Paul Novosad
@novosad.bsky.social
Econ prof @dartmouth, founder devdatalab.org
r2: a morass of disjointed streams of consciousness
🤷♂️
r2: a morass of disjointed streams of consciousness
🤷♂️
Academic creativity is underrated and technical methods are overrated.
Take this excellent paper on Zero Sum Thinking—simple in retrospect, just an excellent idea!
Graph shows rise in zero-sum thinking with Millennial and later U.S. birth cohorts.
Congrats to @s-stantcheva.bsky.social
Take this excellent paper on Zero Sum Thinking—simple in retrospect, just an excellent idea!
Graph shows rise in zero-sum thinking with Millennial and later U.S. birth cohorts.
Congrats to @s-stantcheva.bsky.social
April 24, 2025 at 7:12 PM
Academic creativity is underrated and technical methods are overrated.
Take this excellent paper on Zero Sum Thinking—simple in retrospect, just an excellent idea!
Graph shows rise in zero-sum thinking with Millennial and later U.S. birth cohorts.
Congrats to @s-stantcheva.bsky.social
Take this excellent paper on Zero Sum Thinking—simple in retrospect, just an excellent idea!
Graph shows rise in zero-sum thinking with Millennial and later U.S. birth cohorts.
Congrats to @s-stantcheva.bsky.social
If you want to learn more, we made an interactive demo of how publication bias works:
devdatalab.org/publication-...
And you can learn more about our paper on judicial bias here: devdatalab.org/judicial-bias
4/
devdatalab.org/publication-...
And you can learn more about our paper on judicial bias here: devdatalab.org/judicial-bias
4/
April 8, 2025 at 1:02 PM
If you want to learn more, we made an interactive demo of how publication bias works:
devdatalab.org/publication-...
And you can learn more about our paper on judicial bias here: devdatalab.org/judicial-bias
4/
devdatalab.org/publication-...
And you can learn more about our paper on judicial bias here: devdatalab.org/judicial-bias
4/
If there's publication bias, studies look asymmetric on the graph — because noisier studies need bigger effects to be statistically significant.
The left graph is a simulation of publication bias. The black points in the right graph are past studies on ingroup bias. 3/
The left graph is a simulation of publication bias. The black points in the right graph are past studies on ingroup bias. 3/
April 8, 2025 at 1:02 PM
If there's publication bias, studies look asymmetric on the graph — because noisier studies need bigger effects to be statistically significant.
The left graph is a simulation of publication bias. The black points in the right graph are past studies on ingroup bias. 3/
The left graph is a simulation of publication bias. The black points in the right graph are past studies on ingroup bias. 3/
In the absence of publication bias, estimates should form a funnel centered around the true effect.
Statistically significant estimates are outside the solid lines — but point mass should be similar just on either side of the line. 2/
Statistically significant estimates are outside the solid lines — but point mass should be similar just on either side of the line. 2/
April 8, 2025 at 1:02 PM
In the absence of publication bias, estimates should form a funnel centered around the true effect.
Statistically significant estimates are outside the solid lines — but point mass should be similar just on either side of the line. 2/
Statistically significant estimates are outside the solid lines — but point mass should be similar just on either side of the line. 2/
Our paper on ingroup bias in the Indian judiciary — we didn't find any — is finally out!
This paper has one of the two best graphs I've ever made — on publication bias in the judicial bias literature.
Black circles are prior studies on ingroup bias. Explanation in 🧵👇🏻
This paper has one of the two best graphs I've ever made — on publication bias in the judicial bias literature.
Black circles are prior studies on ingroup bias. Explanation in 🧵👇🏻
April 8, 2025 at 1:02 PM
Our paper on ingroup bias in the Indian judiciary — we didn't find any — is finally out!
This paper has one of the two best graphs I've ever made — on publication bias in the judicial bias literature.
Black circles are prior studies on ingroup bias. Explanation in 🧵👇🏻
This paper has one of the two best graphs I've ever made — on publication bias in the judicial bias literature.
Black circles are prior studies on ingroup bias. Explanation in 🧵👇🏻
This drives me crazy
February 13, 2025 at 6:00 PM
This drives me crazy
What's the deal with birth order effects? Is it generally agreed that there's a causal effect of first-born on wages and education, as suggested by this figure from Black (2018) ?
via @natehilger.bsky.social kiddingaround.substack.com/p/what-we-le...
Black (2018): direct.mit.edu/rest/article...
via @natehilger.bsky.social kiddingaround.substack.com/p/what-we-le...
Black (2018): direct.mit.edu/rest/article...
January 29, 2025 at 5:45 PM
What's the deal with birth order effects? Is it generally agreed that there's a causal effect of first-born on wages and education, as suggested by this figure from Black (2018) ?
via @natehilger.bsky.social kiddingaround.substack.com/p/what-we-le...
Black (2018): direct.mit.edu/rest/article...
via @natehilger.bsky.social kiddingaround.substack.com/p/what-we-le...
Black (2018): direct.mit.edu/rest/article...
Job market students, consider doing an awesome postdoc with us at Dartmouth! Position could be filled by development, trade, applied.
Because we don't have a PhD program, you can get a lot of time with faculty here.
Also it's in New Hampshire ⛷️🚵♂️🏔️🛶
Link below, deadline is in 2 weeks, please share!
Because we don't have a PhD program, you can get a lot of time with faculty here.
Also it's in New Hampshire ⛷️🚵♂️🏔️🛶
Link below, deadline is in 2 weeks, please share!
January 6, 2025 at 9:48 PM
Job market students, consider doing an awesome postdoc with us at Dartmouth! Position could be filled by development, trade, applied.
Because we don't have a PhD program, you can get a lot of time with faculty here.
Also it's in New Hampshire ⛷️🚵♂️🏔️🛶
Link below, deadline is in 2 weeks, please share!
Because we don't have a PhD program, you can get a lot of time with faculty here.
Also it's in New Hampshire ⛷️🚵♂️🏔️🛶
Link below, deadline is in 2 weeks, please share!
Health economists, is this graph correct?
Suggests % spending on healthcare tends to rise with GDP, and in these terms U.S. spending is just normal.
The source is: some blogger
randomcriticalanalysis.com/why-conventi...
Suggests % spending on healthcare tends to rise with GDP, and in these terms U.S. spending is just normal.
The source is: some blogger
randomcriticalanalysis.com/why-conventi...
December 12, 2024 at 4:21 PM
Health economists, is this graph correct?
Suggests % spending on healthcare tends to rise with GDP, and in these terms U.S. spending is just normal.
The source is: some blogger
randomcriticalanalysis.com/why-conventi...
Suggests % spending on healthcare tends to rise with GDP, and in these terms U.S. spending is just normal.
The source is: some blogger
randomcriticalanalysis.com/why-conventi...
Trending sticker searches
December 9, 2024 at 12:17 AM
Trending sticker searches
Last, this is why descriptive work is underrated.
No preposterous IV, no incomprehensible structural black box, there is just a new fact about the world.
If the descriptive work is done well — and it is not easy — the fact permanently enters everyone's headspace and must be contended with.
No preposterous IV, no incomprehensible structural black box, there is just a new fact about the world.
If the descriptive work is done well — and it is not easy — the fact permanently enters everyone's headspace and must be contended with.
December 6, 2024 at 7:19 PM
Last, this is why descriptive work is underrated.
No preposterous IV, no incomprehensible structural black box, there is just a new fact about the world.
If the descriptive work is done well — and it is not easy — the fact permanently enters everyone's headspace and must be contended with.
No preposterous IV, no incomprehensible structural black box, there is just a new fact about the world.
If the descriptive work is done well — and it is not easy — the fact permanently enters everyone's headspace and must be contended with.
Don't miss Ricardo's JMP on the changing backgrounds of Colombia's central bankers
ricardosalasdiaz.com/jmp
12/12
ricardosalasdiaz.com/jmp
12/12
December 6, 2024 at 7:10 PM
Don't miss Ricardo's JMP on the changing backgrounds of Colombia's central bankers
ricardosalasdiaz.com/jmp
12/12
ricardosalasdiaz.com/jmp
12/12
Some supporting evidence:
Seth Zimmerman's paper finds that elite schools in Chile get people into elite jobs — but only for boys from private schools!
Zero effects of elite education on men who went to public schools and on women! 11/N
t.co/Ikvv5uyBzV
Seth Zimmerman's paper finds that elite schools in Chile get people into elite jobs — but only for boys from private schools!
Zero effects of elite education on men who went to public schools and on women! 11/N
t.co/Ikvv5uyBzV
December 6, 2024 at 7:10 PM
Some supporting evidence:
Seth Zimmerman's paper finds that elite schools in Chile get people into elite jobs — but only for boys from private schools!
Zero effects of elite education on men who went to public schools and on women! 11/N
t.co/Ikvv5uyBzV
Seth Zimmerman's paper finds that elite schools in Chile get people into elite jobs — but only for boys from private schools!
Zero effects of elite education on men who went to public schools and on women! 11/N
t.co/Ikvv5uyBzV
My interpretation rant does not fit into BSky's word limit.
December 6, 2024 at 7:10 PM
My interpretation rant does not fit into BSky's word limit.
The paper is "Where did the global elite go to school," by Ricardo Salas-Diaz and Kevin Young. 9/N
t.co/DXwDCWMsGu
t.co/DXwDCWMsGu
December 6, 2024 at 7:10 PM
The paper is "Where did the global elite go to school," by Ricardo Salas-Diaz and Kevin Young. 9/N
t.co/DXwDCWMsGu
t.co/DXwDCWMsGu
The flatter distribution of national leaders is b/c most elites still get educated in their own countries.
The U.S. is 2nd place for nationals of every non-U.S. country, with some notable outliers — 34% of elite Brazilians went to U.S. colleges, 36% of elite Indians. 8/N
The U.S. is 2nd place for nationals of every non-U.S. country, with some notable outliers — 34% of elite Brazilians went to U.S. colleges, 36% of elite Indians. 8/N
December 6, 2024 at 7:10 PM
The flatter distribution of national leaders is b/c most elites still get educated in their own countries.
The U.S. is 2nd place for nationals of every non-U.S. country, with some notable outliers — 34% of elite Brazilians went to U.S. colleges, 36% of elite Indians. 8/N
The U.S. is 2nd place for nationals of every non-U.S. country, with some notable outliers — 34% of elite Brazilians went to U.S. colleges, 36% of elite Indians. 8/N
Now by sector, elite colleges are more concentrated for corporate and NGO positions than for billionaires and international orgs.
But for country leadership (central bankers and heads of state), the distribution is much more mixed, and Oxford edges out Harvard. (Yale still DNF)
But for country leadership (central bankers and heads of state), the distribution is much more mixed, and Oxford edges out Harvard. (Yale still DNF)
December 6, 2024 at 7:10 PM
Now by sector, elite colleges are more concentrated for corporate and NGO positions than for billionaires and international orgs.
But for country leadership (central bankers and heads of state), the distribution is much more mixed, and Oxford edges out Harvard. (Yale still DNF)
But for country leadership (central bankers and heads of state), the distribution is much more mixed, and Oxford edges out Harvard. (Yale still DNF)
My friends from Yale have long claimed it's about getting into Harvard or Yale, and it doesn't matter much between them. Maybe Princeton too.
Friends, the data points for Yale are not even high enough to get labeled. If there is a #2, it is Stanford.
5/N
Friends, the data points for Yale are not even high enough to get labeled. If there is a #2, it is Stanford.
5/N
December 6, 2024 at 7:10 PM
My friends from Yale have long claimed it's about getting into Harvard or Yale, and it doesn't matter much between them. Maybe Princeton too.
Friends, the data points for Yale are not even high enough to get labeled. If there is a #2, it is Stanford.
5/N
Friends, the data points for Yale are not even high enough to get labeled. If there is a #2, it is Stanford.
5/N
Here is graph for non-U.S. nationals. Harvard continues to stand out, Oxford and Cambridge are now also above the pack, but not close to Harvard. 4/N
December 6, 2024 at 7:10 PM
Here is graph for non-U.S. nationals. Harvard continues to stand out, Oxford and Cambridge are now also above the pack, but not close to Harvard. 4/N
Among U.S. nationals, it’s even more stark. Harvard represents 16% of elites — literally off the chart. 3/N
December 6, 2024 at 7:10 PM
Among U.S. nationals, it’s even more stark. Harvard represents 16% of elites — literally off the chart. 3/N
The paper is Ricardo Salas-Diaz and Kevin Young. They collected elite biographies across a few important international domains.
Corporate elites: CEO and board
IO: World Bank, IMF, Basel Committee, etc
National Elites: Head of state & central banker
3rd sector: think tanks 2/N
Corporate elites: CEO and board
IO: World Bank, IMF, Basel Committee, etc
National Elites: Head of state & central banker
3rd sector: think tanks 2/N
December 6, 2024 at 7:10 PM
The paper is Ricardo Salas-Diaz and Kevin Young. They collected elite biographies across a few important international domains.
Corporate elites: CEO and board
IO: World Bank, IMF, Basel Committee, etc
National Elites: Head of state & central banker
3rd sector: think tanks 2/N
Corporate elites: CEO and board
IO: World Bank, IMF, Basel Committee, etc
National Elites: Head of state & central banker
3rd sector: think tanks 2/N
Fascinating paper on where 6000 global elites went to college. Billionaires, CEOs, heads of state, central bankers, etc.
In a word: Harvard.
Fully 10% of global elites went to Harvard. Elite US schools are over-represented (23% IvyPlus), but nobody comes close to Harvard.
🧵
In a word: Harvard.
Fully 10% of global elites went to Harvard. Elite US schools are over-represented (23% IvyPlus), but nobody comes close to Harvard.
🧵
December 6, 2024 at 7:10 PM
Fascinating paper on where 6000 global elites went to college. Billionaires, CEOs, heads of state, central bankers, etc.
In a word: Harvard.
Fully 10% of global elites went to Harvard. Elite US schools are over-represented (23% IvyPlus), but nobody comes close to Harvard.
🧵
In a word: Harvard.
Fully 10% of global elites went to Harvard. Elite US schools are over-represented (23% IvyPlus), but nobody comes close to Harvard.
🧵
Dunno why I waited this long
December 6, 2024 at 2:08 AM
Dunno why I waited this long
With you on this one too
December 5, 2024 at 1:41 PM
With you on this one too
Gotta appreciate the honesty from this predatory journal — they review work "without giving a second thought"
December 2, 2024 at 3:09 PM
Gotta appreciate the honesty from this predatory journal — they review work "without giving a second thought"