The US Census Bureau thread

[saybanana]

Originally Posted by edale

I

Looking at the map of tracts that grew vs declined, it's pretty clear the Valley + DTLA is responsible for most of the city's population gain. Gentrifying areas such as Highland Park, Echo Park, and East Hollywood all saw big declines, despite increases in development. No doubt a reflection of smaller households, as crowded immigrant neighborhoods transition. Hollywood saw some impressive growth, same with some Westside pockets. South LA was a bit of a mixed bag.
l]

I have family in Highland Park and there were probably 100 or less new units built in the last 20 years. It's pretty much a neighborhood that doesn't drastically look different. In about 2 Square miles. Actually I think all 100 units were built in the last 10 years. The vast majority were home renovation, flipped homes. I think echo park, silver lake, eagle Rock are similar. I do notice that in Highland Park, that there are clearly less children among the new affluent homeowners versus 30+ years ago when many immigrants from Asia and Latin America moved in and had 3+ kids.and schools were overcrowding. The kids grew up and moved out or stayed with parents and had their own kids.

[iheartthed]

Originally Posted by Steely Dan

While I'm sure shitty land management plays a role too, there's a bigger commonality connecting all of the cities on the "biggest losers" list: RUST BELT.

that is not a coincidence.

There doesn't really seem to be a strong correlation between worst performing cities and worst performing large metros, which leads me to believe that some cities problems are more policy than economic. Probably the best example of that in the country is Baltimore, and one could even argue that St. Louis is more like Baltimore than Pittsburgh. For instance, St. Louis metro has never really shrank, and even Detroit's MSA is only down 3% from its peak population. But these are the two worst performing major cities in the country. The worst performing large metros, on the other hand, are places like Pittsburgh (-14%), Buffalo (-14%), and Cleveland (-11%), which are all down double digits at the metro level.

Coincidentally, 2020 is the first census that Buffalo's MSA has grown since 1970, and it is the first census that the city of Buffalo has grown since 1960. Pittsburgh's MSA declined in the 2010s at one of the slowest rates in the past half century, Pittsburgh's MSA grew for the first time in half a century in the 2010s, and Pittsburgh city also came very close to posting the city's first population increase since 1960.

^ for St. Louis City I think it was a combination of national and local policy (highways, sprawl subsidies, and drastic urban renewal demolition) and economic loss. St. Louis had a very strong local economy that was largely sucked away by the coasts following deregulation (banking and advertising in particular). And, of course, the Great Divorce.

[SIGSEGV]

Originally Posted by craigs

Speaking for myself, yes--clicking every tract of 10,000+ ppsm or more, and I'm not aware of any tables yet. And I think I speak for everyone: yes, it does indeed take forever.

Surely there's a way to scrape that off the website

[Crawford]

Are there any (I assume Python?) proficient folks who could scrape this 20k+ data nationally? I want to do NYC region but no way am I going tract by tract.

[LosAngelesSportsFan]

Originally Posted by saybanana

I have family in Highland Park and there were probably 100 or less new units built in the last 20 years. It's pretty much a neighborhood that doesn't drastically look different. In about 2 Square miles. Actually I think all 100 units were built in the last 10 years. The vast majority were home renovation, flipped homes. I think echo park, silver lake, eagle Rock are similar. I do notice that in Highland Park, that there are clearly less children among the new affluent homeowners versus 30+ years ago when many immigrants from Asia and Latin America moved in and had 3+ kids.and schools were overcrowding. The kids grew up and moved out or stayed with parents and had their own kids.

There have been hundreds of units built in Eagle Rock and the immediate surrounding area just in the last few years, mostly on eagle Rock Blvd. They are mostly small lot developments of 15 to 25 units scattered around but it adds up. A lot more needs to be built though

[RST500]

Originally Posted by edale

I was just looking at the 2010 vs 2020 comparison for LA County, and one thing really jumped out at me. LA County's white population declined by 1.7 million! That seems like an astounding decline. By percentage, the white population dropped from 50% to 33%.

I believe the white population is declining, and probably a lot of it is moving to Orange and Riverside Counties. But I have to think some of this is census methodology changing. The amount of people responding "two or more" rose by a million, from 400,000 to 1.4M. Seems like these might be related.

But LA County became less white, less Black, more Asian, and more multi racial/other.

Looking at the map of tracts that grew vs declined, it's pretty clear the Valley + DTLA is responsible for most of the city's population gain. Gentrifying areas such as Highland Park, Echo Park, and East Hollywood all saw big declines, despite increases in development. No doubt a reflection of smaller households, as crowded immigrant neighborhoods transition. Hollywood saw some impressive growth, same with some Westside pockets. South LA was a bit of a mixed bag.

Cool tool to check out: https://data.indystar.com/census/tot...nia/050-06037/

This Google doc from this Twitter handle ( https://twitter.com/IDoTheThinking/s...57977070907398 ) shows the changes in California by each city by demographic:

https://docs.google.com/spreadsheets...gid=1511452762

For instance the City of LA had 1,086,908 Whites in 2010 and 1,126,052 Whites in 2020.

[mhays]

It would be much too hard for the Census Department to put out a spreadsheet of census tracts with land areas and 2010 and 2020 populations. It would also be too hard to use the calc function to derive percentage growth from that. (sarcasm)

They do need to be 100.000% accurate, unlike us. And new tracts would complicate the data. But come on.

[UrbanRevival]

Originally Posted by iheartthed

Coincidentally, 2020 is the first census that Buffalo's MSA has grown since 1970, and it is the first census that the city of Buffalo has grown since 1960. Pittsburgh's MSA declined in the 2010s at one of the slowest rates in the past half century, and Pittsburgh city also came very close to posting the city's first population increase since 1960.

While you're correct that the City of Pittsburgh's population declined very minimally (under 3,000) over the past decade, it's worth offering a correction on the MSA numbers you referenced: the metro grew for the first time since at least the 1960s by nearly 15,000 people (0.6% percent); +27,000 (or 2.2%) in Allegheny County specifically: https://www.post-gazette.com/local/2...s/202108120132

[ChiSoxRox]

Originally Posted by Steely Dan

Do you happen to have a list of the suburban Chicago tracts over 20K ppsm? I'm curious to know where the ones outside of Evanston and the Oak Oark/Cicero/Berwyn cluster are.

For each city:
Tract number, then population

Des Plaines
8062.01 4908
Apartment megacomplex north of downtown

Elmwood Park
8107.01 4865

Berwyn
8150 4254
8149 6557

Cicero
8138.02 4958
8139 6768
8137.02 4199
8136 5443
8135 7169

Oak Park
8128.01 3387
8123.01 4978

Evanston
8094.02 3374
Most of Evanston falls in the 15k to 20k range, just missing. This tract is the downtown high rises.

So effectively Des Plaines and then the city neighborhoods that stayed outside the City. A 10k threshold is much more interesting in the burbs. Then you pick up cores of Waukegan, Joliet, etc. and apartment complexes all over the metro. Even a Kenosha tract.

[iheartthed]

Originally Posted by UrbanRevival

While you're correct that the City of Pittsburgh's population declined very minimally (under 3,000) over the past decade, it's worth offering a correction on the MSA numbers you referenced: the metro grew for the first time since at least the 1960s by nearly 15,000 people (0.6% percent).

You're right and I did see that when I calculated it by hand. The Greater Pittsburgh Wiki page is out of date, and I still had that number in my head when I wrote the response. I'll edit it.

[Steely Dan]

Originally Posted by Crawford

Maybe those increasingly Mexican west side inner burbs like Bridgeview, Burbank, Oak Lawn, etc. Bungalow belt areas that likely have higher than average household sizes.

Nope, census tracts in those communities are pretty much all in that 5K - 10K ppsm range. There's nothing out that way approaching 20K ppsm.

Looking at the census map, the bulk of 20K+ ppsm census tracts in Chicagoland outside of city limits are in the Oak Park/Cicero/Berwyn cluster - basically west side city neighborhoods that avoided annexation.

The only other 20K+ ppsm tracts I could find out in the burbs were 1 in Evanston and 1 in Des Plaines.


In cool news for inner ring Chicago burbs, both Oak Park and Evanston grew by ~5%, which is great growth for older built out pre-war burbs. Cicero and Berwyn only grew by around 2%, but it's still great to see both of them positive too.



EDIT: ChiSox addressed the issue while I was typing my response. Good call on catching that one tract in Elmwood park, another "west side city neighborhood that avoided annexation". That was the only one I didn't find.

Quick question: are Cicero, Berwyn, OP and Evanston the top 5 densest municipalities in IL, along with Chicago.

[ChiSoxRox]

Originally Posted by Steely Dan

Quick question: are Cicero, Berwyn, OP and Evanston the top 5 densest municipalities in IL, along with Chicago.

Wikipedia has a list from the 2010 Census I'll see about updating. But since none of the Illinois cities had drastic changes, the ranking probably still holds. So 7 Illinois municipalities over 10k ppsm.

1. Stone Park. By Maywood and Bellwood, small even by near west suburb standards.

2. Cicero

3. Berwyn

4. Elmwood Park

5. Chicago

6. Oak Park

7. Harwood Heights (entirely inside Chicago by O'Hare)

As for Evanston: 9,445 ppsm in 2010. The north and west sides of Evanston see quite the density drop towards Wilmette and Skokie. Since Evanston is its own township, it's outlined nicely on the Census map.

ETA: Evanston was up 4k in the 2020 census, so it's just past 10k ppsm again. Harwood Heights also had growth, so Evanston is likely 8th. Still, it won't have caught the top 5.

[dave8721]

I did Miami/South Florida which was a pain since there are 10k census tracts extending 70+ miles from the core so finding the 20k ones wasn't easy with those awful colors on the census map.

Combining with previous posts:
Los Angeles: 1,832,775
Chicago: 1,236,115
Miami/Ft.Lauderdale: 368,418
Seattle: 156,537
Twin Cities - 51,643
Milwaukee - 41,037
Madison - 35,514
Columbus - 31,592
Champaign - 22,271

[ChiSoxRox]

I volunteer this weekend to handle NYC for >20k.

I'll see if I can get a python scraper to behave. If not, well I'll sum up everything outside the five boroughs, then probably quicker to tally whatever in the boroughs doesn't make it, and subtract.

[dave8721]

If i had to take a guess I get the list of populations living about 20k would look something like this:
NY
LA
Chicago
Philly
SF
DC
Miami
Boston (DC, Miami and Boston would probably be a close race)
Houston
Seattle (maybe above Houston)
Dallas

[Crawford]

I think Miami and Boston will top DC.

I mean, it's tough to guess bc census tracts are highly variable, but the DC area doesn't have many high density tracts, and lot of the density is really corridor-based rather than neighborhood-based (so for example Connecticut Ave. is lined with apartment blocks but neighborhoods right off the Ave. are low density suburban).

Baltimore should on that list too, probably not too far from Miami, Boston and DC. Baltimore has almost nothing super high density, but there are many rowhouse neighborhoods that should top 20k.

[mhays]

Originally Posted by dave8721

If i had to take a guess I get the list of populations living about 20k would look something like this:
NY
LA
Chicago
Philly
SF
DC
Miami
Boston (DC, Miami and Boston would probably be a close race)
Houston
Seattle (maybe above Houston)
Dallas

For Houston I get 81,279.

[cabasse]

found the api endpoint with a sniffer, and playing with it with postman at mtgis-server.geo.census.gov/arcgis/rest/services/PL_Press_Release_2020/PopDen/MapServer/15/query.
looks like it's a get call with the following params when you click on a tract:

f: json
returnGeometry: true
spatialRef: esriSpatialRelIntersects
geometry: {"xmin":-9395670.984029967,"ymin":4000525.228632153,"xmax":-9395556.32848754,"ymax":4000639.8841745807,"spatialReference":{"wkid":102100,"latestWkid":3857}}
geometryType: esriGeometryEnvelope
inSR: 102100
 outFields: OBJECTID,SHAPE_Length,SHAPE_Area,AFFGEOID,GEOID,STATEFP,COUNTYFP,TRACTCE,NAME,NAMELSAD,STUSPS,NAMELSADCO,STATE_NAME,ALAND,AWATER,POP2020,POPDEN2020,VOTEAGE_PCT,VOTEAGE_TOT,GQ_PCT,GQ_TOT,GQ_IN_PCT,GQ_IN_TOT,GQ_NON_PCT,GQ_NON_TOT,WHITE_PCT,WHITE_TOT,BLACK_PCT,BLACK_TOT,AIAN_PCT,AIAN_TOT,ASIAN_PCT,ASIAN_TOT,NHOPI_PCT,NHOPI_TOT,SORACE_PCT,SORACE_TOT,TWORACE_PCT,TWORACE_TOT,WHITE_C_PCT,WHITE_C_TOT,BLACK_C_PCT,BLACK_C_TOT,AIAN_C_PCT,AIAN_C_TOT,ASIAN_C_PCT,ASIAN_C_TOT,NHOPI_C_PCT,NHOPI_C_TOT,SORACE_C_PCT,SORACE_C_TOT,HISPLAT_PCT,HISPLAT_TOT,NOHISPLAT_PCT,NOHISPLAT_TOT,HU_TOT,HUOCCUPIED_TOT,HUOCCUPIED_PCT,AREALAND_SQMI
outSR: 102100

but you can throw different queries at it based on what i'm seeing here Query (Map Service/Layer)—ArcGIS REST APIs | ArcGIS Developers

as an example, i tried this and it seemed to pull back a list of IDs or coordinates for what i assume are each tract in the state: mtgis-server.geo.census.gov/arcgis/rest/services/PL_Press_Release_2020/PopDen/MapServer/15/query?where=STATE_NAME='Georgia'&f=json

[Phil McAvity]

Does anyone know when the metro area populations will be released from last year's census?