This is a funny one. Remember how Google used to have the slogan, “Don’t be evil,” but then they abandoned it? Something similar seems to have happened with the U.S. Census Bureau.
Dale Lehman pointed me to the story:
I’m sure this is not news to you. But I don’t think the usual lag is appropriate given the seriousness of this.
From the linked item, by economist Justin Wolfers:
“Protection from inappropriate influence is a hallmark of scientific integrity.” That line governed the Census Bureau and then the White House quietly disappeared it. . . .
Three weeks ago, I would have told you that the American government couldn’t suppress economic data like this. But under the new rules that the administration just put forward, the White House can direct its political operatives to kill uncomfortable numbers. . . .
Three weeks ago the United States had a policy that protected the free flow of scientific information and barred political meddling in how Census Bureau findings are presented. Today, it doesn’t.
To be clear: I’m not (yet) saying that our government is cooking the statistical books. Instead, I’m saying that it took the necessary steps to make that legal, and possible. In recent weeks, the Trump administration quietly rewrote the rules that kept political interference out of our economic statistics. And almost nobody noticed.
I didn’t notice either! Contra Lehman’s assumption, this was news to me. I guess that’s how news spreads now: Wolfers posts this on his blog, Lehman sees it and emails the link to me, I post it here and you see it, etc. Sometimes Tyler Cowen or Alex Tabarrok notices items from this blog and they post it on Marginal Revolution. Or maybe Krugman will post on it. The New York Times or Wall Street Journal might cover it too.
What I kind of wonder is how Wolfers heard about it. He’s a user of public data, so it makes sense that he’d be directly interested in the topic. But how did he find out about this obscure announcement? Was he just rooting through the census website one day and noticed something different? Did he see this on some news site? Was it sent to him by an anonymous tipster (I get lots of that sort of email!)? I’m curious. Wolfers does point to this NPR report from last month, so my guess is that someone noticed it and pointed Wolfers to it.
From the NPR article:
The Commerce Department updated its policy for promoting a “culture of scientific integrity” in January 2025, days before Trump’s second inauguration.
A new Trump administration order bans the Census Bureau and the Bureau of Economic Analysis from using statistical “noise,” or data for fuzzing survey results, to protect people’s privacy in their statistics.
The 2025 version included bans on attempts to shape or interfere in data collection, statistical analysis and other scientific activities “against well-accepted scientific methods and theories or without scientific justification.”
The department’s policy, the 2025 version said, involves ensuring scientific findings at its agencies “are not suppressed, delayed, or altered for political purposes and are not subjected to inappropriate influence,” as well as making sure there’s “independent review of facilities, methodologies, and other scientific activities as appropriate to ensure scientific integrity.”
That language is missing, however, in the Trump administration’s revision of the policy, according to an NPR analysis of text posted on the Commerce Department’s website.
And while the revision adopts a definition for “scientific integrity” that is similar to the previous version, it no longer says that “protection from inappropriate influence” is among the “hallmarks of scientific integrity.”
Spokespeople for the Commerce Department and Census Bureau did not respond to NPR’s requests for comment.
I guess the strategy of dodging comments worked, at least for awhile, because for three weeks nobody else seemed to have picked up on this story.
NPR continues:
While most of the public may not be familiar with the policy, Nancy Potok, a former deputy director at the bureau who has also served at the Commerce Department, sees the removed language as “a major change that eliminates the guardrails against political interference in the statistical and scientific data products of the department.”
Potok has been tracking how statistical agencies have been responding to a years-long, bipartisan effort to make sure the federal government produces trustworthy information. That included a 2018 law that led to a rule known among federal data experts as the “Trust Regulation.” The rule requires that statistical agencies “determine the policies and practices that ensure objectivity of its statistical activities” and produce data that is “impartial and free from undue influence and the appearance of undue influence.”
“The fact that there is a conscious effort to revise the department administrative orders and not incorporate things that are in the [Trust Regulation], instead of going in the other direction, is definitely troubling and needs some explanation,” Potok adds.
Why did the government do this?
For former Census Bureau Director Robert Santos, a Biden nominee, the close timing of the policy revision and the bureau’s release of an unauthored report with claims about noncitizen voting is striking. “It’s entirely possible that this was done to reconcile the Department of Commerce and the administration’s own actions by publishing an external partisan document about voting on the Census Bureau website,” Santos says.
And Wolfers brings on the indignation:
On August 19, the Commerce Department posted a new directive called “Scientific Integrity” — note the scare quotes. . . . They posted this directive very quietly. There was no press release, no announcement, and no press conference. For the first time in his life, Howard Lutnick was quiet. The new version just appeared online. To analyze the changes, I [Wolfers] actually had to go to the Wayback Machine to find the prior version of this policy, because it had been wiped from the Commerce website. There’s something pretty bleak about a country’s rules about scientific integrity surviving only because a volunteer internet archive grabbed a copy.
The Jan 2025 rules are here, and the new rules are here.
What do I think about all this? I’m not sure.
On one hand, the above story sounds horrible, and I think it’s disgraceful how government officials have been throwing around slogans such as “gold standard science” while suppressing reports, promoting discredited or flat-out fake science, and adding corruption to the science funding process, while their friends in the media are resorting to ridiculous defenses of indefensible policies.
On the other hand, this updated document could just be some overly busy bureaucrats trying some stupid idea. Also, if that 2025 policy was such a good idea, why did they wait until 2025 to institute it?
If “Protection from inappropriate influence is a hallmark of scientific integrity” is such an important slogan, where was it in the Census Bureau’s first 235 years? There must be more to this story.
On the whole, though, I guess I’d go with Noem’s Razor here: Never attribute to stupidity that which is adequately explained by malice.
In short: I think it’s reasonable to assume that this new document was motivated by some inappropriate reason, such as a desire to steal or to cover up some past or anticipated future offense, or from a general ideological position opposing objective knowledge and open scientific inquiry, or from a political attitude that independent science should be weakened because it represent a source of power that’s not under control of the government or its allies. Also, though, I guess they’re stupid. Just like the Soviets under Lysenko, an analogy that keeps coming to mind.
There’s an alternative explanation that this is all just a bunch of paperwork jockeys trying to justify their own existence, but Noem’s Razor suggests otherwise.
Never mind the politics. When talking about “data,” the important, overriding issue is whether to use a singular verb or a plural verb. In the above article, “data” is also an adjective. It may come as a shock to some, but in some fields, e.g., geodesy, the plural of datum is datums. And, I just found out that “datums” is the proper term in needlework as well.
—————————–
“To protect their terminology from being confused with computers, specialized fields—like geodesy, mechanical engineering, and precision artisans—intentionally broke away from standard grammar. They created the word “datums” specifically to signal: “We are talking about physical geometry and measurements, not computers and statistics.”
When talking about data, it might also be amusing to mention:
– Data, the fictional character from one of my favorite childhood movies “The Goonies”
– Data, the fictional character from Star Trek
– DATA, the techno-pop band
– Data, a genus of moths
– DATA, the non-governmental organization founded by Bono
– DATA, the Design and Technology Academy in San-Antonio
Thanks Wikipedia for making this, possibly amusing and educational, comment possible!
It’s extremely relevant (or not) that all of these mentions, “data” would be take a singular verb.
I looked up some data about the Datas and found out that Data from the Goonies is the OG Data as that movie is from 1985 and according to wikipedia Data from Star Trek had his first appearence in 1987.
Side note: 80’s movies rock! The Goonies, Gremlins, Back To The Future, Indiana Jones, E. T., Star Wars, Stand By Me, Die Hard!
Quote from above: “I looked up some data about the Datas and found out that Data from the Goonies is the OG Data as that movie is from 1985 and according to wikipedia Data from Star Trek had his first appearence in 1987. ”
Interesting datum regarding the dates of first appearances of the Datas!
Gonna have to remember that for my possible bar named Poolbeer located in Alaska, and the pub quiz I will be orginazing there regularly should it become a reality.
It’s weird to me that there are no substantive comments on this post.
Yeah, I don’t know—maybe everyone who read it either agrees with you, and so there’s nothing to say, or they disagree but they have no good arguments against your point so they’re rather not think about the whole thing!
I guess it’s just not news because this is consistent with how science has been treated. The bigger news is the proposal to not count “non-citizens” and not collect data on race and sexual orientation (link to NYTimes). They certainly understand that the best defense against unwanted data is to not have the data.
Kaiser:
Yeah, there’s a numbness. Disappearing data is nothing compared to shooting people in boats etc.
Government economic stats have always been heavily manipulated and biased. The main purpose is to manipulate public sentiment. The tricks like “avoid collecting inconvenient data” sound standard rather than novel as well.
If someone is starting from the premise this a new thing, there is little to learn from any discussion.
The real story is meta. Why are NYT/NPR/etc telling their audiences about it now, when previously they would avoid, minimize, or ridicule the topic? What does *that* imply?
Quote from above: “If someone is starting from the premise this a new thing, there is little to learn from any discussion.
The real story is meta. Why are NYT/NPR/etc telling their audiences about it now, when previously they would avoid, minimize, or ridicule the topic? What does *that* imply?”
Ding, ding!
+1
And another possible reason why some people do not even engage with the sort of surface level, or possibly even naive or skewed, reporting or opinion pieces anymore, as they may leave out crucial facts, views, reasoning, and more relevant and real issues.
Anon:
You write, “Government economic stats have always been heavily manipulated and biased. The main purpose is to manipulate public sentiment. The tricks like ‘avoid collecting inconvenient data’ sound standard rather than novel as well.”
I don’t think you’re correct here, at least not for the United States. Federal government statistics have a reputation for high quality, and the Federal statistical agencies have a reputation for seriousness and professionalism. That’s one reason why recent policies have been so disturbing.
I agree that decisions taken by the statistical agencies can serve political goals. Nothing can be completely insulated from politics. But I see no evidence that “the main purpose is to manipulate public sentiment” or that they “have always been heavily manipulated and biased.” Do you have any evidence for this claim?
Sure, GDP includes government spending and CPI ignores where the government spending causes inflation like housing, healthcare, education, stock market, etc.
Actual – fake inflation = “growth”
There are of course many other games with hedonics, relative weighting, shrinkflation, estimating then revising stats months later, and so on. But that is the primary one.
Anoneuoid
That is a very strange comment you make. Apparently you don’t like government spending and so find the government statistics distorted because it includes that spending. Despite the many problems with how GDP is measured, I always like to point out that it is essentially equal to the total income earned in the economy. So, while you may think the government spending $X on building a missile should not be counted as a positive (a sentiment I have some agreement with), that $X was income for a number of people. Everybody might prefer an GDP measure that only includes spending/income on goods and services we deem “worthwhile,” but that would render these statistics totally subjective (and political). Instead, the GDP data is consistently collected and largely avoids such subjective judgements. Some still remain and are well known: labor performed outside of markets (e.g. household labor, predominantly performed by women) is excluded and has been the subject of much criticism. But the fact that these cases are well known and relatively few is testament to the fact that these government statistics are less subject to political forces than they could be (I would say, a lot less).
Similarly, omission of a link between government spending and inflation is an enormous stretch. The connection is not as obvious as you suggest. And, to the extent that government spending does increase the rate of inflation, the real GDP should show less growth if government spending is increasing. Real GDP growth is one of those things reported by the government which I would say is far less influenced by political forces than it could be (or likely will be in the future). Again, there are well known issues: inflation measures usually fail to capture consumer’s ability to substitute cheaper goods/services for more expensive ones, so the CPI generally overstates inflation (this is ignoring the distributional issues which we have frequently discussed on this blog: a more granular view of CPI might reveal very different experiences for different subgroups). It is these known controversies that are evidence of the general high quality of these government statistics and their resistance to political pressure.
I am not denying that there is government pressure, nor am I saying that these statistics are untainted by such influence. But if you compare past US government statistics with China’s, for example, there is a real difference. And sweeping that difference under the rug makes it seem that these recent changes to the oversight of government statistics are not important. I think that is a dangerous view.
Dale, I just happen to be spending time looking at these questions at the moment. You will find some results and discussion including some input from Blair Fix who points out some important political adjustments…
https://mastodon.sdf.org/@dlakelan/117254295847091960
I think with a little massaging perhaps, both you and I will agree that the *nominal* GDP is *completely meaningless*. This is because it measures an amount of currency exchanged, but the ability of that currency to buy things is completely left out of the question.
My student and I built a dimensionless ratio GDP / (cost of buying a certain basket of international commodities each year). The results STRONGLY directly relate to economic consumer sentiment results. Whereas official inflation adjusted GDP PPP numbers were both weakly and *negatively* related to consumer sentiment.
I think GDP nominal is probably OK, it’s the “real” GDP that’s both *supposed* to be meaningful, and also simultaneously completely fictional.
Blair Fix pointed out that for example the relationship between nominal rent and CPI rent suggests that the government has imputed a 3x improvement in “quality” of the houses being rented today compared to those a few decades ago. This despite my house being constructed in 1946 and still basically the same floor, walls, electrical, etc. Also, “quality” is both in principle unobservable, and not actually a part of the *price* index. So smuggling “quality adjustments” into a price index creates a completely wrong impression of a hard observed statistic which is in fact complete fiction and negatively correlated with peoples expressed reality.
Quote from above: “It’s weird to me that there are no substantive comments on this post.”
I usually dislike most talk about politics, politicians, news events, etc., so I don’t commment on those things usually. On top of that, certain topics and people seem to be associated with significantly more sub-optimal or bad or manipulative writing and discussing by originators of the text or source and by commenters subsequently discussing this text or source. I don’t have the energy or will to go into all of that, but here’s just one example that may make something clear.
This is a quote from a person who wrote something talked about in this blog post if I am understanding things correctly:
“On August 19, the Commerce Department posted a new directive called “Scientific Integrity” — note the scare quotes. . . . They posted this directive very quietly. There was no press release, no announcement, and no press conference. For the first time in his life, Howard Lutnick was quiet.”
Now, for me, these few sentences alone might have an undertone and subtle or not so subtle word use that really annoys me and make it seem like I am reading something from a lawyer or politician or con-artist. I don’t have the energy and will to try and find out why this person uses these kinds of words, or whether they are appropriately used, etc.
Why not simply write or talk about the directive and what is good or bad or the same or different between the new and the previous version and leave sentences like “for the first time in his life, Howard Lutnick was quiet” out of things. I don’t know whether that is true, and I don’t have the energy and time to try and figure that out while I am trying to understand this directive and what’s possibly different and how and why.
Another example from this short quoted section is the following. Why write “note the scare quotes” here in this section? What is the point of that? What might that do to the reader? If I look at the linked document dated august 19th the title Scientific Integrity on top is without scare quotes, and at the beginning of the document the crucial terms are defined under section 3 using scare quotes to indicate them which seems pretty useful to me. What’s the big problem here? Am I missing something about the scare quotes? Why have I now spend time and energy reading and thinking about scare quotes?
Anyway, this is just one section, and these are just a few sentences, etc. It’s all so tiresome for me seen from multiple perspectives, so I don’t even read things like this anymore, which in turn results in me not commenting on the substance. Perhaps it’s not even deserving of the term “substance, perhaps it’s just slop or porridge written by AI and/or some dude with psychological issues and a lack of proper discernment and capabilities. I don’t know who all these people are writing all these things and I don’t care.
Anyway, I rather put my time and energy in other things, like wondering whether it is appropriate and useful to ponder why it looks to me that university students aren’t being educated well (enough) in logic and reasoning. At least, I did not receive a proper education in this if I remember correctly. Could such things be useful for many other things aside from scienctific activities as well, for instance with assessing a piece written by some dude who writes some stuff on his newsletter website?
Quote from above: “Could such things be useful for many other things aside from scienctific activities as well, for instance with assessing a piece written by some dude who writes some stuff on his newsletter website?”
Logic and reasoning seems pretty important to me, but I am no expert. Again, I wasn’t educated in this all so I may make mistakes here, but please allow me to say one more thing about this all using the topic of the blog post as an example.
Here’s a section quoted in the blog post:
“That language is missing, however, in the Trump administration’s revision of the policy, according to an NPR analysis of text posted on the Commerce Department’s website.
And while the revision adopts a definition for “scientific integrity” that is similar to the previous version, it no longer says that “protection from inappropriate influence” is among the “hallmarks of scientific integrity.””
Now, if I am looking at the definition of “scientific integrity” in the document dated august 19th I can see the term with “scare quotes” (or simply “quotation marks”) defined as: “Scientific Integrity” is defined as the adherence to professional practice, ethical behavior, and the principles of honesty, objectivity, and transparency when conducting, managing, peer reviewing, using the results of, and communicating about Scientific Activities and Scientific Information.”
Now, what does “protection from inappropriate influence” have to do with this definition of “scientific integrity”? Perhaps that protection could be seen as a pre-requisite or a wish or who knows what, but from a certain perspective one could argue, and perhaps even clearly see, that it has little to do with the actual definition of “scientific integrity” and with the possible emphasis in this definition on the actual behavior and principles and responsibilities of people.
So, perhaps a different focus (maybe even a more appropriate one) on a certain definition and interpretation of scientific integrity might already explain some things here. I don’t have the energy or will to go and investigate more than I already have, but do want to note that the document seems to further state tthat: “This Order otherwise is in addition to, and does not alter the requirements of any other applicable federal statutes, regulations, or policies, or other Operating Unit or Department administrative orders. These include, but are not limited to:”
A list then follows of a whole lot of things that perhaps might be about “protection from inappropriate influence” in one way or another. I don’t know what is meant with that exactly, but one of the things referred to is something called 2CRF part 200 where a lot of things can be read that might have to do with “protection from inappropriate influence” concerning certain situations or scenarios. For instance:
“Costs incurred in attempting to improperly influence, either directly or indirectly, an employee or officer of the executive branch of the Federal Government to give consideration or to act regarding a Federal award or a regulatory matter are unallowable.”
As might often be the case, it could be useful when discussing things to define terms. Sometimes that already makes things clear, or provides some clarity, or makes writing or discussing something more fruitful.
For those interested: “The manipulation of official statistics as corruption and ways of understanding it” (2021) by A. V. Georgiou
“It should be noted that the goal of manipulation of official statistics is to alter the perception of the reality that these statistics are supposed to depict so as to serve certain political interests. Thus, the manipulation of statistics can take the form of altering the statistical results themselves or it can take the form of altering the perception of the statistics. 9 Thus corruption in official statistics can appear with both these forms of manipulation, and the phenomena that are included in the present typology include phenomena from both these forms and not just from the ‘expected’ one – i.e., distorting the numerical figures themselves.” (p. 92)
This paper mentions some examples of manipulation of official statistics in Box 1, and makes a distinction between certain types which I will now summarize:
Type A phenomena include the use of the ‘crudest’ instruments in the manipulation of official statistics:
1. Direct alteration of statistical results
2. Not applying existing standards
3. Suppression of statistics
Type B phenomena include the use of what may be considered as more ‘sophisticated’ instruments in the manipulation of statistics:
4. Selective use of upstream data
5. Withholding upstream data
6. Indirect modification of methodology on ad hoc basis
7. Managing existing standards
8. Distorting the form of the release
9. Suppression of the production of statistics
10. Undermining the credibility of official statistics
Type C phenomena include the use of the most ‘sophisticated’ instruments in the manipulation of statistics:
11. Biased standard setting
The paper you are referring to offers a number of examples of government corruption of statistics from several countries. I think it is worth noting the US example where the government official responsible was demoted as a result. We know that countries differ greatly in corruption (e.g. https://www.transparency.org/en/cpi/2025). So, I think it is important to consider that paper in the context of Anoneuoid’s comment above (and Andrew’s response).
I accept that government statistics are subject to political forces. What is – and what is not – reported is influenced by politics, as well as the accuracy of these statistics. While I believe that is true everywhere, I do not think it is equally true everywhere. The problems are more severe in some places than in others. This post is specifically about a change in US policy regarding statistical collection and reporting. Officially removing the veneer of scientific oversight and replacing it with explicit political oversight is a major change. And, while Anon may say we should never have trusted any government data to begin with, I think that is a mistake. This recent change erodes whatever limited trust was there before. You don’t have to have faith in government statistics, to believe they have some credibility. And they will now have less credibility. Rather than sweeping them all into a grand statement that corruption is everywhere, I think the changes are real and important to recognize. Trust is not absolute, but it is important. And there is now less of it.
+1
+10
Quote from above: “+10”
Data manipulation! Multiple upvotes by a single person in a single comment.
Penalty flag on the field, 10 yard penalty, loss of down.