The Official 2020 Census Thread

[jimrtex]

Unweighted average for 5/4

Total 0.27%
Internet 0.20%
Paper & Phone 0.07%

The Internet was up from Sunday, and broke a 6-day slide.

Paper was way down. It appears that they may be reporting paper returns a day later than expected. Perhaps they are having to decontaminate incoming mail, which is coming in containers on semi-trucks.

It appears that there is still significant manual effort in scanning paper forms.

In their operational planning document, they were still considering using extractors, which would open the envelopes on three sides and pull the form (it sounds more like a paper cutter than a letter opener).

Paper forms are batched up to 64 sheets at a time, and the spine must be trimmed so the sheets can be fed. It sounds like there are all kinds of potential failures that must be corrected, such as torn or damaged forms, forms that are stapled, spindled, or mutilated, forms that won't register in the scanner, or won't scan. The sheets are scanned, then OMR'ed (optical mark recognition) and OCR'ed (optical character recognition), handwriting that can't be read has to keyed in (there is apparently AI-assisted software). There is a possibility of having to translate responses from other than English or Spanish. There might be irregular envelopes, or missing pages, etc. The scanners have to be cleaned periodically.

You don't dump 1000 envelopes in a hopper and push a button/

They said that they had to sort the responses before scanning, and were considering whether the USPS could do this. I'm not sure why this is needed, but maybe they like to keep returns from a given state together. This might explain why a state has a good day on some days and not the next.

The envelopes used by the Census Bureau, including response envelopes have a bar code, which the USPS sorting equipment scans and reports to the Census Bureau. They can track returns that are in the mail, and avoid sending nag postcards or doing NRFU if a return is on the way.

[jimrtex]

57.3% (+0.5)

Internet was about 0.2% yesterday, paper/phone around 0.3%

The next 3 days will be important, maybe 0.4% on average before it drops to 0.5 on the weekend.

Sunday target: 59%

And: MI is only 3% away from their final 2010 rate ...

Unweighted average:

Total 0.48%
Internet 0.17%
Paper&Other: 0.31%

The paper returns had an eastern bias, especially if we remove those from Hawaii:

Eastern: 0.41%
Western: 0.26%

Eastern Paper:

0.7% DE
0.6% PA, NC
0.5% VA, TN, MD, WV, SC, OH, AL, MI
0.4% LA, KY, IN, FL, AR, WI, MO, MN, MA, ME
0.3% RI, NY, NJ, NH, MS, IL, GA, CT, IA, DC
0.2% VT

Western Paper:

0.3% WA, TX, OR, OK. NV, CA, AZ, SD, ND, NE, MT, KS,
0.2% WY, NM, ID, CO, AK
0.1% UT

Hawaii Paper:

HI 0.6%

The assignment of the MN to LA tier of states to the east is based on results from last Thursday when the disparity was even greater.

The two paper-processing centers are in Jeffersonville, IN and Phoenix, AZ. Return envelopes are addressed to either of those two processing centers. The assignment likely has to do with USPS concerns such as bulk postage costs and time for delivery (the Census Bureau would likely prefer delivery within two days to avoid sending out mail to "non-respondents"). There may also be concerns about load balancing, and redundancy.

They wouldn't necessarily want the volume to be split 50:50, but they would want enough capacity in each, in case one center went off-line and they had to truck all the forms to the other center. It is possible that some states are divided, so that perhaps mail from Kansas City goes west, and that from St. Louis goes east.

I assume that Hawaii was so high because of transportation issues. There may be flight scheduling issues between Hawaii and the mainland, and between islands. Or perhaps bulk mail from Hawaii goes by ship.

In 2000 and 2010 about 160 million paper forms were processed. The Census Bureau estimated that around 20 million paper forms would be processed in 2020, and their capacity planning assumed 30 million (or around 1 million per day over 30 days). In 2010, there were three processing centers, and the Jeffersonville, IN center was the smallest. But with a much lower volume, it could easily be the larger.

They have received close to 15 million forms, with about 3 million in the last 7 days. If they are having to maintain social distancing and performing decontamination, they might be close to an adjusted capacity.

Responses were over 12% for the rolling 7-day period ending after the Census Day. By 4/10 it had dropped to 3.8%, and by 4/17 to 2.7%. The bottom was 2.4% on April 22.

It has since picked up to just over 3% likely because of sending out paper questionnaires, and the 5th mailing, a postcard threatening to send an enumerator out.

We should start to see a renewed decay, slowed somewhat by the restart of Update/Leave (but remember that is only 5% of the country).

[jimrtex]

If 4.5% of ALL housing units are Update/Leave and we assume that they have a 60% response rate eventually, that should add ca. 3% to the current 57.3% in the coming 2 months.

Only 13/52 states + DC + Puerto Rico are re-starting Update/Leave right now, so it will be in phases.

That’s why I say 2 months, because the re-starts for the various regions and states will take plenty of time, the filling-out takes time, the return and processing takes time ...

And there might not even be 60% response, maybe just 30% or 40%, because many of these visited housing units might be in decay (WV) or not inhabited any longer.

It is a bad assumption that Update/Leave TEA will match the Self Response TEA response rates.

The Update/Leave TEA have a greater share of vacant and non-existent housing. They have more second homes. Remember the former trailer park in Audubon, IA. It has been in decline for some time, and eventually just some trees and a road. Now it has been replaced with an industrial building of some sort. The field worker will go there and confirm that there are no trailers there. But they remain as part of the response rate denominator.

These areas will have a slightly lower rate because of demographics. There will be about 2% bump from these areas.

If someone responds by Internet it counts as soon as they press submit. If the use a paper questionnaire, it counts as soon as it is scanned. You're assuming that it has to be processed. Puerto Rico doesn't count for the US Census.

[jimrtex]

57.7% (+0.4)

Unweighted Average:

Total 38%
Internet 14%
Paper&Phone 24%

The seven-day total is at 3.1%, but will likely dive tomorrow as last Thursday's anomaly (1.0%) is dropped.

There may be a cultural/demographic trend emerging as paper becomes dominant. All southern states were at 3%, While the Midwest, which was the strongest early performer was at 2%.

Hawaii was the top overall performer for the second day in a row.

[jimrtex]

Best Responding Cities

Strathmoor Village, KY becomes largest town to reach 90% response rate (2018 est. 666)

Tarnov, NE joins Harbine, NE atop 20-49 category. Centennial, CO back in lead in 100K-200K table.

1 to 9	McMullen, AL	Pickens	65.2%	Small all-black town outside Aliceville, west of Tuscaloosa. Likely 16/23 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	76.9%	Harbine is tied by another Nebraska hamlet. Southwest of Lincoln, a bit north of the Kansas/Nebraska line. Likely 20 of 26 responses.
(do)	Tarnov, NE	Platte	76.9%	Named by Polish immigrants for Tarnow, Poland, 137 miles north of Harbine.
50 to 99	North River, ND	Cass	100.0%	North of Fargo along Red River. All 23 households have responded.
100 to 199	Lincolnshire, KY	Jefferson	90.5%	Another eastern suburb of Louisville (57/63)
200 to 499	Broeck Pointe, KY	Jefferson	90.7%	Eastern suburb of Louisville (97/107)
500 to 999	Strathmoor Village, KY	Jefferson	90.3%	Eastern suburb of Louisville (262/290?)
1K to 2K	St. Regis Park, KY	Jefferson	88.0%	11 miles east of Louisville (492/559)
2K to 5K	Howards Grove, WI	Sheboygan	88.0%	Farming community in the northern part of the county, seeing modest suburban growth.
5K to 10K	Huntington Woods, MI	Oakland	89.4%	Immediately northwest of Pleasant Ridge, includes part of the Detroit Zoo.
10K to 20K	Dardenne Prairie, MO	St. Charles	86.1%	Northwest St. Louis suburb, incorporated in 1983, has octupled in population since 1990 Census.
20K to 50K	Andover, MN	Anoka	84.7%	Just north of Coon Rapids, with half of the population added since 1990.
50K to 100K	Maple Grove, MN	Hennepin	80.9%	Northwest of Minneapolis, about 20 minutes in light traffic.
100K to 200K	Centennial, CO	Arapahoe	79.8%	Suburb southeast of Denver, largest city in the US at time of incorporation.
200K to 500K	Lincoln, NE	Lancaster	71.0%	State capital and home of University of Nebraska.
500K to 1M	Louisville/Jefferson County metro government (balance), KY	Jefferson	69.1%	Lousville and unincorporated portions of county prior to government merger.
1M to 2M	San Jose, CA	Santa Clara	66.3%	High-tech Silicon Valley giant, areas along southern edge are over 75% response rates.
2M to 5M	Chicago, IL	Cook, DuPage	50.1%	Three cities in this range are tightly bunched.
5M to 10M	New York, NY	Kings, Queens, New York, Bronx, Richmond	46.3%	USA's largest city made up of Brooklyn, Queens, Manhattan, The Bronx, and Staten Island boroughs.

Largest city to have exceeded certain thresholds.

New categories at 5% thresholds added. Dardenne Prairie, MO; Naperville, IL; San Jose, CA; and Phoenix, AZ are leaders in 85%+, 75%+, 65%+, and 55%+ classes, respectively.

Strathmoor Village, KY becomes largest town over 90%. Livonia, MI becomes largest city over 80%% (it is about 35% ahead of Detroit which is just a few miles to the east). Chicago becomes largest city over 50%.

100%=	North River, ND	55
90%+	Strathmoor Village, KY	666
85%+	Dardenne Prairie, MO	13,360
80%+	Livonia, MI	93,971
75%+	Naperville, IL	148,304
70%+	Lincoln, NE	287,401
65%+	San Jose, CA	1,030,119
60%+	San Diego, CA	1,425,976
55%+	Phoenix, AZ	1,660,272
50%+	Chicago, IL	2,705,994

[jimrtex]

58.1% (+0.4)

59% for Sunday is still on target.

62-64% by the end of May and 65-69% by the end of June.

It’s hard to project when exactly the final 2010 rate of 66.5% will be reached ... I guess between June 20 and 30.

70% probably in late July or early August.

Unweighted average:

Total 0.41%
Internet 0.12%
Paper&Phone 0.29%

Internet is down 24% from a week ago.

With the drop of last Thursday's anomaly, the weekly gain is 2.5% which is back where it was when the paper forms begin to come in.

Since April 16, total response has increased by 8.3%, Internet response by 3.7% and Paper by 4.6%. Paper share is 55% over that period.

Through April 16, paper returns were only 12%. The overall paper share is now up to 18%. Around 15.8 million paper forms have been returned.

Clearly this last few weeks have been sustained due to the questionnaires being included with the fourth mailing (previous to that 75% of the households had been told to use the Internet or .............wait.

There was an eastern bias in the paper returns for Thursday:

East 0.33%
West 0.24%

This is either due to a backlog in Jeffersonville being worked off, a backlog in Phoenix, or demographic differences.

Utah had the lowest mail return of 0.1%. But so far only 7% of Utahn responses have been by paper. This is likely due to Mormons who do genealogy online appreciating the value of the old census forms.

If 2.5% weekly rate is sustained, 66.5% will be reached in 3.5 weeks. If it drops to 2.0% the 2010 result will be reached in 4.7 weeks. But those additional weeks will result in further decay.

If the weekly decline is 23% compounded, the final response will be 66.5%.

Date	Weekly	Cumulative
14-May	1.9%	60.0%
21-May	1.5%	61.5%
28-May	1.1%	62.7%
4-Jun	0.9%	63.5%
11-Jun	0.7%	64.2%
18-Jun	0.5%	64.7%
25-Jun	0.4%	65.1%
2-Jul	0.3%	65.5%
9-Jul	0.2%	65.7%
16-Jul	0.2%	65.9%
23-Jul	0.1%	66.0%
30-Jul	0.1%	66.1%
6-Aug	0.1%	66.2%
13-Aug	0.1%	66.3%
20-Aug	0.1%	66.3%
27-Aug	0.0%	66.4%
3-Sep	0.0%	66.4%
10-Sep	0.0%	66.4%
17-Sep	0.0%	66.4%
24-Sep	0.0%	66.5%
1-Oct	0.0%	66.5%
8-Oct	0.0%	66.5%

If the decay is 30%, the final response will be 63.9%. If the decay is 20%, the final response will be 69.0%.

[jimrtex]

Another 9 states will re-start update/leave operations next week (22 in total now):

Link (NPR)

Link (CB)

This is an updated estimated effect of the resumption of the Update/Leave. The number of respondents is rounded to two significant digits. Thus Alabama has between 1350K and 1450K respondents (or an error of up to 3.5%). Oklahoma has between 975K and 985K respondents (or an error of up to 0.5%). The response rate is rounded to 0.1%, and thus has a potential error of about 0.1% (for response rates around 50%).

The error in the number of respondents carries forward into the calculation of SR HU and to a lesser extent in the SR Response rate. As before, the SR HU assumes a 5% response rate from the UL TEA. The first 13 states only restarted on May 6 (and likely have spent the first few days, just getting ready to restart field operations (e.g. PPE still being shipped, and field workers being contacted to find if they are still interested in working, some may be older and reluctant to have any public contact). For the 9 new states, this preparatory work may be started now since it may only involve a few persons in the Area Census Offices (ACO).

The assumption of a 5% response rate, has a minimal effect on the estimated final rate. If 10% from UL TEA have responded, then there will be fewer new responses after Update/Leave is resumed.

The composite response rate for the 22 states is 56.9%. The effect of adding in UL areas is 2.0% to 58.9%.

Estimates of Effect of Update/Leave Resumption

State	Responses	Self-Response	UL HU	SR HU	SR TEA Response	SR Response Rate	UL Responses	Estimate Response Rate	Change
Alabama	1400	56.1%	82	2414	1396	57.8%	40	57.6%	1.5%
Alaska	120	38.6%	110	201	115	57.0%	53	54.0%	15.4%
Arkansas	810	53.1%	77	1448	806	55.7%	36	55.2%	2.1%
Florida	5700	55.8%	172	10043	5691	56.7%	83	56.5%	0.7%
Georgia	2600	54.3%	106	4682	2595	55.4%	50	55.2%	0.9%
Idaho	480	60.9%	71	717	476	66.4%	40	65.5%	4.6%
Kansas	840	62.7%	33	1307	838	64.2%	18	63.9%	1.2%
Kentucky	1300	61.8%	88	2016	1296	64.3%	48	63.9%	2.1%
Maine	370	48.5%	112	651	364	56.0%	53	54.8%	6.3%
Mississippi	780	53.7%	55	1398	777	55.6%	26	55.3%	1.6%
Missouri	1800	58.9%	131	2925	1793	61.3%	68	60.9%	2.0%
Montana	290	49.5%	113	473	284	60.1%	58	58.4%	8.9%
North Carolina	2700	54.0%	165	4835	2692	55.7%	78	55.4%	1.4%
North Dakota	230	57.7%	42	357	228	63.9%	23	62.9%	5.2%
Oklahoma	980	52.1%	188	1693	971	57.3%	92	56.5%	4.4%
Oregon	1200	60.8%	27	1947	1199	61.6%	14	61.4%	0.6%
Pennsylvania	3600	60.9%	197	5714	3590	62.8%	105	62.5%	1.6%
Tennessee	1900	57.8%	9	3278	1900	57.9%	4	57.9%	0.1%
Utah	740	62.1%	89	1103	736	66.7%	50	66.0%	3.9%
Vermont	170	49.2%	62	284	167	58.9%	31	57.3%	8.1%
Washington	2100	63.4%	88	3224	2096	65.0%	49	64.7%	1.3%
West Virginia	450	45.6%	289	698	436	62.4%	153	59.7%	14.1%

[jimrtex]

Another 9 states will re-start update/leave operations next week (22 in total now):

Link (NPR)

Link (CB)

The 9 new states have relatively small percentages of UL.

Alabama	3.3%	6-May
Alaska	35.4%	6-May
Arkansas	5.0%	6-May
Idaho	9.0%	6-May
Maine	14.7%	6-May
Mississippi	3.8%	6-May
Montana	19.3%	6-May
North Dakota	10.5%	6-May
Oklahoma	10.0%	6-May
Tennessee	0.3%	6-May
Utah	7.5%	6-May
Vermont	17.9%	6-May
West Virginia	29.3%	6-May
Florida	1.7%	13-May
Georgia	2.2%	13-May
Kansas	2.5%	13-May
Kentucky	4.2%	13-May
Missouri	4.3%	13-May
North Carolina	3.3%	13-May
Oregon	1.4%	13-May
Pennsylvania	3.3%	13-May
Washington	2.7%	13-May

They may have been chosen because the UL TEA are concentrated in certain TEA. For example in Pennsylvania 184K of 197K UL Housing Units are serviced from three ACO (Cranberry Township, State College, and Allentown).

While Cranberry Township is a wealthy Butler County suburb, the ACO covers the western strip except for Allegheny County. The State College ACO also stretches from border to border, including the most rural areas along the New York border. The Allentown ACO mainly serves the areas from the Lehigh Valley on North. There do not appear to be any UL in Northampton or Lehigh counties, so that the Allentown ACO is directing actual field operations to the north.

ACO areas appear to be based on load balancing so each would need comparable numbers of field workers, particularly during NRFU.

The Reading ACO covers Berks and Schuylkill, but also Chester and Delaware. The few 1000 UL HU for the area are in the outer counties. The Norristown ACO is responsible for Montco and Bucks. The UL TEA are in the areas of Bucks furthest from Philadelphia. Potentially, the UL could be delayed some given the small number (3000).

If we assume 15 minutes per household, 6 hours per day, 5 days per week, 3000 HU would require 25 worker-weeks. That could be 5 workers over 5 weeks (through mid-June) or 10 concentrated in the last two weeks.

In Kentucky most of the UL are in the Lexington ACO which includes the southeastern part of the state, which is similar to West Virginia. The Lexington ACO goes north to include the Cincinnati suburbs, but there are few if any UL TEA in that area. Similarly, the UL in the Louisville ACO are in the western part of the state, remote from any cities other than Owensboro or Bowling Green.

[jimrtex]

58.3% (+0.2)

Nasty.

But rounded on the ranking page, it’s 0.3%

Unweighted Average:

Total 0.25%
Internet 0.10%
Paper&Phone: 0.15%

Internet is down 36% from a week ago Friday.

7-day change 2.3%.

[jimrtex]

Least Responsive Cities

Lots of changes. In 20 to 49 range, added descriptions of Kupreanof, AK and North Lilbourn, MO.

In 50 to 99 category, added description of Kasaan, AK; Ohiopyle, PA; and Quinn, SD. Centrahoma, OK gets a response.

For 100 to 199 range, added description of Aldora, GA. Anawalt, WV; Deer Creek, OK; Interior, SD; and Toyah, TX get responses.

At 200 to 499, Washingtonville, PA gets a response. This leaves Bryce Canyon City, UT as the largest city with zero responses (est. 2018, 224).

For 500 to 999, both St. Francis, SD and War, WV get responses, and St. Francis leaves the table. This means War is now tied with Agua Dulce, TX and Driscoll, TX for last place. Agua Dulce is largest town under 1% (est. 2018, 812).

For 10K-20K, Jackson, WY moves ahead of Española, NM, pushing the town near Santa Fe into last place.

In 200K-500K, Laredo, TX pushes ahead of Newark, NJ.

1 to 9	Fontana Dam, NC	Graham	0.9%	The dam was built during WWII to supply electricity for Alcoa. It has now been converted to a resort, with between 106 and 177 cottages.
10 to 18	Dering Harbor, NY	Suffolk	0.0%	Two towns in this category. Summer homes on Shelter Island off eastern end of Long Island
(do)	Pine Valley, NJ	Camden	0.0%	Often rated top golf course in country, with a very exclusive membership.
20 to 49	Granger, MO	Scotland	0.0%	Six towns in this this category.  Declining farm town just south of Iowa.
(do)	Kupreanof, AK	Petersburg Borough	0.0%	Across Wrangell Narrows from city of Petersburg.
(do)	North Lilbourn, MO	New Madrid	0.0%	Overwhelmingly black town north of Lilbourn which itself has a significant black population (35%). West of New Madrid and Kentucky Bend
(do)	Putnam, OK	Dewey	0.0%	West Oklahoma town, drying up for decades.
(do)	Rendville, OH	Perry	0.0%	Smallest village in Ohio, with 4% of 1890 population.
(do)	Rockham, SD	Faulk	0.0%	About 85 miles west of Watertown, with less than 1/10 of 1920 population
50 to 99	Alexander, KS	Rush	0.0%	Seven towns with no response in this range. Trading post on Santa Fe Trail. Namesake Alexander Harvey's daughter married William Bent.
(do)	Hitchita, OK	McIntosh	0.0%	Small town in eastern Oklahoma.
(do)	Kasaan, AK	Prince of Wales-Hyder Census Area	0.0%	On eastern side of Prince of Wales Island. Take plane to Ketchikan and then ferry to Hollis, and then about 2 hour drive.
(do)	Ohiopyle, PA	Fayette	0.0%	Lumbering town, now tourist town - whitewater on Youghiogheny; Great Allegheny Passage biking and Frank Lloyd Wright's Fallingwater.
(do)	Quinn, SD	Pennington	0.0%	Home of Badlands Observatory, 60 miles east of Rapid City.
(do)	White Bird, ID	Idaho	0.0%	Panhandle town in decline after US 95 bypassed.
(do)	Wood Lake, NE	Cherry	0.0%	Cherry is the big county in the Sandhills of northern Nebraska.
100 to 199	Aldora, GA	Lamar	0.0%	One of three towns with no response for this size. On US 341 bypass around Barnesville, tucked-in behind tire cord factory, and ounty high school.
(do)	Batesland, SD	Oglala Lakota	0.0%	On Pine Ridge reservation, Lakota name of Íŋyaŋ Šála translates to "Red Stone".
(do)	Ratliff City, OK	Carter	0.0%	South Oklahoma, way off I-35.
200 to 499	Bryce Canyon City, UT	Garfield	0.0%	Incorporated in 2007 under quirky law that permitted incorporation by owners of 50% of land value, in this case resort owner Syrett family.
500 to 999	Agua Dulce, TX	Nueces	0.7%	Three-way tie. East of Alice, known for its sweet water.
(do)	Driscoll, TX	Nueces	0.7%	Between Robstown and Bishop with notorius speed-trap.
(do)	War, WV	McDowell	0.7%	Mining town with 1/6 of 1950 population, with 3/400-odd responses.
1K to 2K	Spring City, UT	Sanpete	1.1%	According to Forbes one of prettiest towns in America, with particular Danish Morman flavor.
2K to 5K	Quartzsite, AZ	La Paz	2.1%	RV's congregate at Quartzite in winter. Low response rate may be due to confusion over which housing units are permanent.
5K to 10K	Big Bear Lake, CA	San Bernardino	4.1%	Lots of second homes, as population in area can swell to 100,000 on summer weekends.
10K to 20K	Española, NM	Rio Arriba, Santa Fe	14.2%	25 miles from Santa Fe, and 18 from Los Alamos. The National Lab is a major employer in Española.
20K to 50K	Paradise, CA	Butte	7.0%	May have lost 90% of its population in the 2018 Camp Fire, when 19,000 buildings were destroyed.
50K to 100K	Deerfield Beach	Broward	34.7%	Highest share of Brazilian (Americans) and Haitian (Americans) in the United States.
100K to 200K	Hartford, CT	Hartford	38.9%	Cities are limited in Connecticut, making an easy commute from suburban towns. Hartford is 84% Hispanic or non-white.
200K to 500K	Newark, NJ	Essex	39.8%	Across Hudson from New York City, 88% Hispanic or non-white.
500K to 1M	Detroit, MI	Wayne	44.2%	Has lost 64% of its population (over 1,000,000) since 1950. 92% Hispanic or non-white.
1M to 2M	Philadelphia, PA	Philadelphia	47.2%	Has lost 1/4 of its population since 1950. 65% non-white or Hispanic.
2M to 5M	Los Angeles, CA	Los Angeles	48.2%	Three cities in this range are tightly bunched.
5M to 10M	New York, NY	Kings, Queens, New York, Bronx, Richmond	47.1%	USA's largest city made up of Brooklyn, Queens, Manhattan, The Bronx, and Staten Island boroughs.

Largest cities to not have reached certain thresholds.

Add 45%, 35%, 25%, and 15% categories, led by Detroit, MI; Deerfield Beach, FL; San Luis, AZ; and San Luis, AZ, respectively. San Luis, AZ which is south of Yuma, AZ on the Mexican border, is 26.5% behind any other city with over 30,000 population, giving it the worst performance in five categories.

Newark, NJ takes over <40% category as a Miami, FL past threshild.

Quartzsite, AZ and Spring City, UT cross 2% and 1% thresholds, respectively. Spring City, becomes new leader in the under 2% response rate, while Agua Dulce takes over the under 1% lead.

After Washingtonville, PA gets a response, Bryce Canyon City, UT becomes largest town with zero responses (est. 2018, 224)

<50%	New York, NY	8,398,748
<45%	Detroit, MI	672,662
<40%	Newark, NJ	282,090
<35%	Deerfield Beach, FL	80,863
<30%	San Luis, AZ	33,490
<25%	San Luis, AZ	33,490
<20%	San Luis, AZ	33,490
<15%	San Luis, AZ	33,490
<10%	San Luis, AZ	33,490
<5%	Big Bear Lake, CA	5,281
<2%	Spring City, UT	1,067
<1%	Agua Dulce, TX	812
0%=	Bryce Canyon City, UT	224

[jimrtex]

Largest Cities Above Thresholds

Andover, MN becomes largest city over 85%; Sterling Heights, MI becomes largest city over 80%.

100%=	North River, ND	55
90%+	Strathmoor Village, KY	666
85%+	Andover, MN	33,072
80%+	Sterling Heights, MI	132,864
75%+	Naperville, IL	148,304
70%+	Lincoln, NE	287,401
65%+	San Jose, CA	1,030,119
60%+	San Diego, CA	1,425,976
55%+	Phoenix, AZ	1,660,272
50%+	Chicago, IL	2,705,994

Largest Cities Below Thresholds

Newark, NJ reaches 40%, leaving Hartford, CT as largest city below that level.

<50%	New York, NY	8,398,748
<45%	Detroit, MI	672,662
<40%	Hartford, CT	122,587
<35%	Deerfield Beach, FL	80,863
<30%	San Luis, AZ	33,490
<25%	San Luis, AZ	33,490
<20%	San Luis, AZ	33,490
<15%	San Luis, AZ	33,490
<10%	San Luis, AZ	33,490
<5%	Big Bear Lake, CA	5,281
<2%	Spring City, UT	1,067
<1%	Agua Dulce, TX	812
0%=	Bryce Canyon City, UT	224

[jimrtex]

Here are the re-opened Update/Leave (UL) states, along with the recent (three week) COVID-19 case rate (per million).

State	New Cases PM	Restart UL
New Jersey	6447
Rhode Island	6385
Massachusetts	5932
New York	5210
District Of Columbia	4923
Connecticut	4454
Delaware	3959
Illinois	3798
Nebraska	3591
Maryland	3291
Iowa	2966
Pennsylvania	2205	13-May
South Dakota	2076
Indiana	2001
Virginia	1852
Mississippi	1848	6-May
Kansas	1758	13-May
Colorado	1758
Louisiana	1691
Michigan	1636
Minnesota	1594
Georgia	1461	13-May
New Mexico	1460
New Hampshire	1278
North Dakota	1220	6-May
Tennessee	1198	6-May
Ohio	1184
Alabama	1052	6-May
Wisconsin	1031
Utah	1025	6-May
California	918
Arizona	880
Kentucky	830	13-May
North Carolina	811	13-May
Nevada	792
Washington	756	13-May
Arkansas	736	6-May
Missouri	729	13-May
Texas	716
Florida	690	13-May
South Carolina	657
Oklahoma	505	6-May
Maine	440	6-May
Wyoming	410
Oregon	324	13-May
Idaho	316	6-May
West Virginia	298	6-May
Vermont	199	6-May
Alaska	87	6-May
Hawaii	40
Montana	30	6-May

A very large share of the Pennsylvania UL Type of Enumeration Areas (TEA) are in the three Area Census Office (ACO) in the northern part of the state (Cranberry Township, State College, and Allentown). While Cranberry Township is in suburban Butler County, its area is the western slice of the state, excluding Allegheny. State College ACO is also border to border in the west-central part of the state, and the Allentown office is in the southern tip of its area. Only 3.3% of Housing Units (HU) in Pennsylvania are Update/Leave.

Kansas has flare-ups around meatpacking plants which drive up the overall case rate for a relatively low-populated state. Kansas also has only 2.5% UL TEA.

Hawaii could be opened up, but there may be logistical problems, particularly if inter-island travel is restricted. Perhaps the state is trying to eradicate the virus as has been done in other isolated island areas (Faroes, Falklands).

Wyoming appears to have a flare-up in Fremont County, which might be associated with the Wind River Reservation. Most reservations are UL TEA.

I suspect South Carolina will be in the next tranche.

Texas and California may be considered to be too large and complex to be included in the early stages of a phased-in program.

[jimrtex]

The increase in responses has declined to such a rate that it will be necessary to use longer time changes.

For the last two days, unweighted average:

Overall 0.27%
Internet 0.17%
Paper&Phone 0.10%

The paper rate will be lower because of the weekend.

The week-on-week increase is 1.9%.

[jimrtex]

59.0% (+0.3%)

Closing in on 60% ...

MI is now just 2% away from their final 2010 rate. I need to update my Excel chart.

For the last three days unweighted average:

Total: 0.49%
Internet: 0.25%
Paper: 0.24%

Week-on-week is down to 1.7%

Best states may be those that were distracted by COVID-19 and more rural states. Minnesota is one of the worst recent performers, there is only so much blood you can't get out of a turnip.

[jimrtex]

Best Responding Cities

Huntington Woods, MI becomes largest city (2018 est. 6312) to reach 90% response rate. The suburb, just north of Detroit includes part of the Detroit Zoo.

S.N.P.J. borough, PA goes top in the 10-19 range. The initials, which apparently are the official name of the borough, are derived from Slovenska Narodna Podporna Jednota, a Slovene fraternal organization. The borough was formed so its lodge could serve liquor on Sundays skirting blue laws. Most of the housing is rental cottages and trailers which might not be permanently occupied.

Broecke Pointe, KY retakes the lead for 200-499, and also the lead in the Louisville suburban derby.

Livonia, MI passes Maple Grove, MN for cities between 50K and 100K.

1 to 9	McMullen, AL	Pickens	65.2%	Small all-black town outside Aliceville, west of Tuscaloosa. Likely 16/23 response.
10 to 18	S.N.P.J., PA	Lawrence	5.1%	Initials of Slovenska Narodna Podporna Jednota, a fraternal organization. Borough was formed to get around blue laws at their lodge.
20 to 49	Harbine, NE	Jefferson	76.9%	Harbine is tied by another Nebraska hamlet. Southwest of Lincoln, a bit north of the Kansas/Nebraska line. Likely 20 of 26 responses.
(do)	Tarnov, NE	Platte	76.9%	Named by Polish immigrants for Tarnow, Poland, 137 miles north of Harbine.
50 to 99	North River, ND	Cass	100.0%	North of Fargo along Red River. All 23 households have responded.
100 to 199	Broeck Pointe, KY	Jefferson	91.8%	Another eastern suburb of Louisville
200 to 499	Strathmoor Manor, KY	Jefferson	90.8%	Enclave of eastern Louisville
500 to 999	Strathmoor Village, KY	Jefferson	90.8%	Enclave of eastern Louisville (262/290?)
1K to 2K	St. Regis Park, KY	Jefferson	88.3%	11 miles east of Louisville (494/559)
2K to 5K	Howards Grove, WI	Sheboygan	88.8%	Farming community in the northern part of the county, seeing modest suburban growth.
5K to 10K	Huntington Woods, MI	Oakland	90.3%	Immediately northwest of Pleasant Ridge, includes part of the Detroit Zoo.
10K to 20K	Dardenne Prairie, MO	St. Charles	86.5%	Northwest St. Louis suburb, incorporated in 1983, has octupled in population since 1990 Census.
20K to 50K	Andover, MN	Anoka	85.8%	Just north of Coon Rapids, with half of the population added since 1990.
50K to 100K	Livonia, MI	Wayne	82.2%	Suburb west of Detroit.
100K to 200K	Centennial, CO	Arapahoe	80.9%	Suburb southeast of Denver, largest city in the US at time of incorporation.
200K to 500K	Lincoln, NE	Lancaster	71.9%	State capital and home of University of Nebraska.
500K to 1M	Louisville/Jefferson County metro government (balance), KY	Jefferson	70.4%	Lousville and unincorporated portions of county prior to government merger.
1M to 2M	San Jose, CA	Santa Clara	67.4%	High-tech Silicon Valley giant, areas along southern edge are over 75% response rates.
2M to 5M	Chicago, IL	Cook, DuPage	51.4%	Three cities in this range are tightly bunched.
5M to 10M	New York, NY	Kings, Queens, New York, Bronx, Richmond	48.2%	USA's largest city made up of Brooklyn, Queens, Manhattan, The Bronx, and Staten Island boroughs.

Largest city to have exceeded certain thresholds.

Huntington Woods, MI becomes largest town over 90%. It is a suburb just north of Detroit with a populations of 6312.

Andover, MN becomes the largest city past 85%; Sterling Heights, MI the largest city over 80%; and Louisville/Jefferson County metro government (balance) the largest city over 70%. This constitutes the city of Louisville and unincorporated areas of Jefferson County prior to creation of the Metro government. It excludes the other incorporated cities in the county.

San Diego, CA passes 65%, making it the top performer in both the 60%+ and 65%+ categories.

100%=	North River, ND	55
90%+	Huntington Woods, MI	6,312
85%+	Andover, MN	33,072
80%+	Sterling Heights, MI	132,864
75%+	Naperville, IL	148,304
70%+	Louisville/Jefferson County metro government (balance), KY	620,118
65%+	San Diego, CA	1,425,976
60%+	San Diego, CA	1,425,976
55%+	Phoenix, AZ	1,660,272
50%+	Chicago, IL	2,705,994

[jimrtex]

Cumulative four days (5/9 through 5/13) unweighted average:

Total 0.66%
Internet: 0.32%
Paper: 0.34%

There may be a slightly higher recent returns from areas more impacted by COVID-19, and rural areas. This does not necessarily appear to be tied to renewed Update/Leave, but perhaps to publicity campaigns.

If the four-day trend holds steady, states will hit their 2010 response rate on the following dates.

Michigan	30-May
Washington	2-Jun
Kentucky	3-Jun
Nevada	7-Jun
Alabama	13-Jun
Ohio	13-Jun
Virginia	14-Jun
Oregon	15-Jun
Indiana	19-Jun
Minnesota	19-Jun
New Jersey	20-Jun
Connecticut	20-Jun
Colorado	21-Jun
Nebraska	22-Jun
Illinois	23-Jun
Maryland	23-Jun
Arizona	24-Jun
Delaware	25-Jun
Tennessee	27-Jun
Pennsylvania	27-Jun
Georgia	28-Jun
Florida	29-Jun
New Hampshire	29-Jun
Louisiana	29-Jun
Massachusetts	29-Jun
Rhode Island	30-Jun
Idaho	30-Jun
Mississippi	1-Jul
Wisconsin	2-Jul
Utah	2-Jul
North Carolina	4-Jul
Missouri	9-Jul
Kansas	9-Jul
Maine	9-Jul
California	10-Jul
Hawaii	10-Jul
South Carolina	10-Jul
Iowa	11-Jul
New York	11-Jul
District of Columbia	12-Jul
Arkansas	12-Jul
Vermont	14-Jul
South Dakota	21-Jul
Texas	28-Jul
West Virginia	28-Jul
Oklahoma	1-Aug
North Dakota	8-Aug
New Mexico	14-Aug
Alaska	16-Aug
Montana	20-Aug
Wyoming	7-Sep

[jimrtex]

The Census Bureau will re-start Update/Leave and NRFU-hiring operations next week in 11 addditional states + Puerto Rico:

https://www.census.gov/newsroom/press-releases/2020/2020-resume-additional-locations.html

That means 33 states + PR in total next week.

Among the big states, TX and CA are not part of the group.

Only the Rochester (NY) ACO is reopening. There are 21 ACO in New York, though the 13 in NYC may have no UL TEA.

The Rochester ACO covers western New York, beginning with Wayne, Ontario, Yates, Schuyler, and Chemung; but excluding Erie and Niagara, which are in the Buffalo ACO.

The Window Rock, AZ ACO is also not reopening yet. It is the only ACO covering two states, and is mainly focused on the Navajo Reservation, though it includes whole counties on both sides of the border.

States not reopening yet:

HI: Unknown. Perhaps there are logistical problems traveling between islands.

CA: Large state.

AZ-NM: Portions covered by Window Rock ACO.

TX: Large state (has as many ACO as total from 11 states)

SD: Maybe problems with reservations, which are mostly Update/Leave. In states like MT, WY, ND, CO, AZ census officials may be working with tribal authorities.

MN, WI, IL, MI, OH: lockdown mentality from governors.

SC: Unknown. Perhaps regional centers recommend states, and this region chose LA.

VA,DC,MD,DE,NJ,NY(exc Rochester),CT,MA,NH: AcelaVirus. RI and PA are exceptional. RI has relatively few UL HU and they are remote from Providence. Most of PA UL are rural north and west.

Of 245 ACO, 103 have or will re-open soon. 142 remain closed.

[jimrtex]

Unweighted Average for last 5 days:

Total 0.79%
Internet 0.39%
Paper&Phone 0.40%

Per Day Average:

Total 0.16%
Internet 0.08%
Paper&Phone: 0.08%

Week-Over-Week: 1.2%

Projecting 5-day rate forward, states will reach 2010 response level by the following date, assuming a linear projection, or a 10% per week exponential decay. In the latter case, 18 states and the District of Columbia will not reach the 2010 level by September 1.

State	Linear	10% Decay
Michigan	30-May	1-Jun
Washington	5-Jun	9-Jun
Kentucky	5-Jun	10-Jun
Nevada	10-Jun	18-Jun
Virginia	14-Jun	25-Jun
Oregon	15-Jun	27-Jun
Ohio	17-Jun	30-Jun
Alabama	17-Jun	30-Jun
Minnesota	22-Jun	12-Jul
Indiana	23-Jun	14-Jul
New Jersey	25-Jun	18-Jul
Colorado	25-Jun	19-Jul
Connecticut	25-Jun	20-Jul
Nebraska	26-Jun	21-Jul
Florida	27-Jun	24-Jul
Illinois	27-Jun	26-Jul
Tennessee	28-Jun	27-Jul
Arizona	28-Jun	27-Jul
Maryland	28-Jun	28-Jul
Mississippi	29-Jun	1-Aug
Delaware	30-Jun	3-Aug
Rhode Island	1-Jul	5-Aug
Idaho	3-Jul	13-Aug
New Hampshire	3-Jul	14-Aug
Georgia	3-Jul	14-Aug
Pennsylvania	3-Jul	15-Aug
Louisiana	4-Jul	19-Aug
Massachusetts	5-Jul	22-Aug
North Carolina	5-Jul	23-Aug
Utah	5-Jul	24-Aug
Hawaii	6-Jul	27-Aug
Wisconsin	6-Jul	30-Aug
California	7-Jul
Arkansas	9-Jul
Kansas	12-Jul
New York	12-Jul
Missouri	14-Jul
Maine	14-Jul
Iowa	14-Jul
South Carolina	17-Jul
District of Columbia	19-Jul
Alaska	14-Aug
Vermont	21-Jul
Texas	24-Jul
Oklahoma	24-Jul
South Dakota	24-Jul
West Virginia	27-Jul
North Dakota	13-Aug
New Mexico	22-Aug
Wyoming	27-Aug
Montana	28-Aug

[jimrtex]

Democrats somehow introduced the Census Bureau request for an operations extension and more funding into their new Coronavirus relief bill:

https://nypost.com/2020/05/15/house-democrats-pass-doomed-3t-coronavirus-relief-bill

But considering it’s unlikely to pass in the Senate, the Census-related extension and funding won’t be granted either ...

It would be better if Ds and Rs would pass a unique Census extension and funding package then and not attach it to other measures that have nothing to do with the Census.

Funding bills have to originate in the House. The Senate can amend it.

[jimrtex]

Unweighted Average for last 6 days:

Total 0.91%
Internet 0.45%
Paper&Phone 0.46%

Per Day Average:

Total 0.15%
Internet 0.07%
Paper&Phone: 0.08%

Week-Over-Week: 1.%

Projecting 6-day rate forward, states will reach 2010 response level by the following date, assuming a linear projection, or a 10% per week exponential decay. In the latter case, 23 states and the District of Columbia will not reach the 2010 level by September 1.

State	Linear	10% Decay
Michigan	1-Jun	4-Jun
Kentucky	6-Jun	10-Jun
Washington	7-Jun	12-Jun
Nevada	9-Jun	16-Jun
Alabama	17-Jun	30-Jun
Virginia	18-Jun	1-Jul
Oregon	19-Jun	3-Jul
Ohio	20-Jun	5-Jul
Indiana	22-Jun	11-Jul
New Jersey	25-Jun	17-Jul
Minnesota	25-Jun	18-Jul
Maryland	27-Jun	24-Jul
Colorado	28-Jun	26-Jul
Nebraska	29-Jun	28-Jul
Delaware	29-Jun	29-Jul
Connecticut	30-Jun	31-Jul
Florida	1-Jul	3-Aug
Arizona	1-Jul	5-Aug
Illinois	1-Jul	5-Aug
Tennessee	3-Jul	10-Aug
Mississippi	3-Jul	13-Aug
Idaho	5-Jul	20-Aug
Rhode Island	6-Jul	23-Aug
New Hampshire	6-Jul	26-Aug
Georgia	7-Jul	30-Aug
Hawaii	8-Jul	1-Sep
Utah	8-Jul	1-Sep
Pennsylvania	8-Jul
Louisiana	9-Jul
Massachusetts	9-Jul
Wisconsin	10-Jul
North Carolina	11-Jul
California	12-Jul
New York	13-Jul
Arkansas	14-Jul
Kansas	15-Jul
Iowa	17-Jul
Missouri	18-Jul
Maine	18-Jul
South Carolina	24-Jul
District of Columbia	25-Jul
Vermont	27-Jul
South Dakota	27-Jul
Oklahoma	29-Jul
Texas	30-Jul
West Virginia	3-Aug
North Dakota	16-Aug
Alaska	23-Aug
New Mexico	28-Aug
Wyoming	2-Sep
Montana	4-Sep

[jimrtex]

Update/Leave restart effect. In the following table, states are ordered by share of the housing units in UL TEA. Of the newly added states, only Wyoming will receive a significant boost. A relatively large share of UL HU in Colorado are likely to be second homes.

No adjustment was made for the exclusion of the Window Rock, AZ ACO to the Arizona and New Mexico numbers. No estimate is made for the Rochester, NY ACO.

State	Responses	Self-Response	UL HU	SR HU	UL HU Share	SR TEA Response	SR Response Rate	UL Responses	Estimate Response Rate	Change	Restart
Alaska	130	40.0%	110	215	33.8%	125	57.9%	54	55.0%	15.0%	6-May
West Virginia	460	46.8%	289	694	29.4%	446	64.2%	158	61.4%	14.6%	6-May
Wyoming	140	49.5%	67	216	23.7%	137	63.3%	36	61.1%	11.6%	20-May
Montana	300	50.5%	113	481	19.0%	294	61.2%	59	59.4%	8.9%	6-May
Vermont	180	50.3%	62	296	17.3%	177	59.8%	32	58.2%	7.9%	6-May
Maine	380	49.6%	112	654	14.6%	374	57.2%	54	56.0%	6.4%	6-May
North Dakota	240	58.6%	42	368	10.3%	238	64.7%	23	63.7%	5.1%	6-May
Oklahoma	1000	53.1%	188	1695	10.0%	991	58.4%	93	57.6%	4.5%	6-May
Idaho	490	61.8%	71	722	9.0%	486	67.4%	41	66.5%	4.7%	6-May
Colorado	1600	62.1%	196	2380	7.6%	1590	66.8%	111	66.0%	3.9%	20-May
Utah	760	63.0%	89	1117	7.4%	756	67.6%	51	66.9%	3.9%	6-May
Arizona	1800	55.8%	179	3047	5.5%	1791	58.8%	89	58.3%	2.5%	20-May
New Mexico	490	46.8%	56	991	5.3%	487	49.2%	23	48.8%	2.0%	20-May
Arkansas	830	54.2%	77	1454	5.0%	826	56.8%	37	56.4%	2.2%	6-May
Missouri	1800	59.8%	131	2879	4.4%	1793	62.3%	69	61.9%	2.1%	13-May
Kentucky	1300	63.0%	88	1975	4.3%	1296	65.6%	49	65.2%	2.2%	13-May
Louisiana	1200	53.7%	92	2143	4.1%	1195	55.8%	44	55.4%	1.7%	20-May
Mississippi	790	54.8%	55	1387	3.8%	787	56.8%	27	56.5%	1.7%	6-May
Nevada	770	58.4%	48	1270	3.6%	768	60.4%	25	60.1%	1.7%	20-May
Nebraska	580	65.9%	30	850	3.4%	579	68.0%	17	67.7%	1.8%	20-May
Alabama	1400	57.4%	82	2357	3.4%	1396	59.2%	41	58.9%	1.5%	6-May
Pennsylvania	3700	62.2%	197	5752	3.3%	3690	64.2%	107	63.8%	1.6%	13-May
North Carolina	2800	55.5%	165	4880	3.3%	2792	57.2%	80	56.9%	1.4%	13-May
Washington	2100	64.6%	88	3163	2.7%	2096	66.3%	50	66.0%	1.4%	13-May
Kansas	850	63.6%	33	1303	2.5%	848	65.1%	18	64.8%	1.2%	13-May
Georgia	2700	55.4%	106	4768	2.2%	2695	56.5%	51	56.3%	0.9%	13-May
Florida	5800	56.9%	172	10021	1.7%	5791	57.8%	84	57.6%	0.7%	13-May
Oregon	1200	62.1%	27	1905	1.4%	1199	62.9%	14	62.8%	0.7%	13-May
Indiana	2000	64.1%	39	3081	1.2%	1998	64.8%	21	64.7%	0.6%	20-May
Rhode Island	280	57.3%	6	483	1.2%	280	58.0%	3	57.8%	0.5%	20-May
Iowa	980	66.3%	14	1464	0.9%	979	66.9%	8	66.8%	0.5%	20-May
Tennessee	1900	59.3%	9	3195	0.3%	1900	59.5%	5	59.4%	0.1%	6-May

[jimrtex]

This last week was quite bad, just +1%.

Hopefully this week will be better as more U/L areas restart and submit their forms.

Target: 61% for next Sunday (+1.4)

62-62.5% by end-May.

Seven-day unweighted average:

Total 0.97%
Internet 0.49%
Paper&Phone 0.48%

Daily average:

Total: 0.14%
Internet: 0.07%
Paper&Phone: 0.07%

Week over week. 1.0%, less than half the previous week, and the 10th straight decline.

More likely 0.8% next week.

[jimrtex]

Eight-day unweighted averages:

Total 1.06%
Internet 0.54%
Paper&Phone 0.52%

Daily averages:

Total 0.13%
Internet 0.07%
Paper&Phone 0.06%

[jimrtex]

Nine-day unweighted averages:

Total 1.15%
Internet 0.60%
Paper&Phone 0.55%

Daily averages:

Total 0.13%
Internet 0.07%
Paper&Phone 0.06%

These may be a tad low since they include two weekends.

Week-over-week is down to 0.9%

Faster states in recent days appear to be concentrated in northeastern areas recovering from the coronavirus. Minnesota is among the lowest. This simply suggests that Minnesotans respond by the 2nd or 3rd notice, and don't need an 8th or 9th notice, like slackers elsewhere.

West Virginia and Alaska are slightly above average, but not clearly showing the effect of the renewal of Update/Leave. McDowell County is still at 5.3%.

[jimrtex]

Ten-day unweighted averages:

Total 1.24%
Internet 0.65%
Paper&Phone 0.59%

Daily averages:

Total 0.12%
Internet 0.06%
Paper&Phone 0.06%

Week-over-week is down to 0.7%

If current rate is sustained 5 states will reach their 2010 level in June (MI 9th, WA 15th, KY 15th, NV 19th, and OR 29th). 28 would reach their 2010 level in July, 13 in August, 4 in September, and 1 in August.

The 5 laggards will be helped by Update/Leave: ND, AK, NM, WY, and MT.

If there is a 10% weekly decay:

4 in June
3 in July
4 in August
6 in September
5 in October
3 in November
3 in December
and 23 never.