The Official 2020 Census Thread

[jimrtex]

Ocasio-Cortez calls for extending census period

Rep. Alexandria Ocasio-Cortez (D-N.Y.) called Thursday for the government to extend the 2020 Census counting period amid the coronavirus pandemic.

Ocasio-Cortez, a staunch progressive, warned that issues with the census related to the coronavirus could lead to skewed results that influence where federal resources are allocated. The New York lawmaker cited statistics showing that in-person reporting for the census is down, a factor aggravated by COVID-19.

“The U.S. Census Bureau has suspended all fieldwork until April 15, and likely will have to suspend further. Self-reporting is currently down across the country compared to 2010,” she wrote for the liberal think tank Data For Progress. “Congress should strongly consider delaying the Census as much as the law allows, in order to ensure we have a comprehensive survey of the national population.”

The census bureau announced in March it is delaying the deadline for counting from the end of July to mid-August, though Ocasio-Cortez said the change was not enough.

Data For Progress released figures showing 78 percent of voters support extending the census deadline, including large majorities of both Democrats and Republicans.

Ocasio-Cortez wrote that the need for a proper count is even higher in 2020 than in the past after the coronavirus laid bare inequalities that would not be rectified by a flawed census.

“My district, which is overwhelmingly working-class and one of the most diverse in the country, is also one of the districts hardest hit by the virus,” she wrote.

“This crisis has only underscored the need for an accurate count - so that resources can be appropriately and quickly distributed in an emergency. However, as things currently stand, those districts hit the hardest by COVID-19, those districts which already suffer greatest from structural racism and inequity, stand to be the most undercounted,” she added.

https://thehill.com/homenews/house/492093-ocasio-cortez-calls-for-extending-census-period

Not a bad idea.

I think they should extend the self-response period well into the fall and keep on advertising until then.

Then, the census takers should hit the road in the fall and the redistricting data should be delayed until summer next year to give the CB more time to process the results.

Reference Day was April 1, so memories will fade as more time goes by ... but during Coronavirus not many people are going to be evicted anyway over the next months, or move to other places. So, the integrity could be guaranteed.

Besides, AOCs claim that response rates are down from 2010 is a bit early because there are different dynamics at play this year (most respond earlier and online, while there is a 2nd wave of mail-in questionnaires coming back soon).

The Census is not comprehensive. It is basically a headcount. It doesn't show any information about income, jobs, education, citizenship, etc. At best getting an accurate frame of addresses will improve the sampling for the ACS.

The self-response rate is only going to creep up. The reason they initially intended to start the NRFU on May 13, was that by then they were simply not going to be getting any more responses two weeks after the 5th mailing.

If they get 0.5% for the next 35 days that will put them around 63%, which is about the same as 2010 after you account for the difference in denominators.

[jimrtex]

I was comparing Texas daily returns with those for Minnesota. It closely followed the same pattern with peak days, etc., but with fewer responses every day.

For example, the peak day for both was April 1, with Minnesota at 3.5% and Texas at 3.1%.

On April 5 and April 8, Texas was barely ahead of Minnesota (by 0.1%). This may be due to Texas having about 38% more non-respondents at this point.

Cities in Texas with more than 20,000, but 5% below the state average are:

Border cities: Pharr, Laredo, San Benito, Weslaco, San Juan, Harlingen, Brownsville, Mission, Edinburg, McAllen, Eagle Pass, and Del Rio.

College Cities: Huntsville (Sam); San Marcos (Colors Are All The Same (Youtube); Stephenville (Tarleton State); and Kingsville (Texas A&M Kingsville).

Nacogdoches (Steve); College Station (whoop); and Plainview (Wayland Baptist) are a bit low, but have other things going on in their economy.

Colleges shut down in early March. Spring break was typically the second week, and they told everyone not to come back.

Galveston: Lots of beach houses, and West End is mostly Update/Leave. Other smaller cities down on the coast (Rockport, Port Lavaca, etc.) are also low. Some of this may also be due to Harvey.

Big Spring. I'm not really sure why this is low. Perhaps it is oil field workers seeking affordable (or any housing). It is about 40 minutes from Midland. As a larger town, it has more of a housing base. Depending on fluctuations in work, it may have more vacancies. It also has some prisons. With a larger population on the interstate it isn't going to be crushed like some smaller towns where the prison is the only business.

[jimrtex]

It's really surprising to see Mississippi to even be on par with the national average. I wonder what could be driving up the state's responses up. Maybe using the internet more is giving the state a bigger boost relative to other states?

I think there may be an effort by the black churches to get response rates up, similar to Souls to the Polls. In Houston, response rates are clearly racial, but with black areas above Hispanic areas. Missouri City with a large black middle class is one of the better performing cities in Texas.

Lafayette (Oxford) and Oktibbeha (Starkville) are below their neighbors - this is about Mississippi, rather than blacks. I thought this may have been due to the shut down of campuses, but they appear to have been a tad lower in 2010 as well. Perhaps adults in their early 20s are less civically engaged even if they are going to a university.

[jimrtex]

The best city is North River, North Dakota a small suburb north of Fargo, with around 55 people.

In 2010 there were 23 housing units, so we can assume a similar number now since the population is estimated to be unchanged.

The return rate of 91.3% is M/N where M and N are whole numbers (there are no fractional responses or housing units - a garage apartment is a whole housing unit, and the 13th and 14th amendments did away with fractional people).

The non-return rate of 8.7% is M'/N where M' = N-M and is also a whole number.

We can try various values of M' and test whether N is a whole number:

N = M'/8.7%.

If M' is 1, N is 11.494, which is not a whole number.
If M' is 2, N is 22.989, which is approximately a whole number.

Moreover, 2/23 is 8.7% rounded to 1/10 of a percent.

21 of 23 housing units have responded.

2 housing units may be vacant, are home to winter Arizonans/Texans/Floridians who have not returned for the Spring (who can blame them with 20 degree temperatures for midweek); or house social deviants who some would subject to $50,000 fines and 5-year prison citizens.

[jimrtex]

Texas has equal or outperformed Minnesota the last 6 days, as it has cut the differential from 14.2% to 13.9%.

[jimrtex]

McDowell, WV response.

Date	Cumulative	Daily
28-Mar	0.6%
29-Mar	0.6%	0.0%
30-Mar	1.0%	0.4%
1-Apr	2.0%	1.0%
2-Apr	2.2%	0.2%
3-Apr	2.6%	0.4%
4-Apr	2.7%	0.1%
5-Apr	2.7%	0.0%
6-Apr	2.8%	0.1%
7-Apr	2.9%	0.1%
8-Apr	2.9%	0.0%
9-Apr	3.0%	0.1%
10-Apr	3.1%	0.1%

It has always been possible for people to respond, even if they did not receive an invitation with a code. There may have been advertisement around Census Day (April 1) since responses quadrupled in a few days.

It is too early to determine whether sending invitations to update/leave areas has an effect. Mailing of questionnaires to Self-Response areas that had not responded began on April 8. The press release that announced that regularly scheduled program, briefly mentioned that Update/Leave areas will be sent an invitation, but gave no details.

Some areas where census takers were originally going to hand-deliver forms in person will now receive a letter in the mail from the Census Bureau reminding them to participate, .... Even if households don’t receive a letter in the mail, the Census Bureau will drop off a census invitation and paper form as soon as it is safe to do so. Census takers will also follow up with all households that do not respond on their own.

I started to see what happens when you respond to the Census without an invitation. One of the first questions asked said to enter a street address where you might have packages delivered, rather than a Rural Route Box number or Post Office Box where you receive mail. It also asked for a description of where you lived, which might be a way for people who don't live in an obviously residential setting, such as the back of a store, could be contacted. I didn't finish up, because it could be considered hacking (which it was, but in the good sense of the word) and some Australian might be egging them on to give me a 50.000€ fine or 5-year sentence in a correctional centre (gaol).

[jimrtex]

I started to see what happens when you respond to the Census without an invitation. One of the first questions asked said to enter a street address where you might have packages delivered, rather than a Rural Route Box number or Post Office Box where you receive mail. It also asked for a description of where you lived, which might be a way for people who don't live in an obviously residential setting, such as the back of a store, could be contacted. I didn't finish up, because it could be considered hacking (which it was, but in the good sense of the word) and some Australian might be egging them on to give me a 50.000€ fine or 5-year sentence in a correctional centre (gaol).

Austrian.

Good to see you tried getting this information. As long as you didn’t send off your attempt, everything is fine. When I tried filling out the Census questionnaire from here (abroad), it wouldn’t let me. There seems to be an IP-blocker.

Which leads me to the question:

How do Americans respond who are temporarily abroad for some months for work ?

They would usually get a mail with an ID code in their US home.

If they try filling it out online abroad, they cannot access. They have only the chance to:

A) use a VPN to simulate a US IP address
B) fill things out when they are home again/wait for the census taker

Also: there are 5-10 million American citizens living abroad. But they won’t get counted, because they do not „live in the US for most of the time“ on April 1.

öezzie öezzie öezzie öi öi öi

I don't know about those who are visiting another country for work or recreation. The census materials are non-forwardable. They could probably contact a consul for assistance, but I suspect many will not be informed.

Incidentally, I know someone who had just begin studying in Australia (for the fall term, since they do things upside down), who has now been repatriated, and is in quarantine in the basement of someone else's house.

If she were still in Australia, she would not be counted in the US Census. But I think she is still doing classwork online. So is her being outside Australia considered to be temporary, similar to if she were on summer break? And is her place of quarantine a usual place of residence, or just where she was on April 1. I don't know the precise timing, so we can assume she boarded a plane on April 1, and was in Hawaii on March 31, and back on the mainland on April 1.

US Government employees, both military and civilian, and any dependents living with them are counted in the US Census - but government records are used. They are also attributed to states for purposes of apportionment of representatives, but not for redistricting. Except for 2020, military personnel who are deployed abroad are treated differently than those stationed abroad.

On the other hand, Congress requires that persons living abroad be permitted to vote in federal elections. So a person might not count when representatives are apportioned, but can vote for those representatives.

This was an issue back after the 2000 Census when Utah and North Carolina were contesting for the 435th seat. North Carolina has a considerable military presence (Fort Bragg and Camp Lejeune ), while Utah had lots of LDS missionaries who were not counted.

Congress directed the Census Bureau to examine whether overseas civilians (which in US-speak includes Mexico) could be counted in the Census. Congress writes the Census Law so they could require that such people could be counted. The Census Bureau found it would be hard to do since there are not records (you don't need a passport to leave the US - you may need one to get back in), and they were dubious of organizations such as the LDS providing the information.

If the US had a registry, they would have a record of these persons. Census forms could be sent to them. It would also facilitate voting. If someone moved between state and updated their address, they could automatically be registered at their new home, and de-registered at their previous home. They would also have an ID. Those living overseas could vote in person at embassies, consuls, military bases, and other locations. ID's could be checked, and the ballots collected and sent back securely to the US. This would also eliminate the need to mail out ballots 6-1/2 weeks before an election.

[jimrtex]

WA and CO had the biggest increases in the past 10 days with 16% each (relative to their final 2010 rate).

You can clearly see that tech-affiliated states with an educated, younger population and with good internet access had a good headstart with the new online component.

Let's see how the paper questionnaires change this ... I expect old voters to return this more in the coming weeks, pushing up the rates everywhere - but more so in the currently lagging states.

Perhaps. But remember in Audubon County, IA, which had the highest response rate in the state (this may still be true) most were questionnaires. One metric used to determine whether to send out paper forms was the percentage of older persons in the population.

[jimrtex]

The Census Bureau has just released an update on the 2020 Census from the Census Bureau director and the Secretary of Commerce Wilbur Ross.

U.S. Department of Commerce Secretary Wilbur Ross and U.S. Census Bureau Director Steven Dillingham Statement on 2020 Census Operational Adjustments Due to COVID-19

It announces that field operations would begin as soon as possible after June 1, and be completed by October 31, 2000.

They are asking for statutory relief from Congress to delay apportionment counts from December 31, 2020 to April 30, 2021, and delivery of PL94-171 data to July 31, 2021.

In general, this is a 3 to 4 month slip in the schedule.

The PL94-171 data is used by legislatures to redistrict with. Ordinarily it is delivered in time that legislatures that meet in the spring can get their work done early, then spend the last couple of months worrying about preserving their seat/important constituent relationships.

A July 31, 2021 release would result in special sessions, or require year-round legislatures to cancel their summer recess/junkets to legislative conferences, and could impinge on the earliest filing deadlines for 2022.

[jimrtex]

48.1% (+0.2)

That’s a painfully slow daily increase.

Yesterday was Easter Sunday.

There is also no mail delivery on Sundays. Paper forms were just sent out last week (likely still continuing).

[jimrtex]

I'm just curious, does anyone know how prisoners are counted?  I mean are they expected to fill out their own responses or does the facility just provide the info or who is there on April 1.

The Census classifies persons as living in households or living in group quarters.

Rather than trying to define characteristics of groups quarters, they tend to give examples:

"Group quarters include facilities such as college/university student housing, residential treatment centers, military barracks, nursing/skilled nursing facilities, group homes, and correctional facilities."

Filling out the form for a household is somewhat analogous to filling out out the form for a group quarters.

Let's say that you filled out the form for your household, which consists of yourself, your wife, two children, and a friend who is staying in a spare bedroom.

There is somewhat of an assumption that one of the two parents would fill out the form, but the only requirement is that they be 15. If you and your spouse were not literate in any language, you might have your teenager complete the form.

The person filling out the form would provide the relationship of the other persons in the household. If you filled out the form, the relationships would be (1) yourself; (2) spouse; (3,4) child; (5) non-relative. If your daughter filled out the form the relationships would be (1) herself; (2,3) parent; (4) sibling; (5) non-relative. I think the relationship information is more a matter of quality control, than actually trying to plot out family trees, but it could be used for statistical purposes.

So you begin to fill out the form, making sure the address is correct, that you own your home, and then enter your name and that of the four others.

You fill out the page for yourself, sex, age, ethnicity, and race, and whether you might be enumerated somewhere else. You could then go ahead and fill out the pages for the other persons in the household.

But you might choose to let the other persons do their own. Race and ethnicity are a matter of self identity. Perhaps you want your wife to choose whether to specify her last name as Dingo-Blingo, Dingo, or Blingo, all of which she uses at different time. And you want your kids to become familiar with the census. You might not know the details for your friend, such as birthday, and whether he might be reported somewhere else.

Now imagine you are the administrator of a group quarters: The 37 persons have a statistical relationship similar to you and your household. The census bureau wants a headcount of everyone who lived at your house, and the 37 people who live at the prison, nursing home, residential treatment center, or dormitory. It would be really hard to count all the persons individually in either case. Just as you knew everyone living at your house, you know (or at least know of) everyone residing at your facility, and it would be a mess to try to contact individuals directly.

The Census Bureau has several methods for a group quarters administrator to choose from to enumerate. This was worked out over the phone earlier in the year (January to March).

(1) Drop off/pickup. An enumerator drops off a set of individual forms (a single page that is basically one page of the household form for an individual. The administrator would be responsible for distributing the forms and collecting them. The administrator would choose the date and time for drop off and pickup. This might be used for a dormitory, where the administrator doesn't have detailed information about residents, and the residents are autonomous. This is roughly equivalent to Update/Leave.

(2) In-Person interviews. An enumerator visits the group quarters and interviews the residents. An enumeration date and time is arranged. This might be considered similar to the non-response follow-up for households.

(2a) Self-enumeration. This is a variant of in-person interviews, where a staff member of the group quarters is deputized as a census enumerator. Rather than a 80-year old civilian trying to interview prisoners in a maximum security prison, a guard might do so.

(3) eResponse. This appears to be the Census Bureau's preferred method. The administrator downloads a template, and then fills it in with data from electronic records - or possibly keyed in. The instructions warn about making sure not to copy confidential data such as Social Security numbers, so it appears that it might assume cut and paste from administrative records. On completion, a files is uploaded to the Census Bureau. This avoids any field operations.

(4) Paper Response. I don't quite understand this one, since it shows a spreadsheet template. If you can get responses into a spreadsheet, it seems pretty weird that you would have an enumerator come and collect a paper printout.

[jimrtex]

For what it's worth, I lived in university housing at the time of the 2010 Census and definitely received my own individual Census form that I mailed directly back to the Census bureau. (I was also accidentally double-counted in the 2010 Census as I learned later that my parents also listed me as still resident at their home when they completed the Census - oops.)

Group housing mechanics are probably going to be used more for group housing where the individuals occupying it can't reasonably fill out and mail their own forms (or complete a form online, this time around), such as prisons, asylums, nursing homes, etc., and less for places like military barracks or university housing.

You likely filled out this form. Your recollection of mailing the form may be faulty, unless you had a street address. Some universities own what are essentially apartment houses, which they only rent out to students. From the perspective of the Census Bureau these look like household residences.

Informational Copy of the Individual Census Questionnaire (PDF)

Notice the third question. There is something similar on the form your parents filled out. It is conceivable that the Census Bureau detected the double-listing, and that has how you found out.

Whites in their late teens are among the few groups that are undercounted.

The method of response is up to the facilities administrator. Universities are likely to have less information about their students. The person in charge of dorms might not know birthdates or race or ethnicity or the parent's address. The admissions office may have all this information for purposes of quota administration.

I'd expect some group homes for persons with various physical or mental disabilities would want the let the resident function as normally as possible, and fill out their own form.

Residential treatment programs might or might not want residents responding. It might be considered rehabilitative or too overwhelming.

Note that military personnel are handled entirely separately. The DOD is sensitive about tying persons to specific locations, and there are all kinds of issues with being deployed vs. stationed, particularly with national guard deployments to the Middle East. Sailors may be considered to be residing on the ship, even when the ship is docked on April 1.

[jimrtex]

https://twitter.com/hansilowang/status/1249790519796027392

Seems like the April and July 2021 dates would mess with states like Virginia and New Jersey that hold elections that year supposedly with the new maps.

Could they just reuse the old maps again or is that illegal?

They'd have no choice but to reuse the old maps unless they conducted their own state censuses, I think.

There are court cases that you don't have to change election scheduled to fit censuses.

One of the companion cases to 'Reynolds v Sims' was from Maryland which holds elections on a four-year cycle, and the next election was in 1968. The SCOTUS said that was fine.

Alabama itself was on a four-year cycle - 1966, 1970, 1974, ... The federal court still had jurisdiction after the 1970 Census and they ordered Alabama to redistrict. When they didn't they went ahead and drew new districts. They used census enumeration districts rather than election precincts, because they would have had to estimate populations. Election precincts are presumably drawn for the convenience of the voter (e.g. easy to get to a familiar polling place), while enumeration districts are work units for census enumerators. After you do these 50 people over here, go do those 100 people over there.

After they had drawn the maps, the court noted that it gave plenty of time to draw new precincts and tell voters where they had to go to vote.

Mississippi failed to draw legislative maps for the 2011 elections, though they attempted to do so. But a federal court ruled that drawing maps by 2012 complied with the state constitution - the attempt to draw maps for 2011 was aspirational.

[jimrtex]

49.1% (+0.5)

It appears that we are starting to see an uptick from the sending out of paper forms to those  who did not respond to the internet invitation.

Minnesota as of April 1,

49.9% total, 46.5% internet, 3.4% paper and phone (93% internet)

Minnesota as of April 12,

57.5% total, 53.7% internet, 3.8% paper and phone (93% internet)

Change since April 1,

7.6% total, 7.2% internet, 0.4% paper and phone (95% internet)

Minnesota as of April 14,

58.8% total, 54.4% internet, 4.4% paper and phone (93% internet)

Change since April 12

1.3% total, 0.7% internet, 0.6% paper and phone (54% internet).

Daily rate 11 days from April 1 to April 12 (this includes the last part of the Census Day Surge)

0.65% internet, 0.04% paper and phone.

Daily rate 2 days from April 12 to April 14

0.35% internet, 0.30% paper and phone.

McDowell, WV is seeing some increase presumably from the mailout to update/leave areas, but only 0.2% each of the last two day.

[jimrtex]

Best responding city by population class. North River, ND just north of Fargo is best performer of all cities, with only one holdout.

1 to 9	McMullen, AL	Pickens	60.9%	Small all-black town outside Aliceville, west of Tuscaloosa. Likely 14/23 response.
10 to 18	Bonanza, CO	Saguache	4.8%	Almost ghost town survived effort to have its government abandoned. At one time had 36 dance halls and 7 saloons. Likely 1/21 response.
20 to 49	Harbine, NE	Jefferson	69.2%	Southwest of Lincoln, a bit north of the Kansas/Nebraska line. Likely 18 of 26 response
40 to 99	North River, ND	Cass	95.3%	North of Fargo along Red River. Likely 22/23 response.
100 to 199	Oakland Acres, IA	Jasper	83.6%	Exurban Des Moines. Most of town is golf course, but not too exclusive ($20 greens fee on weekends). Likely 56/67 response.
200 to 499	Broecke Pointe, KY	Jefferson	80.4%	11 miles northeast of Louisville. Nice, but not as exclusive as name would suggest. Likely 78/97 response.
500 to 999	Trout Valley, IL	McHenry	84.7%	Southeast part of county near Cook panhandle. Was modular home showcase (166/196? response)

[jimrtex]

Funny how Pendleton County, WV lags their 2010 response rate by more than 40% right now, while Madison County, VA outperforms their 2010 rate by 17% already.

There are not that many counties right now that are outperforming their 2010 rates at this early stage already ...

And the 2 counties are not that far apart, maybe a few 100 kilometers.

Pendelton is extremely rural. By looking at a map of the census tracts you can tell exactly where it is located.

It has about 7000 people with 700 in the county seat of Franklin, and no other towns. About 1.7% of the housing units were self-response, the rest are update/leave. It also appears that there is a large number of second homes in the county, likely for those in Washington DC who prefer the mountains to the beach or bay. It is closer to Washington, DC than Charleston, WV, for those who can't afford the Greenbrier, or prefer something more rustic.  In 2010 about 36% of the housing units in the county were unoccupied. Unlike McDowell where there may be houses that are abandoned, they just don't have permanent residents. These units would have been removed from the denominator for the 2010 response rate.

Madison is 100% self-response. It is a long commute from the NoVa suburbs and has had modest growth in the latter part of the 20th Century. It is on the eastern slope of the Blue Ridge and would be a pleasant spot for retirement if one wanted proximity to medical facilities in NoVa. The last Democrat to carry Madison County was FDR 3rd term in 1940.

[jimrtex]

jimrtex asked me to try to calculate an alternative 2010 final response rate metric. To do this, I had to estimate the percentage of HHs in each tract that are self-report tracts vs. Update/Leave or others. I then multiplied the 2010 Final rate in each tract by the 2020 Self Report rate. Here's the resultant map based on the most recent data released on 4/15, in gif form:

Does the data you have for response rates have population numbers?

If you click on the map on the Census Bureau response page, you end up on public.tableau.com, which is apparently a platform that permits creation of public data visualization.

The rankings page of the visualization allows filtering by population range.

It appears that the visualization believes that the population of The Highlands, KS (sic, it would alphabetized between Thames and Thomastown) and Highland, KS are both 1001. The Highlands is in Reno County and was newly incorporated in 2017. I can not find anything on the census pages, including the 2018 place estimates. Highland is in Doniphan County, and the estimate of 1001 appears to be accurate.

But the response rates for the two cities are different.

The response rate rankings would lead you to believe that The Highlands is the response leader in the 1000 and 1999 range, but it has nowhere the number of houses to support that. It is built around a golf course. Oddly The Highlands golf course is now Crazy Horse golf course. There is also a Crazy Horse in Salina, KS so I suspect the two are related.

Tableau looks interesting. I don't know how difficult the learning curve is.

[jimrtex]

jimrtex asked me to try to calculate an alternative 2010 final response rate metric. To do this, I had to estimate the percentage of HHs in each tract that are self-report tracts vs. Update/Leave or others. I then multiplied the 2010 Final rate in each tract by the 2020 Self Report rate. Here's the resultant map based on the most recent data released on 4/15, in gif form:

Does the data you have for response rates have population numbers?

If you click on the map on the Census Bureau response page, you end up on public.tableau.com, which is apparently a platform that permits creation of public data visualization.

The rankings page of the visualization allows filtering by population range.

It appears that the visualization believes that the population of The Highlands, KS (sic, it would alphabetized between Thames and Thomastown) and Highland, KS are both 1001. The Highlands is in Reno County and was newly incorporated in 2017. I can not find anything on the census pages, including the 2018 place estimates. Highland is in Doniphan County, and the estimate of 1001 appears to be accurate.

But the response rates for the two cities are different.

The response rate rankings would lead you to believe that The Highlands is the response leader in the 1000 and 1999 range, but it has nowhere the number of houses to support that. It is built around a golf course. Oddly The Highlands golf course is now Crazy Horse golf course. There is also a Crazy Horse in Salina, KS so I suspect the two are related.

Tableau looks interesting. I don't know how difficult the learning curve is.

I’ll have to check the data. I think population and household PERCENTS are in the tract relationship file, but not neccesarily raw counts. I’ve now calculated HH numbers, so that could be added for 2020 tracts.

I do have the population data from the 2014-18 ACS for places/MCDs, but I don’t think the data is in the shapefile I uploaded to Mapbox. I could use min show to put a floor in, but would have to program to put a ceiling in.

I’ve never gotten the hang of Tableau, but some find it easier to use. Mapbox requires cobbling together some Javascript to get the maps to work.

How do you get tract names? The csv file just has ID's.

1600000US2070300 Thayer, KS
1600000US2070357 The Highlands, KS???
1600000US2070625 Timken, KS

[jimrtex]

Best Responding Cities

Have added in some more categories. Two more households have responded in Broeck Pointe. I have discounted The Highlands, KS of having more than 1000 persons. It appears to be confused with Highland, KS.

1 to 9	McMullen, AL	Pickens	60.9%	Small all-black town outside Aliceville, west of Tuscaloosa. Likely 14/23 response.
10 to 18	Bonanza, CO	Saguache	4.8%	Almost ghost town survived effort to have its government abandoned. At one time had 36 dance halls and 7 saloons. Likely 1/21 response.
20 to 49	Harbine, NE	Jefferson	69.2%	Southwest of Lincoln, a bit north of the Kansas/Nebraska line. Likely 18 of 26 response
40 to 99	North River, ND	Cass	95.3%	North of Fargo along Red River. Likely 22/23 response.
100 to 199	Oakland Acres, IA	Jasper	83.6%	Exurban Des Moines. Most of town is golf course, but not too exclusive ($20 greens fee on weekends). Likely 56/67 response.
200 to 499	Broecke Pointe, KY	Jefferson	82.5%	11 miles northeast of Louisville. Likely 80/97 response.
500 to 999	Trout Valley, IL	McHenry	84.7%	Southeast part of county near Cook panhandle. Was modular home showcase (166/196? response)
1K to 2K	Hudson, IL	McLean	77.2%	Just above Normal. It appears to be apocryphal that it is named for the burg in New York.
2K to 5K	Pleasant Ridge, MI	Oakland	79.5%	Southeast part of county, 10 Mile Road and Woodward Avenue. Landmark was Hedge's Wigwam.
5K to 10K	Huntington Woods, MI	Oakland	82.5%	Immediately northwest of Pleasant Ridge, includes part of the Detroit Zoo.
10K to 20K	Dardenne Prairie, MO	St. Charles	77.8%	Northwest St. Louis suburb, incorporated in 1983, has octupled in population since 1990 Census.

[jimrtex]

The census maps have been updated with today's data:

https://www.cinycmaps.com/index.php/2020-census

If you open the Options menu on the top right, Min Show will do the following now:

For tract maps: putting a decimal between 0.1 and 1 will white out precincts that have less than that 2020 response rate.
For place/MCD maps: putting an integer > 0 will white out areas with a 2018 population less than what's inputted.

Hit Options again to close the options menu.

I'll have to think about how to add a Max Show if I have time over the weekend.

I can't find The Highlands.

[jimrtex]

Well, the map is of incorporated places, and The Highlands were a CDP in 2018. As you know, Census doesn't provide population estimates for CDPs as part of their Population Estimates Program, just incorporated places. I'd have to use the ACS data for that, probably 5-year in a lot of cases. That's stale.

Is Census providing PCT return data for CDPs or just incorporated places? I haven't checked. If they are, I might be able to add a new map this weekend.

The Highlands was incorporated in 2017, but the Census Bureau may not have had that information in time for 2018 estimates. The Census Bureau response rate map is actually a Tableau window embedded in the 2020census.gov web page.

Once you click on Rankings, you are on an actual Tableau page.

I think that they are just using percentage values (rounded to 0.1%) because there was a ranking tie between Waukesha, WI and Carver, MN which is quite unlikely without rounding.

But there has to be population data, because the filters would not work without it.

I first set a filter of 1000 to 1999, and The Highlands shows up as the best responding city. So I did a little research and came across articles about its incorporation. But their is no Wikipedia page or much other information. I added a link on the Reno County Wikipedia page, and was going to produce at least a minimal page for the city. The city does have a Facebook page, so I might have asked them to fill in some more information. If you do Google search, you are just as likely to come across articles about Highland, which is an older town on a square grid in Doniphan County. The Highlands appears to just be the houses around the golf course on about a section of land.

If you start trimming the filter down, you eventually get to 1001 to 1001 where the only cities in Kansas are The Highlands with a response rate of 77.9% and Highland with a response rate of 34.9%. 1001 is the 2018 estimate for Highland.

I think the relationship between the Census Bureau and Tableau is likely to be somewhat fuzzy. The Census Bureau realize they aren't nimble enough to do good presentation. They concentrate on data collection. If they contracted out the response rate app, there would have been cost overruns and it wouldn't be live on time. The Census Bureau hosts hackathons to produce apps accessing their data.

It is good advertising for Tableau to be hosting a Census Bureau visualization. They would probably be willing to do some hand-holding to help the Census Bureau get an visualization running. Maybe it was an intern.

Maybe if there was no population for The Highlands (N/A not 0), the visualization would crash and someone tweaked the data.

I don't know. I have contacted Tableau and they said it is the Census Bureau's visualization, which is quite reasonable. I emailed the Census Bureau email address for the response rate page, but just today and haven't got a response yet.

The data set (.csv file) has a Geographic ID that would fit between Thayer and Timken, KS, so I think that is the ID for The Highlands, but there is no population data on that site, or any other Census Bureau web site that I can locate.

I just found it on the 2019 shapefiles for places, just over a square mile at 109.04W 46.02N. No population numbers.

The response rates don't appear to include CDP's. It calls them "cities", which is colloquial and not something the Census Bureau would use since the meaning is very state-specific.

p.s. I can't seem to get the crosswalk file to load. Excel says it is trying to repair it, but can not.

[jimrtex]

I thought you were talking about Highlands, TX, which is a CDP. Neverhtheless, The Highlands, NV? are in my data, but not in the 2018 generalized Place shp. I'd need to download the 2019 Place shape, but unless Census recently added it, they haven't released a generalized 2019 file yet.

Here's a chart of all Places (including CDPs, I think) with a response rate over 77%:



Key:
DecFSRR2010 = 2010 Final Response Rate
A16 = 2020 Response Rate from 4/16 Data
DiffA16 = 2020 - 2010
NWA16 = 2020 - 0.494, the national response rate on 4/16/20
10NWA16 = 2020 - 0.665, the national 2010 Self-Response Rate


I'll see if I can find the Crosswalk file on my computer and e-mail it to you.

The Highlands, KS

ID 2070357

Which is on the 2019 Places shapefile.

The 77.9% for the 2020 response rate might well be right. It is midwestern community built around a golf course. I could probably count the the number of houses from the satellite image.

A population of 1001 is clearly wrong.

Adding in the Internet response rate of 32.6%, the only numbers between 50 and 150 housing units that round to 77.9% and 32.6% (rounded to 0.1%).

67/89 total response, 29/89 internet response and
74/95 total response, 31/95 internet response.

So assuming a household average of between 2.1 and 2.9 persons a population between 187 and 262.

I don't think they are showing response rates for CDP's. That they are using the term "cities" would indicate some sort of informal group within the Census Bureau or loosely attached such as interns.

Highlands CDP, TX would not have a 77.9% response rate. CDP's Harris County, 2529.01 and 2529.02 have response rates of 35.9% and 43.0%. The 2010 population was 7522. Very blue collar (ship channel) and a significant Hispanic population (40% in 2010).

[jimrtex]

Oh man ... some people are so stupid to believe they are getting Trump$$$ because they filled out the census form:

https://twitter.com/brittabble/status/1251016187107893248

My Census Form said my community would get millions of $$$ based on filling out the form.

It wouldn't surprise me if the same folks who harvest absentee ballots were harvesting census forms and promising Trump $$$. They could help establish bank accounts for direct deposit and then get cash $$$ for the respondent.

[jimrtex]

New NYT article:

https://www.nytimes.com/2020/04/18/us/coronavirus-census.html

The bad news:

Responses by mail, the standard method in decades past, are muddled because social distancing has reduced the staff that processes forms. The bureau had planned to send a new batch of forms by now to every household that had not responded to earlier mailings, but virus-caused slowdowns at the bureau and the Postal Service have delayed that until the month’s end.

So ... it seems not every household that has not responded yet has gotten a paper questionnaire (planned for April 8-16).

If they get them out by end-April, it would take until mid-May or longer until the returns are processed and show up in the rates.

60-65% response would be pretty good for end-May.

The news article is muddled.

The first two mail outs were sent out to 95% of households. They were customized by address (each address has a unique 12 character pass code).

The first two mail-outs could have been prepared in advance, and likely were sorted in mail route order, and probably injected into the USPS system and regional centers. Even if you discover some late addresses, it is straightforward to add those automatically. They don't have to be pre-sorted. Perhaps it costs an extra fraction of a cent to mail those. Mail is not stuffed in envelopes by hand.

It might be that returned questionnaires have to be opened by hand. You have to check for forms with jelly on them, those with staples, etc. the scanners themselves are probably amazing, detecting upside-down and backwards pages, missing pages, etc. They will be OCR'ed and classified, and likely set aside. Any checking might wait.

By March 30th, midpoint of the 3rd mailing, 36% of households had responded. This would not count those who were held up by USPS delays, or delays in the Census Bureau mail room, where returned forms had not been scanned by Bureau employees standing further apart than usual. Maybe there were also delays due to UV-radiating paper returns. Perhaps the return envelopes are coded. Those might count as a self-response. The same might be true of internet responses are incomplete. Someone mentioned that they found a way to to skip a question. Follow-up can occur later, if they want workers to concentrate on those filling out the form.

5% of the non-respondents were in the update/leave group. So around 59%, in the first group had apparently not responded. It would not hurt to have the post card sent to those who had responded a day or so earlier. The printing of the postcards can be fully automated, since they only have to be customized with the code and address. So the post cards would have gone out on time or nearly so.

By April 12th, midpoint of the 4th mailing, the national response rate was 48%. This mail out included a questionnaire for those who had not received one initially. Again this mailing could be automated.

Since April 12, returns from Minnesota have been 1.4% internet (53.7% to 55.1%) while non-internet returns have gone up 3.8% to 5.1% (1.3% increase). So 1/2 of the most recent returns from the midpoint of the questionnaire mail out have been on paper. These would include time for delivery, filling in, taking to the post office, and return to census bureau, which would explain that it was not instantaneous.

We can conclude that the 4th mailing to the self-response went out on time, or near to it.

The final mailing, a reminder post card wasn't scheduled until April 20. NRFU wasn't scheduled to begin until May 13. This last post card was designed to get those last few to trickle in. A few percent more still cuts down on costs, but expecting slackards to respond voluntarily become extremely remote.

What has been delayed has been the update/leave representing about 5% of housing units, but a smaller share of households. The update portion is to survey in part to discover housing units that would not be expected to respond (vacant, non-habitable, etc.). It was hoped that these areas would have fallen into the five-mailing plan once the initial contact was made.

The Census Bureau announced that they were going to attempt to mail to these housing units. But this might take some coordination with the USPS, since mailings would go to RR boxes or postal boxes.

[jimrtex]

Best Responding Cities

Completes all categories, and updates percentages. One new response in Harbine, and two each in Oakland Acres, and Trout Valley help them keep their leads. New leader in the 1K to 2K category: St. Regis Park, KY

1 to 9	McMullen, AL	Pickens	60.9%	Small all-black town outside Aliceville, west of Tuscaloosa. Likely 14/23 response.
10 to 18	Bonanza, CO	Saguache	4.8%	Almost ghost town survived effort to have its government abandoned. At one time had 36 dance halls and 7 saloons. Likely 1/21 response.
20 to 49	Harbine, NE	Jefferson	73.1%	Southwest of Lincoln, a bit north of the Kansas/Nebraska line. Likely 19 of 26 response
40 to 99	North River, ND	Cass	95.3%	North of Fargo along Red River. Likely 22/23 response.
100 to 199	Oakland Acres, IA	Jasper	83.6%	Exurban Des Moines. Most of town is golf course, but not too exclusive ($20 greens fee on weekends). Likely 58/67 response.
200 to 499	Broecke Pointe, KY	Jefferson	82.5%	11 miles northeast of Louisville. Likely 80/97 response.
500 to 999	Trout Valley, IL	McHenry	85.7%	Southeast part of county near Cook panhandle. Was modular home showcase (168/196 response)
1K to 2K	St. Regis Park, KY	Jefferson	78.1%	12 miles east of Louisville, incorporated in 1953.
2K to 5K	Pleasant Ridge, MI	Oakland	80.0%	Southeast part of county, 10 Mile Road and Woodward Avenue. Landmark was Hedge's Wigwam.
5K to 10K	Huntington Woods, MI	Oakland	83.3%	Immediately northwest of Pleasant Ridge, includes part of the Detroit Zoo.
10K to 20K	Dardenne Prairie, MO	St. Charles	78.5%	Northwest St. Louis suburb, incorporated in 1983, has octupled in population since 1990 Census.
20K to 50K	Andover, MN	Anoka	77.1%	Just north of Coon Rapids, with half of the population added since 1990.
50K to 100K	Maple Grove, MN	Hennepin	73.2%	Northwest of Minneapolis, about 20 minutes in light traffic.
100K to 200K	Centennial, CO	Arapahoe	72.0%	Southeast of Denver, incorporated in 2001 as most populous incorporation at that time.
200K to 500K	Boise City, ID	Ada	64.1%	State capital, "city" is part of official name.
500K to 1M	Seattle, WA	King	61.0%	18th largest city, and fastest growing over 500,000 since 2010, slightly exceeding Austin.
1M to 2M	San Jose, CA	Santa Clara	59.1%	High-tech Silicon Valley giant, areas along southern edge are over 75% response rates.
2M to 5M	Chicago, IL	Cook, DuPage	43.8%	Three cities in this range are tightly bunched.
5M to 10M	New York, NY	Kings, Queens, New York, Bronx, Richmond	39.5%	USA's largest city made up of Brooklyn, Queens, Manhattan, The Bronx, and Staten Island boroughs.