100%. I live in a hamlet of a larger town in the US, and was curious what the population of my hamlet is.
There’s a Wikipedia page for the hamlet, but it’s empty. No population data, etc.
I’d much rather see no data than a LLM’s best guess. I’m guessing a LLM using the data would also perform better without approximated or “probably right” information.
Yes, however we lack essential data to run density estimates until late 1800s. Even cities didn't have accurate population counts, much less rural regions.
I looked around the local area here and it turns out to be quite reasonable, but there are also more than a couple misses, where the more notable town/village in a pair is suppressed.
I am a little skeptical about how dire this will really be. Census takers are allowed to guesstimate if non-responses reasonably make them believe they are undercounting:
>In instances where the bureau is unsure of the number of residents at an address after a field visit, its population characteristics are inferred from its nearest similar neighbor (hot-deck imputation). This practice has effects across many areas, but is seen by some as controversial.[6] However, the practice was ruled constitutional by the U.S. Supreme Court in Utah v. Evans.
Roughly that's all it's showing. I'd be far more interested in the cartogram of land value/population. I suspect it would do much more to show some of the oddities that I'm sure exist. (i.e. Park City where a lot of people have really expensive vacation homes for skiing and don't live there.)
Yeah I asked for an estimate of the percentage of the US population that lives in the DMV area (DC, Maryland, Virginia) and it was off by 50% of the actual answer, which I only realized when I realized I shouldn’t trust its estimate for anything important
Note that the registry for this is not central but managed by the towns instead. So it's very decentralized and data in this aggregated form doesn't exist. An often used trick by towns is to pretend some person didn't move away from the town, thus not decreasing the number of people who live there, thus increasing the number of various funds that are paid based on how many citizens the town has. So those registries often aren't 100% accurate.
I live in a college town. Some census tracts get sub 40% response rates. At least 5k students in one college town not counted or counted in the wrong district. There are plenty of reasons to doubt census data.
I hit the WSJ paywall so I don't know what they say about it, but according to the AP article on this topic the state's population estimates don't use census data
>The state's population estimate comes from a number of sources, including birth and death counts, the number of new driver's licenses and address changes, school enrollments and federal tax returns.
Indeed. The data is only on a census block level and could have false positives and negatives. Unfortunately, the data is not more granular than the census block level.
As far as I could find for people who aren’t counted by either census workers or themselves like you mentioned, the Census Bureau tries to get their information through other methods.
This can be proxies, like other people who live in the neighborhood, other government data sets like social security records and postal service data, or through "imputation" which leverages statistical modeling based on the data they do know.
All of the sources I can find suggest that the census is extremely accurate at the large city level, and only obfuscated at lower levels because of the data privacy protections they've put into place.
There’s a Wikipedia page for the hamlet, but it’s empty. No population data, etc.
I’d much rather see no data than a LLM’s best guess. I’m guessing a LLM using the data would also perform better without approximated or “probably right” information.
reply