The Official 2020 Census Thread

[jimrtex]

Steve Dillingham, Director of the Census Bureau appeared before the House Committee on Oversight and Reform.

Prepared Statement (PDF)

(1) Census Bureau continues to count everyone, while exploring means of excluding illegal aliens.

(2) The OMB has made a supplemental request for $1 billion.

(3) 62.6% of households have responded. This includes 79 million Internet, 18 million paper, and 1.4 million phone.

(4) Update/Leave was delayed from March to May, and is now 99.6% complete. The remainder are in tribal lands with little or no access.

(5) College students who lived in dorms are counted as part of ongoing group quarters operations. I suspect that college originally intended to let students fill out their form, and instead will provide lists of students - the method more likely to be used by prisons.

The Census Bureau has requested off-campus addresses of students that they may have. Ordinarily, these students would have been counted as part of the self-response (individual students or groups of students sharing an apartment are considered households by the households. Areas near college campuses would have been reached during Early NRFU in April.

(6) Soft NRFU began on July 16 in 6 ACO, and on July 20 in 6 more ACO. 32 ACO began on July 30.

40 more areas were to announced for an August 6 start. These have been announced. They are overwhelmingly in the Northeast, with COVID-19 apparently being the primary concern.

The remaining ACO are expected to begin on August 11. Given the concern about COVID-19, I wonder whether there will be further delays, or counties, etc. to be covered will be very selective.

Plans for a seventh mailing, including forms is being prepared for September. They are also planning use of e-mail and possible texting to low response census blocks. I suspect that they will have very low contact information for non-respondents in low response areas. If someone has lived in their house for 20 years, there may be well-established address-email or phone relationships, but the residents are more likely to have already responded. Someone who has just move into an apartment or left or split is unlikely to have a known relationship, and did not respond to the census.

(7) They are over-hiring temporary workers. Some of those hired are not showing up for training, or are not showing up for work. I suspect temporary workers skew towards older retired persons, or homemakers, who would consider themselves at risk for COVID-19, particularly when training stresses use of PPE equipment. People at home are unlikely to be wearing masks. Some workers will show up one day, and no more. This may not be immediately obvious since enumerators and their supervisors will be working from home. The worker may not inform the supervisor. Work assignments are intended to be made on a batch basis, with workers having some control over their schedule.

(8) The Census Bureau is working with various partners. Among anecdotes:

In Chicago, Mayor Lori Lightfoot recently recruited 33-year-old Adam Hollingsworth ("the Census Cowboy") and his horse "Robin" to ride through the 10 communities in the city with the lowest census response rates to encourage residents to respond to the 2020 Census. Lightfoot compared it to using the Bat-Signal in the fictional city of Gotham.

[Tender Branson]

(3) 62.6% of households have responded. This includes 79 million Internet, 18 million paper, and 1.4 million phone.

What ?

"We are now at a 62.6 percent response rate, with more than 92 million households counted."

79 + 18 + 1.4 = 98.4 million

Does it mean they have not processed 6.4 million forms yet, or did they remove 6.4 million duplicate entries or a combination of those ?

[Tender Branson]

Southern NJ doesn’t start enumerating until August 14 (like most of the US):

ACO Status Map for NRFU

[jimrtex]

Egg Harbor: Atlantic, Camden, Cape May, Cumberland, Gloucester, and Salem counties or

Toms River: Burlington, Monmouth, and Ocean counties?

It doesn't really matter, but I was just curious. The start of NRFU for these areas has not been announced, but will likely be at the time of the general startup of NRFU on August 16.

As I understand it, enumerators will be operating on a piece-work basis. You will get driving instructions on your census laptop, as well as directions to the houses/apartment units that haven't responded. Let's assume a block has 15 houses, and somewhere between 30% and 90% have responded. So there would be between 2 and 11 houses. If it is an apartment complex, the first contact will be with the manager, who may be able to tell you which units no longer are used for housing (converted to storage or a laundry room, etc., or were vacant on April 1). They might also be able to provide contact addresses for tenants who have moved out.

An enumerator might not be able to work Tuesdays, or after 3:00 pm, so the scheduling could take that into account. This will also offer flexibility if some areas require more followup visits. The scheduling software can also take into account where the enumerator lives - I assume you are paid mileage, so you wouldn't be working in both Camden and Atlantic City. Ability to speak and understand Spanish could be useful in Camden, for example.

The director indicated that they are over-hiring. Some enumerators might not show up for in-person training, or might not show up for work, or quit after the first day. Using your six-foot pole (issued as part of your PPE by the Census Bureau) to fend off friendly grandmothers who offer you a lemonade on their porch swing can be stressful.

The director of an Area Census Office (600,000 or so housing units) is at an actual physical location, may have a one-year contract, to help setup the office which may have no more than 10 persons, all working on short-term contracts. But below that it would all be temps. Even the supervisors above the enumerators will be working from their homes.

[jimrtex]

(3) 62.6% of households have responded. This includes 79 million Internet, 18 million paper, and 1.4 million phone.

What ?

"We are now at a 62.6 percent response rate, with more than 92 million households counted."

79 + 18 + 1.4 = 98.4 million

Does it mean they have not processed 6.4 million forms yet, or did they remove 6.4 million duplicate entries or a combination of those ?

I suspect they just goobered the numbers.

The Internet responses, except for non-ID responses should be counted almost instantaneously since they can be verified as entered. Paper responses are counted when they are successfully scanned - (e.g. if a form has 1's that look like 7's or use a funny ligature for a double-s it won't be counted until this can be resolved).

[jimrtex]

Thirty-three-day averages (since 6/28/2010)

Total 1.04% (0.032% per day)
Internet 0.84% (0.026% per day)
Total 0.20% (0.006% per day)

Roughly 41% of responses have been in the last week (21% of the time period). There is a small increase in the share of paper responses suggesting the impact of soft NRFU, where enumerators leave a paper form and another Internet invitation on first contact).

Michigan reached its 2010 response rate on June 15.
Washington reached its 2010 response rate on June 30.
Nevada reached its 2010 response rate on July 20.
Kentucky reached its 2010 response rate on July 27.

Idaho is projected to reach its 2010 response rate on August 1 (gap is 0.1%). Idaho is a soft NRFU state, beginning on July 16.

0 other states in August
6 states in September
9 states in October
7 states in November
11 states in December
12 states + DC  in 2021

Green-7: Reached (4): MI, WA, NV, KY
Green-6: Next 14 days (1): ID
Green-5: Next 28 days (0):
Green-3: August (0):
Red-3: September (6): CO, UT, NH, AZ, VA, ME
Red-4: October (9): AL, OR, MN, CT, NE, OH, SD, MD, NJ
Red-5: November (7): IL, VT, FL, KS, CA, IN, MA
Red-6: December (11): NM, AK, OK, DE, MS, WY, LA, AR, WV, MO, GA
Red-7: 2021 (13): NY, RI, IA, HI, TX, DC, MT, WI, PA, NC, ND, TN, SC

[jimrtex]

This map illustrates the net change over the past four weeks (June 28 to July 26). It is starting of soft NRFU in states such as ID and ME. There is an overall increase likely due to the final postard. The national average was 0.77% (the median was 0.74%, and mode 0.8%, with a long tail on the higher increase side). States with lower response rates tended to have larger increases - this may simply be an effect of a certain percentage of non-respondents responding from a larger pool of non-respondents.

1.6% X
1.5%
1.4%
1.3% X
1.2% XX
1.1% XX
1.0% XXXXX
0.9% XXXXXX
0.8% XXXXXXXXXXXXX
0.7% XXXXXXXXX
0.6% XXXXXXX
0.5% XXXX
0.4% XX

The top states were NM (1.6%), AK (1.3%), ID and WY (1.2%), and ME and VT (1.1%)

The biggest gainers were those with large shares of Update/Leave households, particularly those with later restarts, as well states where soft NRFU has begun. Note, by the time the 4-week measurement period began, Update/Leave had been going on for 8 weeks in the earliest starters such as WV.

[Tender Branson]

Idaho becomes the 5th state to surpass its final 2010 response rate.

This is particularly funny because I read a few years ago that anti-government far-right retirees went to Northern ID in droves.

[jimrtex]

Statement from U.S. Census Bureau Director Steven Dillingham: Delivering a Complete and Accurate 2020 Census Count

Today, we are announcing updates to our plan that will include enumerator awards and the hiring of more employees to accelerate the completion of data collection and apportionment counts by our statutory deadline of December 31, 2020, as required by law and directed by the Secretary of Commerce.

...

The Census Bureau continues its work on meeting the requirements of Executive Order 13880 issued July 11, 2019 and the Presidential Memorandum issued July 21, 2020. A team of experts are examining methodologies and options to be employed for this purpose. The collection and use of pertinent administrative data continues.

The July 21 memorandum called for the exclusion of alien visitors without authorization to be resident in the United States.

[Beet]

Statement from U.S. Census Bureau Director Steven Dillingham: Delivering a Complete and Accurate 2020 Census Count

Today, we are announcing updates to our plan that will include enumerator awards and the hiring of more employees to accelerate the completion of data collection and apportionment counts by our statutory deadline of December 31, 2020, as required by law and directed by the Secretary of Commerce.

...

The Census Bureau continues its work on meeting the requirements of Executive Order 13880 issued July 11, 2019 and the Presidential Memorandum issued July 21, 2020. A team of experts are examining methodologies and options to be employed for this purpose. The collection and use of pertinent administrative data continues.

The July 21 memorandum called for the exclusion of alien visitors without authorization to be resident in the United States.

Time for a lawsuit.

Between this and news of Trump's decision to end the Census before the count is over, it's clearly not happening this year. A stopgap count should be reported to fulfill Constitutional duties, and then the real Census without the politics involved should be conducted again in 2022 if Biden wins.

Another option is to have an "amended" Census by going back in 2021 and plugging the gaps, then using that for redistricting before the 2022 elections.

[jimrtex]

Statement from U.S. Census Bureau Director Steven Dillingham: Delivering a Complete and Accurate 2020 Census Count

Today, we are announcing updates to our plan that will include enumerator awards and the hiring of more employees to accelerate the completion of data collection and apportionment counts by our statutory deadline of December 31, 2020, as required by law and directed by the Secretary of Commerce.

...

The Census Bureau continues its work on meeting the requirements of Executive Order 13880 issued July 11, 2019 and the Presidential Memorandum issued July 21, 2020. A team of experts are examining methodologies and options to be employed for this purpose. The collection and use of pertinent administrative data continues.

The July 21 memorandum called for the exclusion of alien visitors without authorization to be resident in the United States.

Time for a lawsuit.

Between this and news of Trump's decision to end the Census before the count is over, it's clearly not happening this year. A stopgap count should be reported to fulfill Constitutional duties, and then the real Census without the politics involved should be conducted again in 2022 if Biden wins.

Another option is to have an "amended" Census by going back in 2021 and plugging the gaps, then using that for redistricting before the 2022 elections.

Who has standing for either lawsuit? Are the cases ripe?

[jimrtex]

Thirty-six-day averages (since 6/28/2010)

Total 1.15% (0.032% per day)
Internet 0.94% (0.026% per day)
Total 0.21% (0.006% per day)

Twenty-nine-day averages (since 7/5/2010)

Total 1.02% (0.035% per day)
Internet 0.85% (0.029% per day)
Total 0.16% (0.006% per day)

The daily average continues to climb as a week with an increase of 0.38% replaces a week with a 0.11% increase over the 4th of July.

Michigan reached its 2010 response rate on June 15.
Washington reached its 2010 response rate on June 30.
Nevada reached its 2010 response rate on July 20.
Kentucky reached its 2010 response rate on July 27.
Kentucky reached its 2010 response rate on August 1.

0 other states in August
9 states in September
10 states in October
9 states in November
8 states in December
9 states + DC  in 2021

Green-7: Reached (5): MI, WA, NV, KY, ID
Green-6: Next 14 days (0):
Green-5: Next 28 days (0):
Red-3: September (9): CO, UT, VA, NH, AZ, ME, OR, AL, CT
Red-4: October (10): OH, MN, NE, IN, SD, IL, VT, MD, NJ, KS
Red-5: November (9): FL, MA, CA, WV, DE, MS, LS, OK, MO
Red-6: December (8): NM, GA, AK, HI, PA, WY, NY, AR
Red-7: 2021 (10): IA. RI, TX, DC, MT, WI, NC, ND, TN, SC

[jimrtex]

Thirty-day averages (since 7/5/2010)

Total 1.06% (0.035% per day)
Internet 0.89% (0.030% per day)
Total 0.17% (0.006% per day)

Michigan reached its 2010 response rate on June 15.
Washington reached its 2010 response rate on June 30.
Nevada reached its 2010 response rate on July 20.
Kentucky reached its 2010 response rate on July 27.
Kentucky reached its 2010 response rate on August 1.

0 other states in August
9 states in September
10 states in October
9 states in November
9 states in December
8 states + DC  in 2021

Green-7: Reached (5): MI, WA, NV, KY, ID
Green-6: Next 14 days (0):
Green-5: Next 28 days (0):
Red-3: September (9): CO, NH, UT, VA, AZ, ME, OR, AL, CT
Red-4: October (10): MN, OH, NE, IN, SD, IL, VT, MD, NJ, KS
Red-5: November (9): FL, MA, CA, OK, WV, DE, MS, NM, LA
Red-6: December (9): HI, MO, GA, AK, PA, NY, AR, WI, RI
Red-7: 2021 (9): WY, IA, TN, TX, DC, MT, NC, ND, SC

[jimrtex]

This map illustrates the net change over the past four weeks (July 5 to Aug 3). It includes the starting of soft NRFU in states such as ID and ME. There is an overall increase likely due to the final postard. The national average was 0.94% (the median was 0.86%, and mode 0.9%, with a long tail on the higher increase side). States with lower response rates tended to have larger increases - this may simply be an effect of a certain percentage of non-respondents responding from a larger pool of non-respondents.

1.6% X
1.5%
1.4% X
1.3% X
1.2% XXXXX
1.1% XXXX
1.0% XXXXXXXX
0.9% XXXXXXXXXXXXXX
0.8% XXXXXXXXXXX
0.7% XXX
0.6% XXX

The top states were ID (1.6%), ME (1.4%), and CT (1.3%), AK, DC, NM, NY, and VT (1.2%)

The biggest gainers were those with statewide or widespread soft NRFU. The effect of late starting Update/Leave is fading (e.g. AK and NM).

[jimrtex]

This map illustrates the net change over the past week (through August 3)

ID has increased 0.8%, and ME 0.6%. Idaho and Maine are early NRFU states as partially so. 9 states increased 0.5%; 19 states 0.4%; 16 states 0.3%, and 5 states 0.2%.

The national increase over the past week was 0.4%. This included the burp last Tuesday.

[jimrtex]

2020 Self Responses relative to 2010 as of August 3.

100%+ (Blue-6): (5) MI, WA, NV, KY, ID
98%+ (Green-9): (3) CO, VA, UT
96%+ (Green-8): (12) NH, OH, OR, MN, AZ, AL, NE, IN, MD, IL, NJ, CT
94%+ (Green-7): (11) ME, SD, KS, FL, WI, DE, MA, MS, IA, CA, PA
92%+ (Green-6): (10) GA, VT, LA, MO, TN, AR, RI, WV, HI, OK
90%+ (Green-5): (4) NC, NY, ND, WY
88%+ (Green-4): (4) TX, DC, AK, SC, NM
86%+ (Green-3): (3) MT

[jimrtex]

Thirty-one-day averages (since 7/5/2010)

Total 1.15% (0.037% per day)
Internet 0.95% (0.031% per day)
Total 0.20% (0.006% per day)

Michigan reached its 2010 response rate on June 15.
Washington reached its 2010 response rate on June 30.
Nevada reached its 2010 response rate on July 20.
Kentucky reached its 2010 response rate on July 27.
Idaho reached its 2010 response rate on August 1.

Colorado is projected to reach its 2010 response rate on August 30 (gap is 0.9%)

9 states in September
10 states in October
12 states in November
6 states in December
7 states + DC  in 2021

Green-7: Reached (5): MI, WA, NV, KY, ID
Green-6: Next 14 days (0):
Green-5: Next 28 days (1): CO
Red-3: September (9): VA, NH, UT, AZ, ME, OR, AL, CT, OH
Red-4: October (10): MD, MN, NE, IN, SD, IL, VT, NJ, KS, MA
Red-5: November (12): CA, FL, OK, WV, DE, MS, MO, HI, NM, LA, GA, AR
Red-6: December (6): PA, NY, AR, IA, WY, TN
Red-7: 2021 (8): WI, RI, TX, DC, NC, MT, ND, SC

[jimrtex]

Thirty-two-day averages (since 7/5/2010)

Total 1.21% (0.038% per day)
Internet 1.01% (0.032% per day)
Total 0.20% (0.006% per day)

Michigan reached its 2010 response rate on June 15.
Washington reached its 2010 response rate on June 30.
Nevada reached its 2010 response rate on July 20.
Kentucky reached its 2010 response rate on July 27.
Idaho reached its 2010 response rate on August 1.

Colorado is projected to reach its 2010 response rate on August 27 (gap is 0.8%)

10 states in September
10 states in October
12 states in November
7 states in December
5 states + DC  in 2021

Green-7: Reached (5): MI, WA, NV, KY, ID
Green-6: Next 14 days (0):
Green-5: Next 28 days (1): CO
Red-3: September (10): NH, VA, UT, AZ, ME, OR, AL, CT, MD, MN
Red-4: October (10): OH, IN, NE, SD, IL, VT, NJ, KS, FL, CA
Red-5: November (12): MA, HI, OK, WV, DE, MS, MO, NM, LA, GA, AR, PA
Red-6: December (7): AR, NY, WI, RI, TX, IA, WY
Red-7: 2021 (6): DC, TN, NC, ND, MT, SC

[Amenhotep Bakari-Sellers]

Trump 50, Biden 44 in TX, TX isnt happening with Hegar, Castri was the one that should of ran

[Tender Branson]

CO is just 0.7 points short of reaching its final 2010 response rate.

It gained 0.8 points during the past 12 days ...

[jimrtex]

Last week (July 27-August 3) the response rate in counties with 0%-5% Update/Leave Housing Units rose by 0.4% from 64.5% to 64.9%. Meanwhile the response rate in counties with 95%-100% Update/Leave Housing Units rose by 0.6% from 26.8% to 27.4%.

The differential between these two extremes dropped from a 0.4% differntial increase to a 0.2% increase. This suggests that the Update/Leave effect has completely dissipated, or possibly that the final postcard beginning July 22 and soft NRFU are pushing up the response rate in self-response areas.

An alternative explanation is that the response rate in remote Update/Leave areas is so low, that the pool of non-respondents is much higher. For example, the share of non-respondents in the 95%-100% Update/Leave areas was 73.2%, 2.06 times as high as the 35.5% share of non-respondents in in the 0%-5% Update/Leave areas.

The response rate for counties with between 45% and 55% Update/Leave Housing Units rose from 38.0% to 38.3%, an increase of 0.3%. This rate is below the midpoint of the 5% minus, and that 95%+ counties. That is, a mix of the two types in a county has a lower response than two counties, one of each type. The increase was less than that at the extremes, which suggests that there is almost no residual Update/Leave effect.

This is a list of the counties with the largest increase of 0.7% or greater in each state. The threshold has been decreased from 0.8% to reflect a decline in higher-responding counties due to completion of Update/Leave. While soft NRFU pushes the response rate it was only going on for the full week in about 5% of the country.

Using last week's threshold of 0.8%, there would have been a decrease in higher-responding counties from 119 to 52. Using the new threshold there were 90 counties (i.e. 38 of the counties had 0.7% increases).

Half of these 90 counties were in Idaho (17), Missouri (9), Texas (9), South Dakota (6), and Maine (5). The large numbers in Idaho, Missouri, and Maine are due to the soft NRFU in those states.

County	State	Change	Total	UL HU%	UL Restart	NRFU Start	Counties > 0.5%
Northwest Arctic Borough	AK	1.4%	22.8%	44%*			1
Apache	AZ	2.0%	18.4%	96%	11-Jun		2
Sierra	CA	0.7%	39.3%	95%	25-May		1
Hartford	CT	0.9%	68.6%	0%	25-May	20-Jul	4
Lincoln	ID	1.2%	52.6%	9%		16-Jul	17
Dearborn	IN	0.7%	72.3%	1%	18-May	20-Jul	2
Floyd	tie	0.7%	73.7%	0%	18-May	20-Jul
Fayette	IA	1.0%	63.5%	2%	18-May		2
Kearny	KS	0.9%	37.9%	7%		20-Jul	2
St. Bernard Parish	LA	1.0%	60.5%	1%	18-May	16-Jul	4
Waldo	ME	0.8%	53.5%	10%		16-Jul	5
Worth	MO	1.1%	53.6%	10%		16-Jul	9
Rosebud	MT	1.1%	36.3%	70%			4
Banner	NE	1.1%	53.9%	27%	18-May		3
McKinley	NM	1.1%	28.5%	66%			3
New York	NY	0.7%	56.3%	0%	8-Jun		3
Orange	tie	0.7%	61.7%	2%	1-Jun
Manhattan	tie	0.7%	54.8%	0%	8-Jun
Graham	NC	1.2%	30.5%	33%			1
Sherman	OR	0.7%	39.7%	11%			1
Union	PA	0.7%	68.7%	6%		20-Jul	1
Dewey	SD	1.2%	24.1%	100%			6
King	TX	1.1%	33.9%	89%	25-May		9
Starr	tie	1.1%	40.8%	19%	25-May
San Juan	UT	1.6%	32.3%	73%	11-Jun		1
Falls Church Independent City	VA	1.0%	80.0%	0%		20-Jul	4
Pierce	WA	0.7%	68.8%	1%		30-Jul	2
Thurston	tie	0.7%	71.6%	0%		30-Jul
Fremont	WY	1.0%	0.7%	73%	18-May		1

*UL restart is omitted for areas where restart was 12 or more weeks ago (11-May or earlier), and is unlikely to have a substantial ongoing impact.

*The other 56% housing units of the Northwest Arctic Borough are Update/Enumerate which will have almost no self-response.

States with no counties on list (new states are underlined:

AL, AR, CO, DE, DC, FL, GA, HI, IL, KY, MD, MA, MI, MN, MS, NV, NH, NJ, ND, OH, OK, RI, SC, TN, VT, WI, and WY. CT(4), IN(2), IA(1), NC(1), PA(1), VA(4), and WV(3) had counties with relatively high increases, after having none listed last week.

There are several themes associated with counties with large increases:

Significant AIAN population: Northwest Arctic Borough, AK; Apache, AZ; Rosebud, MT; McKinley, NM; Graham, NC; Dewey, SD; and San Juan, UT;

Significant Hispanic population: Starr, TX.

Satmar Hasidic Community: Orange, NY (Palm Tree town, Kiryas Joel). Over a two-week period, responses in Palm Tree increased by 15.3%. The next highest town was Woodbury at 1.5% (Woodbury is immediately east of Palm Tree). If Palm Tree is excluded from the county, the two-week increase is a quite ordinary 0.8%. With Palm Tree included, the countywide increase was 1.7%, high enough to make Orange County the statewide leader, and among the top 100 counties, nationwide.

It appears that the Internet response increase was greater than the total increase - perhaps an Internet response replaced a paper form from some households.

Soft NRFU: Connecticut (Hartford and eastern); Idaho; Indiana (southern); Kansas (all but northeast); Louisiana (New Orleans area); Maine; Missouri (northern); Pennsylvania (central); Virginia (opposite Washington, DC); Washington (Puget Sound); and West Virginia.

Small Size: Sierra, CA (17 households);  Banner, NE (4); Sherman, OR (6); and King, TX (2).

Unknown: Fayette, IA (2nd highest in Iowa was Winneshiek, Fayette's neighbor to north; 3rd and 4th, Chickasaw and Buchanan are also neighbors). Clearly a regional phenomena.

Queens and New York (Manhattan), NY. Perhaps related to a large minority population. Kings (Brooklyn), Bronx, and Rockland were also high relative to the state, so perhaps a publicity campaign.

There were 62 counties (2.0% of 3142 total) that had no gain (less than 0.1%) this last week. This was a modest increase from 37 the previous week. The mode was 0.3%, and the median 0.305%. Overall the range was more concentrated. 33.1% of counties had an increase of 0.3%, 77.8% had an increase of between 0.2% and 0.4%.

[jimrtex]

CO is just 0.7 points short of reaching its final 2010 response rate.

It gained 0.8 points during the past 12 days ...

Three of the four ACO in Colorado started NRFU on July 30. This may have been too late to register an effect by August 3 (which is actually through Sunday the second).

The NRFU effect may have two direct causes: (1) Publicity associated with the start of NRFU - the three NRFU will all be in the immediate Denver media markets; (2) Enumerators leave a census invitation and paper form on their first visit. Even if someone from the household is home, they may prefer to self-respond rather than be interviewed through masks at six-foot distance.

"Muffly"
"Wltzat"
"Muffily"
"IdnUstmt!"

Removing mask:

"Emily"

Removing mask:

"Thank you. Would it be easier if I just left a form for you to fill out?"

[jimrtex]

Thirty-three-day averages (since 7/5/2010)

Total 1.26% (0.038% per day)
Internet 1.05% (0.032% per day)
Total 0.21% (0.006% per day)

Michigan reached its 2010 response rate on June 15.
Washington reached its 2010 response rate on June 30.
Nevada reached its 2010 response rate on July 20.
Kentucky reached its 2010 response rate on July 27.
Idaho reached its 2010 response rate on August 1.

Colorado is projected to reach its 2010 response rate on August 24 (gap is 0.7%)

10 states in September
9 states in October
12 states in November
8 states in December
5 states + DC  in 2021

Green-7: Reached (5): MI, WA, NV, KY, ID
Green-6: Next 14 days (0):
Green-5: Next 28 days (1): CO
Red-3: September (10): VA, NH, UT, AZ, ME, OR, AL, CT, OH, MD
Red-4: October (9): MN, IL, IN, NE, SD, VT, NJ, MA, KS
Red-5: November (12): FL, CA, HI, NM, OK, WV, DE, MS, MO, LA, GA, AR
Red-6: December (8): PA, AR, NY, TX, TN, WI, RI, WY
Red-7: 2021 (6): IA, NC, DC, ND, MT, SC

[Tender Branson]

WA is now 2% above its final 2010 rate, surpassing MI (1.3% above).

WA has made big gains over the past week, MI not.

But both states are far above the US average.

[jimrtex]

Thirty-six-day averages (since 7/5/2010)

Total 1.40% (0.039% per day)
Internet 1.18% (0.033% per day)
Total 0.22% (0.006% per day)

Twenty-eight-day* averages (since 7/13/2010)

Total 1.28% (0.046% per day)
Internet 1.08% (0.039% per day)
Total 0.20% (0.007% per day)

*I decided that I had messed up on the time interval. It is likely I became confused by the date shown on the web page, and the date in the data file. The web page shows yesterday's date (e.g. 8/9/2020) while the file shows the release date of today (8/10/2020). It is possible that the data file released around 1:00 PM EDT (UTC-4) is through midnight of the day before. In any event, the state-weighted average of 1.28% is for files issued on Monday four weeks apart). Using the correct number of days, increases the average change per day by about 3%.

Michigan reached its 2010 response rate on June 15.
Washington reached its 2010 response rate on June 30.
Nevada reached its 2010 response rate on July 20.
Kentucky reached its 2010 response rate on July 27.
Idaho reached its 2010 response rate on August 1.

Colorado is projected to reach its 2010 response rate on August 22 (gap is 0.6%).

Oregon is projected to reach its 2010 response rate on September 1.
Virginia is projected to reach its 2010 response rate on September 2.
Maine is projected to reach its 2010 response rate on September 3.
Utah is projected to reach its 2010 response rate on September 7.
New Hampshire is projected to reach its 2010 response rate on September 7.
Arizona is projected to reach its 2010 response rate on September 7.

8 states in remainder of September
10 states in October
10 states in November
7 states + DC in December
3 states in 2010 (at one time projections stretched into 2020, the latest is now February 2021).

Green-7: Reached (5): MI, WA, NV, KY, ID
Green-6: Next 14 days (1): CO
Green-5: Next 28 days (6): OR, VA, ME, UT, NH, AZ
Red-3: September (8): OH, AL, CT, IN, MD, MN, IL, KS
Red-4: October (10): NJ, NE, CA, SD, HI, VT, MA, FL, DE, OK
Red-5: November (10): LA, NM, WV, MS, MO, GA, AR, PA, AR, NY
Red-6: December (8): TN, TX, WI, RI, IA, WY, NC, DC
Red-7: 2021 (3): ND, SC, MT