David Nield
@dnield.bsky.social
Data scientist. Canada-born, Ohio-raised, Los Angeles-residing. New father to a beautiful baby daughter, Ohio State football fan, R stats enthusiast, seasonal anime enjoyer, and on-and-off chess hobbyist.
Reposted by David Nield
The problem with #rstats is portability. Shit like `df$x` has to be converted to `df€x` in Europe and `df£x` in the UK but nobody talks about this.
March 14, 2025 at 12:42 PM
The problem with #rstats is portability. Shit like `df$x` has to be converted to `df€x` in Europe and `df£x` in the UK but nobody talks about this.
Reposted by David Nield
a thought i’ll put out there: it is not the ideological moderation that is the electoral problem, it is that selecting for candidates who chase popular issue positions is necessarily selecting for candidates who exude the kind of focused-group affect that voters genuinely hate
March 14, 2025 at 10:17 AM
a thought i’ll put out there: it is not the ideological moderation that is the electoral problem, it is that selecting for candidates who chase popular issue positions is necessarily selecting for candidates who exude the kind of focused-group affect that voters genuinely hate
Reposted by David Nield
"When a survey question defines tariffs as taxes on imports, for example, fewer people say they support them. When a survey question calls out specific countries, more are in favor of tariffs on China than on other countries."
March 5, 2025 at 1:07 AM
"When a survey question defines tariffs as taxes on imports, for example, fewer people say they support them. When a survey question calls out specific countries, more are in favor of tariffs on China than on other countries."
Reposted by David Nield
Here's a cool, simple technique I learned from Joey Cherdarchuk ( @obumbratta.com ): in a choropleth map, your legend can double as a histogram - showing both the colour coding ánd the data distribution.
#maps #dataviz
#maps #dataviz
February 25, 2025 at 8:51 AM
Here's a cool, simple technique I learned from Joey Cherdarchuk ( @obumbratta.com ): in a choropleth map, your legend can double as a histogram - showing both the colour coding ánd the data distribution.
#maps #dataviz
#maps #dataviz
Reposted by David Nield
recent scholarship agrees that the invention of shredded cheese was a collaborative effort, a history that runs contrary to the claims by some proponents of the so called "grate man theory"
February 23, 2025 at 7:04 PM
recent scholarship agrees that the invention of shredded cheese was a collaborative effort, a history that runs contrary to the claims by some proponents of the so called "grate man theory"
Reposted by David Nield
The theory is that price gouging is a good thing because the people who get the most utility from a good are the most Willing To Pay.
Just a coincidence I guess that people with more money are more Willing To Pay
Just a coincidence I guess that people with more money are more Willing To Pay
January 16, 2025 at 5:20 PM
The theory is that price gouging is a good thing because the people who get the most utility from a good are the most Willing To Pay.
Just a coincidence I guess that people with more money are more Willing To Pay
Just a coincidence I guess that people with more money are more Willing To Pay
Reposted by David Nield
People don't really think any of this! Overestimates of small proportions are a quirk of general innumeracy, not genuine misperceptions.
See research by @brianguay.bsky.social et al:
—PNAS forthcoming: www.brianguay.com/files/guay_2...
—Psychonomic Bulletin 2017: link.springer.com/article/10.3...
See research by @brianguay.bsky.social et al:
—PNAS forthcoming: www.brianguay.com/files/guay_2...
—Psychonomic Bulletin 2017: link.springer.com/article/10.3...
January 14, 2025 at 2:43 PM
People don't really think any of this! Overestimates of small proportions are a quirk of general innumeracy, not genuine misperceptions.
See research by @brianguay.bsky.social et al:
—PNAS forthcoming: www.brianguay.com/files/guay_2...
—Psychonomic Bulletin 2017: link.springer.com/article/10.3...
See research by @brianguay.bsky.social et al:
—PNAS forthcoming: www.brianguay.com/files/guay_2...
—Psychonomic Bulletin 2017: link.springer.com/article/10.3...
Reposted by David Nield
Target Losing Market Share As More Americans Opt To Forgo All Earthly Possessions
theonion.com/target-...
theonion.com/target-...
January 14, 2025 at 8:00 PM
Target Losing Market Share As More Americans Opt To Forgo All Earthly Possessions
theonion.com/target-...
theonion.com/target-...
Reposted by David Nield
Outraged by Jan 6, he spent two years getting inside the top ranks of militias like the Oath Keepers. He was stunningly successful. He penetrated a new generation of militia leaders, which included doctors, cops & government attorneys.
This is his story. www.propublica.org/article/ap3-...
This is his story. www.propublica.org/article/ap3-...
A Mole Infiltrated the Highest Ranks of American Militias. This Is What He Found.
Outraged by the Jan. 6 Capitol riot, a wilderness survival trainer spent years undercover climbing the ranks of right-wing militias. He didn’t tell police or the FBI. He didn’t tell family or friends....
www.propublica.org
January 4, 2025 at 12:48 PM
Outraged by Jan 6, he spent two years getting inside the top ranks of militias like the Oath Keepers. He was stunningly successful. He penetrated a new generation of militia leaders, which included doctors, cops & government attorneys.
This is his story. www.propublica.org/article/ap3-...
This is his story. www.propublica.org/article/ap3-...
Reposted by David Nield
Midwest Peace Talks Shattered By Illinois Toll-Booth Bombing
theonion.com/midwest-peac...
theonion.com/midwest-peac...
January 4, 2025 at 6:44 PM
Midwest Peace Talks Shattered By Illinois Toll-Booth Bombing
theonion.com/midwest-peac...
theonion.com/midwest-peac...
Reposted by David Nield
In economics, editors, referees, and authors often behave as if a published paper should reflect some kind of authoritative consensus.
As a result, valuable debate happens in secret, and the resulting paper is an opaque compromise with anonymous co-authors called referees.
1/
As a result, valuable debate happens in secret, and the resulting paper is an opaque compromise with anonymous co-authors called referees.
1/
December 24, 2024 at 2:44 PM
In economics, editors, referees, and authors often behave as if a published paper should reflect some kind of authoritative consensus.
As a result, valuable debate happens in secret, and the resulting paper is an opaque compromise with anonymous co-authors called referees.
1/
As a result, valuable debate happens in secret, and the resulting paper is an opaque compromise with anonymous co-authors called referees.
1/
Being a pedant here but the French revolution was not sparked by increasing wealth inequality, but the exact opposite: an increasingly wealthy and educated bourgeoisie resentful of the inherited privileges of an aristocratic class. The leaders of the third estate in the estates general were rich.
December 23, 2024 at 4:35 AM
Being a pedant here but the French revolution was not sparked by increasing wealth inequality, but the exact opposite: an increasingly wealthy and educated bourgeoisie resentful of the inherited privileges of an aristocratic class. The leaders of the third estate in the estates general were rich.
Reposted by David Nield
"This is a new right based on a populist working class nativism except that it’s the worlds richest man, an immigrant, slashing the welfare state and taxes as some kind of libertarian shock therapy. Absolute clown country."
-Matt Bruenig
-Matt Bruenig
December 22, 2024 at 8:03 PM
"This is a new right based on a populist working class nativism except that it’s the worlds richest man, an immigrant, slashing the welfare state and taxes as some kind of libertarian shock therapy. Absolute clown country."
-Matt Bruenig
-Matt Bruenig
Reposted by David Nield
a very nice talk with a lot of insights and lessons! the slide deck at cbergmeir.com/talks/bergme...
December 21, 2024 at 4:33 PM
a very nice talk with a lot of insights and lessons! the slide deck at cbergmeir.com/talks/bergme...
Reposted by David Nield
Experts Recommend Putting Injured Person In Some Sort Of Cylindrical Tank Filled With Fluid
theonion.com/experts-reco...
theonion.com/experts-reco...
December 19, 2024 at 4:54 PM
Experts Recommend Putting Injured Person In Some Sort Of Cylindrical Tank Filled With Fluid
theonion.com/experts-reco...
theonion.com/experts-reco...
Reposted by David Nield
That a dude with a couple cameras and a plate of appetizers can make something worth $82.5 million and a media company looks at that and goes "couldn't we make even more money if we had AI do it?" tells you exactly how stupid these people are
The $82.5 million sale of "Hot Ones" studio First We Feast is part of a broader strategy at BuzzFeed to shift away from human editors, writers and content producers in favor of artificial intelligence, which result in "high-margin, tech-enabled revenue lines."
BuzzFeed sells "Hot Ones" studio for $82.5 million as it pursues more AI-driven content
The company said the sale will allow it to depend less on human-created content in favor of "high-margin, tech-enabled revenue lines."
thedesk.net
December 13, 2024 at 1:31 PM
That a dude with a couple cameras and a plate of appetizers can make something worth $82.5 million and a media company looks at that and goes "couldn't we make even more money if we had AI do it?" tells you exactly how stupid these people are
Maybe an unpopular opinion, but I hate the term "skeet" to refer to bluesky posts. Please tell me that it's just a tongue-in-cheek colloquialism that can die out rather than the official term that the company uses.
December 10, 2024 at 11:15 PM
Maybe an unpopular opinion, but I hate the term "skeet" to refer to bluesky posts. Please tell me that it's just a tongue-in-cheek colloquialism that can die out rather than the official term that the company uses.
Reposted by David Nield
How I feel when people say "if the death rate from cancer is declining, then the death rate from another cause must be rising"
December 10, 2024 at 10:40 PM
How I feel when people say "if the death rate from cancer is declining, then the death rate from another cause must be rising"
Reposted by David Nield
Here is an experiment on hacking the TSB model for intermediate time series to accommodate availability constraints. We provide a way to add availability constraints to forecast the "true" demand and avoid just predicting zeros.
juanitorduz.github.io/availability...
juanitorduz.github.io/availability...
November 26, 2024 at 7:25 AM
Here is an experiment on hacking the TSB model for intermediate time series to accommodate availability constraints. We provide a way to add availability constraints to forecast the "true" demand and avoid just predicting zeros.
juanitorduz.github.io/availability...
juanitorduz.github.io/availability...
Reposted by David Nield
Reposted by David Nield
New paper: Do social media algorithms shape affective polarization?
We ran a field experiment on X/Twitter (N=1,256) using LLMs to rerank content in real-time, adjusting exposure to polarizing posts. Result: Algorithmic ranking impacts feelings toward the political outgroup! 🧵⬇️
We ran a field experiment on X/Twitter (N=1,256) using LLMs to rerank content in real-time, adjusting exposure to polarizing posts. Result: Algorithmic ranking impacts feelings toward the political outgroup! 🧵⬇️
November 25, 2024 at 8:32 PM
New paper: Do social media algorithms shape affective polarization?
We ran a field experiment on X/Twitter (N=1,256) using LLMs to rerank content in real-time, adjusting exposure to polarizing posts. Result: Algorithmic ranking impacts feelings toward the political outgroup! 🧵⬇️
We ran a field experiment on X/Twitter (N=1,256) using LLMs to rerank content in real-time, adjusting exposure to polarizing posts. Result: Algorithmic ranking impacts feelings toward the political outgroup! 🧵⬇️
Reposted by David Nield
I straight up think it is dishonest to make this claim without noting that a) the mobilization against the Muslim ban involved a thousands of actual people flooding airports and b) coincided with a massive mobilization of people on the ground in Washington D.C.
November 22, 2024 at 5:13 PM
I straight up think it is dishonest to make this claim without noting that a) the mobilization against the Muslim ban involved a thousands of actual people flooding airports and b) coincided with a massive mobilization of people on the ground in Washington D.C.
Reposted by David Nield
an overlooked aspect of all these “bluesky is an echo chamber” pieces is that the writers bizarrely think that the only reason people use social media is to talk about politics, which is true only for a subset of users
November 22, 2024 at 3:21 PM
an overlooked aspect of all these “bluesky is an echo chamber” pieces is that the writers bizarrely think that the only reason people use social media is to talk about politics, which is true only for a subset of users