief it is to have just submitted the first complete draft of my book manuscript to @ucpress.bsky.social 😮💨🥳
5+ years in the making.
249 pages.
82,000 words.
0 ChatGPT. 😉
A million thanks to all those who gave me guidance and inspired me along the way! 🙏
5+ years in the making.
249 pages.
82,000 words.
0 ChatGPT. 😉
A million thanks to all those who gave me guidance and inspired me along the way! 🙏
June 3, 2025 at 5:00 PM
Hard to express what a relief it is to have just submitted the first complete draft of my book manuscript to @ucpress.bsky.social 😮💨🥳
5+ years in the making.
249 pages.
82,000 words.
0 ChatGPT. 😉
A million thanks to all those who gave me guidance and inspired me along the way! 🙏
5+ years in the making.
249 pages.
82,000 words.
0 ChatGPT. 😉
A million thanks to all those who gave me guidance and inspired me along the way! 🙏
For those currently writing and/or grading papers, always good to keep this distinction in mind
April 26, 2025 at 5:01 PM
For those currently writing and/or grading papers, always good to keep this distinction in mind
It makes sense, yep. Unfortunately, though, people can easily forget over time...
BUT, per your idea about taking actual econ courses, there is definitely some evidence that those with higher education are better on this question
BUT, per your idea about taking actual econ courses, there is definitely some evidence that those with higher education are better on this question
April 23, 2025 at 7:09 PM
It makes sense, yep. Unfortunately, though, people can easily forget over time...
BUT, per your idea about taking actual econ courses, there is definitely some evidence that those with higher education are better on this question
BUT, per your idea about taking actual econ courses, there is definitely some evidence that those with higher education are better on this question
Interesting idea, yes, and makes perfect sense.
But alas, it doesn't really look like older generations are answering this much better than younger generations. Surprising...
But alas, it doesn't really look like older generations are answering this much better than younger generations. Surprising...
April 23, 2025 at 6:50 PM
Interesting idea, yes, and makes perfect sense.
But alas, it doesn't really look like older generations are answering this much better than younger generations. Surprising...
But alas, it doesn't really look like older generations are answering this much better than younger generations. Surprising...
Do most people understand how inflation works? That drops in the inflation rate don't mean drops in prices? 🤔
No. And misunderstanding is significantly associated with vote choice in 2024.
Thanks to @verasight.bsky.social and @peterenns.bsky.social
for letting me field this question 🙏
No. And misunderstanding is significantly associated with vote choice in 2024.
Thanks to @verasight.bsky.social and @peterenns.bsky.social
for letting me field this question 🙏
April 22, 2025 at 7:34 PM
Do most people understand how inflation works? That drops in the inflation rate don't mean drops in prices? 🤔
No. And misunderstanding is significantly associated with vote choice in 2024.
Thanks to @verasight.bsky.social and @peterenns.bsky.social
for letting me field this question 🙏
No. And misunderstanding is significantly associated with vote choice in 2024.
Thanks to @verasight.bsky.social and @peterenns.bsky.social
for letting me field this question 🙏
Had a great time presenting in @stecula.bsky.social 's PhD class on null results in experiments. Thank you again, Dom!
If anyone who teaches experiments would like me to present for a course in the Fall, just email or DM me--happy to do it! 😁
If anyone who teaches experiments would like me to present for a course in the Fall, just email or DM me--happy to do it! 😁
April 21, 2025 at 7:14 PM
Had a great time presenting in @stecula.bsky.social 's PhD class on null results in experiments. Thank you again, Dom!
If anyone who teaches experiments would like me to present for a course in the Fall, just email or DM me--happy to do it! 😁
If anyone who teaches experiments would like me to present for a course in the Fall, just email or DM me--happy to do it! 😁
Had a great time presenting on null / non-significant results in survey experiments to Samara Klar's Ph.D. class at University of Arizona! 😁
🚨If you teach a course on experiments in the Fall and would like me to present, feel free to email/DM me--I'd be happy to do it!
polisky
🚨If you teach a course on experiments in the Fall and would like me to present, feel free to email/DM me--I'd be happy to do it!
polisky
April 10, 2025 at 1:54 PM
Had a great time presenting on null / non-significant results in survey experiments to Samara Klar's Ph.D. class at University of Arizona! 😁
🚨If you teach a course on experiments in the Fall and would like me to present, feel free to email/DM me--I'd be happy to do it!
polisky
🚨If you teach a course on experiments in the Fall and would like me to present, feel free to email/DM me--I'd be happy to do it!
polisky
Next, our key hypothesis: what might reduce BB is if partisans see the OTHER party engaging in BB.
We theorize that this makes bias easier to recognize and, b/c the other party is doing it, renders it a “bad” thing to do.
Using 2 pre-registered survey experiments, we find support for this. 👍
We theorize that this makes bias easier to recognize and, b/c the other party is doing it, renders it a “bad” thing to do.
Using 2 pre-registered survey experiments, we find support for this. 👍
March 28, 2025 at 3:07 PM
Next, our key hypothesis: what might reduce BB is if partisans see the OTHER party engaging in BB.
We theorize that this makes bias easier to recognize and, b/c the other party is doing it, renders it a “bad” thing to do.
Using 2 pre-registered survey experiments, we find support for this. 👍
We theorize that this makes bias easier to recognize and, b/c the other party is doing it, renders it a “bad” thing to do.
Using 2 pre-registered survey experiments, we find support for this. 👍
We then find that among Trump supporters (but not Biden supporters), BB is associated with higher support for a leader breaking rules and higher support for political violence.
March 28, 2025 at 3:07 PM
We then find that among Trump supporters (but not Biden supporters), BB is associated with higher support for a leader breaking rules and higher support for political violence.
First, we introduce the idea of “blatant bias” (BB)—we define it as support for a politician *regardless* of what the politician does.
We find similar levels of BB among both Biden & Trump supporters, though Trump supporters showed a large jump in the most recent wave of @prripoll.bsky.social data.
We find similar levels of BB among both Biden & Trump supporters, though Trump supporters showed a large jump in the most recent wave of @prripoll.bsky.social data.
March 28, 2025 at 3:07 PM
First, we introduce the idea of “blatant bias” (BB)—we define it as support for a politician *regardless* of what the politician does.
We find similar levels of BB among both Biden & Trump supporters, though Trump supporters showed a large jump in the most recent wave of @prripoll.bsky.social data.
We find similar levels of BB among both Biden & Trump supporters, though Trump supporters showed a large jump in the most recent wave of @prripoll.bsky.social data.
🚨New WP! What is political bias in its most extreme form? What factors are associated with it? And how can it be reduced?
At next week's @MPSAnet conf., @iganson.bsky.social & I will present our paper "Rejoining the Cult" that offers answers these questions: go.shr.lc/4j9bg9j 🧵
At next week's @MPSAnet conf., @iganson.bsky.social & I will present our paper "Rejoining the Cult" that offers answers these questions: go.shr.lc/4j9bg9j 🧵
March 28, 2025 at 3:07 PM
🚨New WP! What is political bias in its most extreme form? What factors are associated with it? And how can it be reduced?
At next week's @MPSAnet conf., @iganson.bsky.social & I will present our paper "Rejoining the Cult" that offers answers these questions: go.shr.lc/4j9bg9j 🧵
At next week's @MPSAnet conf., @iganson.bsky.social & I will present our paper "Rejoining the Cult" that offers answers these questions: go.shr.lc/4j9bg9j 🧵
A perfect encapsulation of our current moment:
Celebrating the destruction of the Dept. of Education whilst misspelling “Department”
Celebrating the destruction of the Dept. of Education whilst misspelling “Department”
March 20, 2025 at 10:48 PM
A perfect encapsulation of our current moment:
Celebrating the destruction of the Dept. of Education whilst misspelling “Department”
Celebrating the destruction of the Dept. of Education whilst misspelling “Department”