>"We need to check more carefully who is credited and who is accountable for statistical agency releases. We need to watch carefully, when the 2025 American Community Survey eventually comes out, whether the ban on noise infusion leads to less data being published and which tables are suppressed. We also need to question seemingly technical methodological changes and scrutinize them for political motivation more closely."
I didn't realize they were delaying the 2025 ACS again and it seems things are just going to get worse if things don't fit the narrative of the administration.
the US statistical system is currently down, I'm estimating that we won't get an accurate picture of what systems have been wiped out until we can get some forensic analysts in there in 2029 or so once we solve a little HR issue we have in the executive branch.
2029? By then the pendulum would swing back to it being considered "deep state intervention" and a "waste of taxpayer funds" especially since "the national debt is higher than ever".
whats crazy is that we got here and it isn't even the midterms yet, who knows how cranky folks are going to get if the midterms don't go well for them.
Absolutely insane how the USA is speed running into Russia.
I was discussing BBQ with some friends the other day and living in Asia there are certain cuts that you basically have to get imported if you want to do it right. Basically every single person in that group, including Americans have decided they wouldn't trust or buy meat produce from the USA anymore as they don't trust it anymore...
except they also seem to speed running crashing the world economy, making it feels more like "speed running Russian government structures while speed running a soviet union future"...
which isn't really "that" surprising, if you consider that, while most people will suffer from a sever world economy crash, the ~top 1% or so will long term profit. Mainly because even iff they lose a huge lot they still have enough left to buy up theoretically well working businesses crashed by the large scale economy crash. Sure that would be very evil, but you only need a few of the richest pushing for it while the rest "just" doesn't push against it for this to happen. Or a few of the richest having even way more money then the rest. And given that both (huge differences in wealth of the wealthiest & lack of morals) seem the case it's hard to believe that the current direction is fully accidental.
Which brings use back to the topic of how any proper free market must not allow any to large concentration of money as it will break the market sooner or later. Or how democracies must not allow too much concentration of political power (or it break sooner or later very reliably historically speaking). Or how the one thing a society preaching tolerance can't tolerate is strong cases of intolerance. All the same kind of "naively it looks like a contradiction" cases where if you look into them it kinda start looking obvious that tolerating that is a bad idea.
Problem is while preventing aggregation of power, wealth, etc. is very viable if properly perused by politics, it's to late for prevention.
But also there isn't really a good fix once a country (or more like most of the world) failed to prevent it.
Trying to forcefully size assets is a severely unpleasant idea. More importantly but it's a thing things which tend to end very very badly.
And the ethical issues with that are pretty obvious, too.
Through also this isn't about millionaires, or the top 10%,1%,0.1%, but about wealth "beyond reason". Not sure where that starts but probably lowest in the the 50-100 millions and definitely in the billions (or trillions). I.e. the point where your wealth gives you trivially more political influence then the votes of the majority of people... I mean in the end about protecting democracy and a (proper) free market. It isn't about communism, no matter how much certain propaganda channels will claim that supposedly wanting a working democracy and free market is communism (or socialism). Through iff, and only if, you put the vague limit of when something start being "unreasonable rich" at a much lower point, e.g. inflation corrected 2 million, then _that_ is making it about communism like stuff. So it's a slippery slope. (Sadly too many people seem to have lost the ability (or will?) to have well differentiated discussions or opinions, always falling into extremes, so discussing such things seems often pointless)
It does feel a bit like people have so much money they could basically drive us into extinction (or close too it) by constantly supporting and funding psychopathic lunatics like Trump, helping them stay in power and watching them do nothing to hurt their business, therefore do nothing to stop anthropomorphic climate change.
I believe that most people are in favor of fixing the problems caused by the improper allocation of additional congressional representation to areas with larger populations of non-citizens. This piece complains about the upcoming changes without going into any depth on the problems these proposed changes will resolve.
I think the burden of proof's on you to show that the "changes" mentioned by TFA will do the "resolution" you want. For now -- as a working statistician who uses Census products -- we have a lot of collateral damage that impacts our ability to reliably represent the status of the United States. As well as a sh*t publication with no one's name on it, and the expectation that the Census will publish fewer and slower going forwards.
If people are in favor of changing the law they can vote for folks to change it. Until then, the government needs to follow the law - especially the highest in the land.
So what do you make of the phrase in Amendment 14 section two; "excluding Indians not taxed."? Do you suppose maybe they were trying to limit the representation to citizens?
And through some impressive legal magic perversity, illegal immigrants in the US pay plenty of taxes. "No taxation without representation", wasn't that a phrase?
Well, at the time, Indians (Native Americans) were the largest populaton of non-citizens residing in the country. That is no longer the case today. (There are 10 times more non-citizens than non-citizen Native Americans today.)
Imagine how much more sense things would make if "Race [dropdown]" and "Hispanic? [Y/N]" weren't separate fields. Or if the available options made sense, or even allowed multiselect instead of having to pick only one. Or allowed for "all my ancestors that I know names of lived in the states, so stop trying to associate me with other places".
.
And, um, I can kind of understand thinking it's wrong to deliberately falsify results even in a controlled manner.
first they are mostly collecting statistics about ethnicity, not race
genetically there is only one human race (which is often best explained as the genetic differences _inside_ a sample group of arbitrary ethnicities being larger then the difference between (a hypothetical average genome) of such groups).
through to be fair, people use race and ethnicity commonly interchangeable
what is more relevant is why should a country even collect either such information?!
the only thing which seems potentially relevant is statistics about stuff like _practiced_ (sub-)culture, religion etc. But nothing really fixed by birth. (The reasons you might want to have that information is because it can affect which of multiple imperfect possible decisions, is the one wich is most likely causing the least amount of unintended issues for peoples daily live.)
And I know "people" also conflate practiced culture and race and ethnicity in daily conversations. But they also tend to tell white looking people with dreadlocks that they are acting very inappropriately, even if that person has a African parent (as a very dump example of the kind of situations which make me fell the need to write this comment. Also not me in the example. Just a fully arbitrary chosen, simple to understand example).
> first they are mostly collecting statistics about ethnicity, not race
From the 2020 form I can see, the question is: "What is this person’s race? Mark X one or more boxes [...]"
> what is more relevant is why should a country even collect either such information?!
Having the true population (or close to it) is very useful for a lot of purposes in statistics/scientific studies - like knowing whether your sample is representative (and how much to undersample/oversample if not), or controlling for race when looking at factors correlated with it.
Is that "difficult" quote from a separate source? I can't find it in the text of the linked report.
Also, the report says that both red and blue states had both overcounts and undercounts, which is a different picture than "over counts in blue states". It is accurate that more blue states had overcounts and more red states had undercounts.
I realise your question is rhetorical, but the report does offer a possible answer: "The 2020 Census faced many challenges such as
conducting fieldwork during the COVID-19 pandemic.
Other challenges to the 2020 Census included con-
troversy around a proposed citizenship question and
changes in the duration of the Nonresponse Followup
operation." It's plausible that those "challenges" could have been experienced unevenly by state, particularly with how politicised the pandemic and its mitigations were in the US.
Also worth pointing out that when the census can't collect individualized data, they estimate it based on existing government data. The quality of data varies significantly depending on the local government, and pretty much all governments were struggling to figure out the current state of things during the pandemic.
So what? The article describes the political implications of the Census Bureau errors. I cited the Census Bureau release itself so you can cross-check whether the factual assertions in the article are accurate.
> and also implies political causes.
> So it's useful to identify this as not something a journalist or credible outlet would actually put their name to.
> An independent analysis by the Pew Research Center and election data experts calculated exactly how those specific errors altered the final map. If the census had perfectly matched the PES estimates, the seats would have shifted as follows: [table showing apportionment errors cost Texas and Florida 1 seat each, and gave Minnesota, Rhode Island, and New York 1 undeserved seat each.
Fucking up 5 House seats for a decade seems like a firing-worthy offense.
> We estimated undercounts of people living in census tracts that had the lowest self-response rates. Specifically, we estimated a statistically significant net undercount among people living in the 20 percent of tracts with the lowest self-response rates.
News hot off the press: people bred from day one to be militantly antagonistic of the federal government end up undercounted when federal government comes to do counting.
Reminds me of the domestic terrorists who were threatening FEMA workers during Helene and then acting surprised when FEMA workers weren't scrambling to walk up to Trump-flagged homes.
> How can you fuck up so monumentally, but then when Trump fires you somehow he’s the problem?
They can both be a problem, it doesn’t have to be mutually exclusive. I’m sure in all the turnover Trump removed some incompetent people. He also probably removed quite a few competent people. The ratio is what matters.
This isn’t a minor fuckup. It skewed House apportionment for a decade. Think of what would happen in the private sector: your team screws up something that blows up a major customer relationship. Do you think the firings are going to be surgical?
> Think of what would happen in the private sector: your team screws up something that blows up a major customer relationship. Do you think the firings are going to be surgical?
Except that you can look at all of the firings the current administration has been behind and the pattern does not suggest that they are intended to remove people who were incompetent.
To the contrary, the large number of reductions (and forced retirements) suggests that the administration is trying to remove political opponents and people who are "disloyal" to it, and to generally weaken institutions that the administration doesn't like or feels threatened by.
Ohio and Utah were overcounted, Illinois was undercounted, and the remaining 37 states plus DC had no statistically significant error either way. Some of the significant results are also marginal: Delaware's estimate of 5.45% has a standard error of 2.82, so the 90% confidence interval barely clears zero.
Linked article is by kook right-wing Breitbart alum Kristin Tate, whitewashed through a relatively non-biased news source (thehill.com) by way of being an opinion piece (for which The Hill explicitly disavows agreement with any presented views).
Census bureau works pretty hard to get as accurate a count as possible, and hires mathematicians specifically to work on privacy. This is about the administration trying to put its finger on the pans
>"We need to check more carefully who is credited and who is accountable for statistical agency releases. We need to watch carefully, when the 2025 American Community Survey eventually comes out, whether the ban on noise infusion leads to less data being published and which tables are suppressed. We also need to question seemingly technical methodological changes and scrutinize them for political motivation more closely."
I didn't realize they were delaying the 2025 ACS again and it seems things are just going to get worse if things don't fit the narrative of the administration.
I guess I'm going to have to be an even bigger stickler to my colleagues for computing those margins of error.
the US statistical system is currently down, I'm estimating that we won't get an accurate picture of what systems have been wiped out until we can get some forensic analysts in there in 2029 or so once we solve a little HR issue we have in the executive branch.
2029? By then the pendulum would swing back to it being considered "deep state intervention" and a "waste of taxpayer funds" especially since "the national debt is higher than ever".
If anybody really cared about the national debt they would stop passing giant tax cuts.
GOP Voters don’t care when the ones they vote for are in power. Meanwhile the Dems reduce the debt and are pilloried for it
Would be nice to be in a stable country.
Sort this: https://en.wikipedia.org/wiki/List_of_modern_sovereign_state...
... by "Date of last subordination", it says there's very few that have been themselves for longer.
This shit aint never getting fixed.
whats crazy is that we got here and it isn't even the midterms yet, who knows how cranky folks are going to get if the midterms don't go well for them.
Absolutely insane how the USA is speed running into Russia.
I was discussing BBQ with some friends the other day and living in Asia there are certain cuts that you basically have to get imported if you want to do it right. Basically every single person in that group, including Americans have decided they wouldn't trust or buy meat produce from the USA anymore as they don't trust it anymore...
> running into Russia.
except they also seem to speed running crashing the world economy, making it feels more like "speed running Russian government structures while speed running a soviet union future"...
which isn't really "that" surprising, if you consider that, while most people will suffer from a sever world economy crash, the ~top 1% or so will long term profit. Mainly because even iff they lose a huge lot they still have enough left to buy up theoretically well working businesses crashed by the large scale economy crash. Sure that would be very evil, but you only need a few of the richest pushing for it while the rest "just" doesn't push against it for this to happen. Or a few of the richest having even way more money then the rest. And given that both (huge differences in wealth of the wealthiest & lack of morals) seem the case it's hard to believe that the current direction is fully accidental.
Which brings use back to the topic of how any proper free market must not allow any to large concentration of money as it will break the market sooner or later. Or how democracies must not allow too much concentration of political power (or it break sooner or later very reliably historically speaking). Or how the one thing a society preaching tolerance can't tolerate is strong cases of intolerance. All the same kind of "naively it looks like a contradiction" cases where if you look into them it kinda start looking obvious that tolerating that is a bad idea.
Problem is while preventing aggregation of power, wealth, etc. is very viable if properly perused by politics, it's to late for prevention.
But also there isn't really a good fix once a country (or more like most of the world) failed to prevent it.
Trying to forcefully size assets is a severely unpleasant idea. More importantly but it's a thing things which tend to end very very badly.
And the ethical issues with that are pretty obvious, too.
Through also this isn't about millionaires, or the top 10%,1%,0.1%, but about wealth "beyond reason". Not sure where that starts but probably lowest in the the 50-100 millions and definitely in the billions (or trillions). I.e. the point where your wealth gives you trivially more political influence then the votes of the majority of people... I mean in the end about protecting democracy and a (proper) free market. It isn't about communism, no matter how much certain propaganda channels will claim that supposedly wanting a working democracy and free market is communism (or socialism). Through iff, and only if, you put the vague limit of when something start being "unreasonable rich" at a much lower point, e.g. inflation corrected 2 million, then _that_ is making it about communism like stuff. So it's a slippery slope. (Sadly too many people seem to have lost the ability (or will?) to have well differentiated discussions or opinions, always falling into extremes, so discussing such things seems often pointless)
It does feel a bit like people have so much money they could basically drive us into extinction (or close too it) by constantly supporting and funding psychopathic lunatics like Trump, helping them stay in power and watching them do nothing to hurt their business, therefore do nothing to stop anthropomorphic climate change.
I believe that most people are in favor of fixing the problems caused by the improper allocation of additional congressional representation to areas with larger populations of non-citizens. This piece complains about the upcoming changes without going into any depth on the problems these proposed changes will resolve.
I think the burden of proof's on you to show that the "changes" mentioned by TFA will do the "resolution" you want. For now -- as a working statistician who uses Census products -- we have a lot of collateral damage that impacts our ability to reliably represent the status of the United States. As well as a sh*t publication with no one's name on it, and the expectation that the Census will publish fewer and slower going forwards.
Do you believe that? Do you also believe most people are in favor of ending prison gerrymandering?
Maybe they are. But that doesn't mean it will get ended.
If "improper allocation" problems are selectively addressed, expect division on partisan lines.
If people are in favor of changing the law they can vote for folks to change it. Until then, the government needs to follow the law - especially the highest in the land.
That change really requires a constitutional amendment, since the relevant article talks about counting persons, not citizens.
So what do you make of the phrase in Amendment 14 section two; "excluding Indians not taxed."? Do you suppose maybe they were trying to limit the representation to citizens?
Citizens? Surely you mean taxpayers?
And through some impressive legal magic perversity, illegal immigrants in the US pay plenty of taxes. "No taxation without representation", wasn't that a phrase?
The illegal immigrants do pay plenty of taxes, but how many of them pay Federal Income Tax? It's not something that you can do without a SSN.
Maybe they were but the compromise they accepted was Indians not taxed, so unless there’s some pretty surprising evidence…
Well, at the time, Indians (Native Americans) were the largest populaton of non-citizens residing in the country. That is no longer the case today. (There are 10 times more non-citizens than non-citizen Native Americans today.)
Ok. So? Still need an amendment.
You mean the pigheaded interpretation of the constitution?
Ignorance supports itself i guess.
Imagine how much more sense things would make if "Race [dropdown]" and "Hispanic? [Y/N]" weren't separate fields. Or if the available options made sense, or even allowed multiselect instead of having to pick only one. Or allowed for "all my ancestors that I know names of lived in the states, so stop trying to associate me with other places".
.
And, um, I can kind of understand thinking it's wrong to deliberately falsify results even in a controlled manner.
Hispanic isn't related to race.
first they are mostly collecting statistics about ethnicity, not race
genetically there is only one human race (which is often best explained as the genetic differences _inside_ a sample group of arbitrary ethnicities being larger then the difference between (a hypothetical average genome) of such groups).
through to be fair, people use race and ethnicity commonly interchangeable
what is more relevant is why should a country even collect either such information?!
the only thing which seems potentially relevant is statistics about stuff like _practiced_ (sub-)culture, religion etc. But nothing really fixed by birth. (The reasons you might want to have that information is because it can affect which of multiple imperfect possible decisions, is the one wich is most likely causing the least amount of unintended issues for peoples daily live.)
And I know "people" also conflate practiced culture and race and ethnicity in daily conversations. But they also tend to tell white looking people with dreadlocks that they are acting very inappropriately, even if that person has a African parent (as a very dump example of the kind of situations which make me fell the need to write this comment. Also not me in the example. Just a fully arbitrary chosen, simple to understand example).
> first they are mostly collecting statistics about ethnicity, not race
From the 2020 form I can see, the question is: "What is this person’s race? Mark X one or more boxes [...]"
> what is more relevant is why should a country even collect either such information?!
Having the true population (or close to it) is very useful for a lot of purposes in statistics/scientific studies - like knowing whether your sample is representative (and how much to undersample/oversample if not), or controlling for race when looking at factors correlated with it.
Sure. And there's that story about Feynman listing his race as "mutt" on some official form because the rest also aren't mutually exclusive.
[flagged]
Is that "difficult" quote from a separate source? I can't find it in the text of the linked report.
Also, the report says that both red and blue states had both overcounts and undercounts, which is a different picture than "over counts in blue states". It is accurate that more blue states had overcounts and more red states had undercounts.
I realise your question is rhetorical, but the report does offer a possible answer: "The 2020 Census faced many challenges such as conducting fieldwork during the COVID-19 pandemic. Other challenges to the 2020 Census included con- troversy around a proposed citizenship question and changes in the duration of the Nonresponse Followup operation." It's plausible that those "challenges" could have been experienced unevenly by state, particularly with how politicised the pandemic and its mitigations were in the US.
Also worth pointing out that when the census can't collect individualized data, they estimate it based on existing government data. The quality of data varies significantly depending on the local government, and pretty much all governments were struggling to figure out the current state of things during the pandemic.
Yeah, good catch. Meant to link to two sources, the census bureau admission and an article describing it. I forgot to paste one of them: https://thehill.com/opinion/campaign/3504546-census-blunder-...
Opinion.
So what? The article describes the political implications of the Census Bureau errors. I cited the Census Bureau release itself so you can cross-check whether the factual assertions in the article are accurate.
The article describes political implications, and also implies political causes.
So it's useful to identify this as not something a journalist or credible outlet would actually put their name to.
> and also implies political causes. > So it's useful to identify this as not something a journalist or credible outlet would actually put their name to.
Why? And is the bottom line point wrong?
Gemini, citing a Pew Research study, seems to agree that Republicans lost 3-5 seats due to apportionment errors: https://share.google/aimode/9yPrkEU1U08abPWzF
> An independent analysis by the Pew Research Center and election data experts calculated exactly how those specific errors altered the final map. If the census had perfectly matched the PES estimates, the seats would have shifted as follows: [table showing apportionment errors cost Texas and Florida 1 seat each, and gave Minnesota, Rhode Island, and New York 1 undeserved seat each.
Fucking up 5 House seats for a decade seems like a firing-worthy offense.
Is your understanding that they found "the person responsible" for this "error" (of people not self-reporting to Census workers?) and then fired them?
That seems to be what you believe happened, but it's not.
> We estimated undercounts of people living in census tracts that had the lowest self-response rates. Specifically, we estimated a statistically significant net undercount among people living in the 20 percent of tracts with the lowest self-response rates.
News hot off the press: people bred from day one to be militantly antagonistic of the federal government end up undercounted when federal government comes to do counting.
Reminds me of the domestic terrorists who were threatening FEMA workers during Helene and then acting surprised when FEMA workers weren't scrambling to walk up to Trump-flagged homes.
> How can you fuck up so monumentally, but then when Trump fires you somehow he’s the problem?
They can both be a problem, it doesn’t have to be mutually exclusive. I’m sure in all the turnover Trump removed some incompetent people. He also probably removed quite a few competent people. The ratio is what matters.
This isn’t a minor fuckup. It skewed House apportionment for a decade. Think of what would happen in the private sector: your team screws up something that blows up a major customer relationship. Do you think the firings are going to be surgical?
Yes? This is neither Cesar's legion, nor The Apprentice.
Although, the guy at the top sure runs his place like a TV show, where all the stuntmen really do die for real.
Yes, the government shouldn’t be run on emotions or like a for profit company
> Think of what would happen in the private sector: your team screws up something that blows up a major customer relationship. Do you think the firings are going to be surgical?
Except that you can look at all of the firings the current administration has been behind and the pattern does not suggest that they are intended to remove people who were incompetent.
To the contrary, the large number of reductions (and forced retirements) suggests that the administration is trying to remove political opponents and people who are "disloyal" to it, and to generally weaken institutions that the administration doesn't like or feels threatened by.
This is especially visible in the military:
https://www.cbsnews.com/news/military-leaders-fired-forced-o...
The quote isn't corroborated by the report.
Ohio and Utah were overcounted, Illinois was undercounted, and the remaining 37 states plus DC had no statistically significant error either way. Some of the significant results are also marginal: Delaware's estimate of 5.45% has a standard error of 2.82, so the 90% confidence interval barely clears zero.
Community Note:
Linked article is by kook right-wing Breitbart alum Kristin Tate, whitewashed through a relatively non-biased news source (thehill.com) by way of being an opinion piece (for which The Hill explicitly disavows agreement with any presented views).
[dead]
[flagged]
Corruption in the US Census Bureau? No fuckin' way!
Census bureau works pretty hard to get as accurate a count as possible, and hires mathematicians specifically to work on privacy. This is about the administration trying to put its finger on the pans