In alt.home.repair, on Sun, 18 Oct 2020 15:58:17 -0700 (PDT), trader_4
wrote:


I
remember that they would start hiring and training 4 or 6 months in
advance. This year, my friend... I have to ask her again. She also
applied to be a contact tracer for Covid. She took online training for
one or both of them and then didn't hear back for a long time, but
anyhow, the question is, when it says in 2010 "Enumerators begin
door-to-door operations to collect census data" were they really out
there going to people's doors on the last day of April.

I'm talking about 2010 here. You suggested "above" they weren't ready to
go on the first day.

No, because of Covid it started around early August. But why do they have
to have started in April?

Question of fact about 2010.



I hate to say it but I was depressed and hadn't returned my census form
and somoene called me, someone I knew. If I could remember who it was,
I could and would call him for more details, but I can't remember. Other
than that, I don't know how to find more about this.
This year from what I can tell it was two months, maybe a little less,
starting in early August, ending early Oct. It probably varied a bit by
region of the country. But the real issue here is what percentage was
completed this year versus previous years and if it's substantially
different, does it make any difference in the result. They will never get
to 100%, there will always be a few percent left. That is then filled in
using estimates and models based on the 97% or whatever they have. So
if one year they get to 98%, while another year they get to 96%, not clear
that it really matters.

I agree with all of this, and I don't expect the census to be 100%.

They are saying that they've been to 99% of all the residences in
Maryland,

I suspect what that said is that 99% is complete, meaning they have census
data for 99%.

In Maryland I believe they said what I said.


but they did that partly by loosening standards on what is
visiting a home. Used to be, on I forget, the second or third trip, the
doorknocker could talk to neighbors and if they said how many were
living there (even if they were occasinally wrong) it counted as a
visit. They weakened that to allow it on the first visit. (My friend
went to some places 3 times before finding a resident who was home and
answsed the door, but she's very diligent. She also liked the job, at
least sometimes, from the interesting places she got to go. One was a
house built in the 1800's for the visit by some German king or
something.)

All that is wrong, AFAIK there is no change in how the census data is recorded.
It's always been that after some number of attempts at an address if they
can't be contacted, then they use proxies, ie a neighbor or landlord.

No, it's not wrong. The number of required visits has been lowered.

That's mostly because some people just refuse all attempts, which is why
they just didn't do it online or via mail.

No. They often have sucess on later visits when they don't on the first
or on the 2nd.


Even though she lives there, she saw a lot of streets she hadn't seen.
OTOH, if I did it, I think t hey would assign me around here and I've
been down almost every street already.

But I digress:

What I don't want isn't for it to be 100% but for it not to favor
Republican areas at the direction of a Republican. I cetainly didn't
suspect H.W.Bush of doing that, as we all know, stumpie has shown plenty
of signs of trying to corrupt the Census. (For others, originnally in
June or so it was the census bureau, his own appointee I suspect, who
wanted it extended that month. Then something made them change their
mind. Maybe someone from a state that was going lose a Repuiblican
district pointed out to them how they could increase Republican
representation by shortening the

There were two issues there. One was ending it early and I agree, that looked
like Trump figuring that illegal aliens and the like would be more resistant
to being counted, so ending it early would count less of them. But the census
has been completed now to a high enough extent that this no longer is a
significant issue.

It's been completed to a high enough extent when they spend as much time
doorknocking as they did in the past, unless one can show a convincing
reason why less doorknocking is not needed. And I must have mentined
in this thread or the other how even in 10, maybe 20 or moe of the days
the census bureau says they were working, they were winding down
instead, more than once. Before Sept 30 and the entire 15 days of
Octobter. and those days count only partially,


The other is Trump wants to exclude illegals from the count. I think he's right
on that, it's not what the framers would have intended. On the other hand the

Not only that, none of the census questions say who is an alien and who
is not, who is illegal and who is not. They have an emergency hearing
before the USSC to get permission to exclude illegals but they have no
bovious way to know who they are, and if, IF, there is an indirect way
it will take a long time and will be inaccrate. I hope the SC justices
ridicule the admin lawyer or his client, but they probably won't


Constitution says all people, and I don't think Trump is likely to prevail in the courts. That is one of of those conservative things, ruling on what the
Constitution says, the letter of the law. If there were some contemporary
discussion at the time showing otherwise, I would be persuaded. But there

AIUI at the time, no one was in the country illegally. There were no
rules against entering, no Ellis Island, no Castle Clinton (well there
was, but it was a defense implacemnet, not immigration). Some of the
news stories say that we haven't excluded illegals for 220 years, and
I'm pretty sure they're sincere when they say that, but none existed for
about a century. Maybe the Chinese Exclusion Act of 1882 created the
first ones, assuming some excluded Chinese managed to get in anyhow.
Was there anything earlier?

No, 1875, "Building on the 1875 Page Act, which banned Chinese women
from immigrating to the United States," assuming some Chinese women got
in anyhow, and some probably did.

But someone else writes "The Chinese Exclusion Act of 1882 was the first
significant law restricting immigration into the United States." so
manybe the rule against women didn't apply much. After all, it's
usually men who travel looking for work.

isn't, because at the time the US was not being flooded with illegals.


Mickey



At some point, people that refuse to cooperate
or where contact can't be made, where somone can't be found to provide the
data, another month of pounding on the door isn't going to change the result.

No one pounds on one door for a month.

Whether it will change depends on how thoroughly they've done it
already. Have they really been back 2 more times? That WAS the
standard.

I think they probably doorknocked for 3 full months in 2010, but we
still don't know how long it was in 2020.


As to the completion date, I read under Herman Hollerith

On the Census site above. Hollerith invented the Hollerith card, the
precursor of what was it called, an IBM card? a rectangle with a corner
missing and little rectangualr punch holes. He got the idea from paper
crds that controlled looms as ealy as 1801, that enabled them to make
complicated patterns in cloth. He started at the Census Bureau, but
later started the company that after he left became IBM.

https://www.census.gov/history/www/c...hollerith.html
Herman Hollerith (1860-1929): Hollerith worked briefly for the Census
Office in the run-up to the 1880 census. This experience, along with
some advice from mentor John Shaw Billings, convinced him that the
Census Office desperately needed a better way to tabulate census data
than hand counting. Hollerith was able to invent a device that did just
that: an electric tabulating machine.....

https://www.census.gov/history/www/i...tabulator.html
1888 Competition
Following the 1880 census, the Census Bureau was collecting more data
than it could tabulate. As a result, the agency held a competition in
1888 to find a more efficient method to process and tabulate data.
Contestants were asked to process 1880 census data from four areas in St
Louis, MO. Whoever captured and processed the data fastest would win a
contract for the 1890 census.....

that the 1880
census counting was completed in 1887. So there is no need to be done
by January 2021.

Census day in every year back to 1930 was April 1
In 1920 it was Jan 1, of all things.
In 1910 it was April 15
In 1900 it iwas June 1

The url above goes back to 1790 but I don't have time to read everything
now.