Originally published at: How a Baseless, Right-Wing Social Media Panic Led to Dramatic Changes at the Census Bureau - TPM – Talking Points Memo
When longtime Indiana politician Sen. Jim Banks (R-IN) lobbed a powerful allegation against the Census Bureau last October, what had been a low-profile, fringe GOP conspiracy theory burst to the fore. Banks said he believed the Census Bureau had injected errors in the decennial population count and survey which were so severe, they netted Democrats…
2 Likes
Trumpers want to ignore the US constitution, math, and science.
Can we sue the GOP for emotional and financial distress?
10 Likes
Right wing crackpots run the U.S. government.
That is my comment.
13 Likes
A. Categorically not a “Biden admin” thing if Census Bureau employees were already using this technique in 2020. But perhaps the language here is loose and the second quote should have started “In the 2020 census, …”
B. Extremely unlikely that Bureau under Biden went from not using noise infusion, to using noise infusion and pissing off Alabama enough to garner a lawsuit in the space of 2 months (January 20th - March whatevereth).
Conclusion: the technique was developed, chosen, and instituted during the Trump I admin.
15 Likes
If it weren’t for bad faith, they’d have no faith at all.
12 Likes
I don’t know why Republicans have such difficulty understanding differential privacy.
The definition is right there on wikipedia:
Let ε be a positive real number and A be a randomized algorithm that takes a dataset as input (representing the actions of the trusted party holding the data). Let im A denote the image of A. The algorithm A is said to provide (ε, δ)-differential privacy if, for all datasets D1 and D2 that differ on a single element (i.e., the data of one person), and all subsets S of im A:
where the probability is taken over the randomness used by the algorithm.
Since differential privacy is a probabilistic concept, any differentially private mechanism is necessarily randomized. Some of these, like the Laplace mechanism, rely on adding controlled noise to the function that we want to compute.
The Laplace mechanism adds Laplace noise (i.e. noise from the Laplace distribution, which can be expressed by probability density function noise(y) ∝ exp(−|y|/λ), which has mean zero and standard deviation
.Now in our case we define the output function of A as a real valued function (called as the transcript output by A) as TA(x) = f(x) + Y ∼ Lap(λ) and f is the original real valued query/function we planned to execute on the database. Now clearly TA(x) can be considered to be a continuous random variable, where
which is at most
We can consider
to be the privacy factor ε. Thus T follows a differentially private mechanism (as can be seen from the definition above).
14 Likes
Thanks for making it clear!
7 Likes
Isn’t the activity that will actually affect apportionment going to happen after trump is a gone goose? How the Census will operate when 2030 comes isn’t his call. This just strikes me as whopping up the voter fraud frenzy to no purpose other than the frenzy.
5 Likes
a disinformation campaign using categorically false information about census administration that has reduced a highly advanced statistical technique to just another arm of the disproven, right-wing delusion
The United States Census is the gold standard of what is, by definition, impossible – determining the number of people in the United States exactly and all at the same moment. Any delay injects error.
Okay. There’s no way to tell everyone in the country to stop what they’re doing, stand there, and wait for Jim Banks to walk by counting you. Instead, the Census Bureau has it down to several months. Plus, the Census has been conducted for …how long has it been, Jim? Why is your life’s ambition to run down the United States of America?
Especially since your damn Party has been running it for decades.
4 Likes
So, how does it work with zeros and ones ; - )
4 Likes
Lost in their own disinformation. They really do think that they will steal history too.
5 Likes
This. But as with what’s beginning to look like everything the Dem (fingers crossed) administration entering office in 2028 will confront, the Census Bureau will be forced to waste valuable time and energy cleaning up this mess. I just hope there are outside groups and some in Congress who monitor this Census Bureau shit so they’ll be as prepared to clean up the mess as Project 25 and others were prepared to start making it out of the gate in January 2025.
3 Likes
Just to note that it’s always been about more than numbers of people. Even in 1790, the categories were:
Name of the head of the household, number of free white males age 16 or over, number of free white males under the age of 16, number of free white females, number of other free persons (excluding slaves – so free Blacks were counted), number of slaves.
Each decade thereafter, categories were refined (age groups, for instance) and questions added, questions deemed important for understanding the shape and character of the population, questions like country of birth, citizenship status (note that non-citizens have always been counted), how many children a woman had given birth to and how many were still living (as “public health” became a thing, in 1900 and 1910). So much more. Always an opportunity to understand ourselves in terms of issues that were pressing at the time.
4 Likes
When you don’t understand how things work your only fallback is a conspiracy theory.
What we’re experiencing right now is what happens when America is at the mercy of really stupid people incapable of being serious.
10 Likes
Just want to add that as far as I, a mere former agronomist, understand it, differential privacy prevents a potential hacker from getting access to one individual’s data, but does not affect the accuracy of the bulk data used for reapportionment.
5 Likes
My celebration of your “clarifying” was a nod in the direction of sober factuality being presented on an opinion site that invites such while allowing those like myself and, occasionally @castor_troy , to opine in, ummm, less than serious contributions. I will add, respectfully, that my understanding of your post is nil. I am an old man for whom all of this electronic communication equipment replacing semaphor flags is nearly unbearable.
5 Likes
Hard to get a frenzied mob without a heaping basketful of frenzy.
2 Likes
My post was also a modest attempt at humor, to be honest.
4 Likes
This just strikes me as whopping up the voter fraud frenzy to no purpose other than the frenzy.
That is Ratbag’s favored modus operandi.
2 Likes
Another thing to take into account regardless of the mathematical explanation and proof thereof, is that the R’s don’t like the privacy aspect of differential privacy. They want to get hold of as much personal data on people as possible in order to manipulate the populace with disinformation.
They don’t want to collect or analyze data properly because it usually proves them wrong…like about noncitizens voting. So they’ll avoid learning to understand what differential privacy and noise injection actually are and how they work like the plague. Same applies to crime statistics, the economy, global warming…you name it.
4 Likes