►
From YouTube: 2020 Census Updates
Description
The census has been in the news this month with an announced rollback of field operations, a return to the statutorily set release date for reapportionment numbers and a promised change in how those numbers will be calculated (with the president’s memorandum signaling undocumented immigrants will not be included). What will these changes—some of which are still in flux or in court—mean for states?
A
Welcome
everyone
we
are
here
today
to
catch
up
on,
what's
going
on
at
the
census,
what's
going
on
with
the
counting
of
people
what's
going
on
with
the
post-counting
processes,
what's
going
on
with
the
deadlines
of
the
census,
a
lot
of
stuff
is
up
in
the
ear
kind
of
been
a
very
unusual
month
for
the
census
and
we're
going
to
get
to
hear
some
about
it.
A
Ncsl
first
started
planning
this
event
in
july,
with
the
idea
of
being
able
to
give
our
members
all
of
the
good
information
they
need
to
know
to
plan
for
next
year's
redistricting
by
the
end
of
august.
So
we
were
going
to
do
this
at
the
end
of
august.
Then
we
realized
that
things
have
changed.
So
we
said:
okay,
well,
we'll
postpone
this
meeting
until
september
10th.
A
A
The
one
thing
that
all
of
that
change
means
is
that
you
will
not
be
getting
final,
firm,
definitive
answers
to
the
key
questions.
You've
got
and
those
questions
are
probably:
when
is
the
data
for
redistricting
going
to
arrive
and
what
will
it
be
made
up
of?
How
will
it
be
constructed
this
time
around?
We
won't
have
final
answers
on
that.
A
I'm
wendy
underhill,
I'm
the
director
of
elections
and
redistricting
at
the
national
conference
of
state
legislatures,
and
I'm
delighted
to
welcome
you
all
here
today.
I
know
that
we've
got
many
legislative
staff,
people,
some
legislators
and
lots
of
other
folks
as
well.
In
fact,
this
is
a
well
attended
event.
If
each
of
you
would
be
willing
to
put
in
the
chat
box
who
you
are,
who
you
work
for
and
if
there's
anything
in
particular
that
you
want
to
get
out
of
today.
A
A
He
is
the
chief
of
the
redistricting
and
voting
rights
data
office
at
the
u.s
census
bureau,
and
he
has
been
a
friend
to
ncsl
for
many
years
and
we
so
much
appreciate
the
way
he
communicates
with
us
and
the
way
he
carries
the
message
to
the
interstices
of
the
census
bureau
that
redistricting
really
is
driving
this
bus
or
should
be
driving
this
bus
at
least,
and
then
we're
going
to
hear
from
tyrion
lowenthal.
A
Who
is
one
of
the
nation's
most
well-known
census
experts
and
she
was
formerly
the
staff
director
for
the
u.s
house
census
oversight
committee,
I'd
like
to
think
of
tyrion
as
the
consummate
census
geek,
and
that
is
a
title
I
aspire
to
earn
myself
at
some
point
before
we
turn
it
over
to
christy.
I
do
want
to
invite
everybody
to
do
exactly
what
you're
doing.
Maybe
I
already
said
that
we've
scheduled
this
for
one
hour.
A
My
thought
is
that
we
would
take
the
questions
that
are
the
most
general
first
and
take
the
more
technical
questions.
Last
with
the
idea
that
if
we
do
run
over
the
one
hour
mark,
those
of
you
who
need
to
get
off
will
have
gotten
the
stuff,
that's
more
generally
applicable,
and
those
of
you
who
are
into
the
data
into
the
weeds
would
stay
on
for
the
more
technical
things
we'll
see
how
that
goes
so
christy,
I
know
you're
busy
admitting
people
right
and
left.
Can
you
do
two
things
at
once?
I'm
sorry!
A
B
Great
hello,
everyone
I'm
going
to
give
a
very
quick
brief
update
on
census
litigation.
B
B
B
But
the
the
bureau
is
temporary,
stopped
from
winding
down
or
altering
any
census
field
operations
until
that
new
court
hearing
is
scheduled
additionally,
with
regard
to
apportion
apportionment
counts.
B
Back
in
july,
the
president
had
directed
the
department
of
commerce
to
only
include
non-citizens
who
were
legally
in
the
u.s
in
the
2020
census.
State
population
totals
and
those
numbers
would
be
used
for
apportioning
seats
in
the
us
house
of
representatives,
thursday.
A
couple
weeks
ago,
a
special
three-judge
court
in
new
york
had
issued
an
opinion
and
order
that
blocked
that
directive
under
the
law.
B
A
total
population
tabulation,
is
what
the
commerce
secretary
is
directed
to
report
to
the
president,
and
the
president
in
turn,
is
then
to
head
off
to
congress
the
number
of
persons
in
each
state.
Therefore,
the
court
order
bans
the
commerce
secretary
to
report
to
the
president
any
information
about
the
number
of
unauthorized
non-citizens
in
each
state.
So
what
will
happen
next
from
what
I
heard
from
this
morning
from
terry
m,
that
the
case
is
being
appealed
to
the
u.s
supreme
court.
A
C
I
am
ready
to
go
and
I
am
planning
to
share
slides
so
just
sound
check.
You
can
hear
me
all.
C
C
As
wendy
mentioned,
we
have
a
long
history
of
working
together
to
make
sure
that
we
both
have
the
information
coming
into
the
census
bureau
that
the
states
want
to
provide
and
that
we're
providing
information
that
states
need
back
out
from
the
census.
So
I
appreciate
the
opportunity
for
this
this
chance
to
talk
to
you.
C
It
has
been
a
bit
of
a
moving
target
as
to
when
we
were
going
to
be
able
to
talk
to
you
and
what
we'd
be
able
to
share
and-
and
I
think
wendy's
instincts
were
correct-
that
we,
we
just
needed
to
have
a
moment
to
talk
to
you
about
whatever
we
could
and
that
this
is
a
good
moment
for
that.
Today's
top
our
presentation
might
be
a
little
disjointed
seeming
and
that
I
have
some
topics
that
are
not
always
addressed
in
the
same
setting.
C
So
I'm
going
to
talk
about
the
progress
on
the
2020
census,
I'm
going
to
talk
a
little
bit
about
imputation
and
editing
virtual
response
data.
That's
something
that's
been:
we've
got
frequent
questions
on,
I'm
gonna
have
a
few
reminders
about
the
redistricting
data
program
and
talk
as
best
as
I
can
about
deadlines.
C
Although
there
is
you'll
notice,
when
I
get
there,
there's
there's
some
fuzziness
there
still
and
then
our
plans
for
disclosure
avoidance
and
then
finally,
I'm
going
to
talk
about
the
cvap
special
tab
that
we
are
working
on
for
from
the
2020
census.
C
So
I
want
to
start
with
the
most
important
message
that
you
can
take
away
from
today,
and
you
can
share
with
constituents
and
your
friends
and
your
family,
and
that
is
that
it's
not
too
late
to
respond
to
the
census.
You
can
still
respond
online,
you
can
still
respond
by
phone
or
you
can
still
respond
by
paper
if
you
have
still
have
that
paper
questionnaires
that
we
sent
earlier
in
the
year.
In
addition,
if
you've
already
responded
and
you
get
a
visit
from
a
census
worker-
please
work
with
them.
C
There's
many
reasons
why
an
enumerator
might
revisit
an
address
that
has
already
responded,
most
of
which
involves
quality
assurance
work.
Your
cooperation
with
these
hard-working
folks
is
important
and
valuable,
and
second,
I
also
wanted
to
point
out
that
we
have
two
new
operations
that
have
started
in
september.
C
But
let's
talk
about
what's
already
finished,
the
census
bureau
was
able
to
successfully
complete
our
remote
alaska
operation.
This
is
almost
like
a
mini
census
for
those
outlying
areas
of
alaska
and
covers
address,
listing
housing
unit
group
quarters,
enumeration
and
transitory
location
enumerations.
C
C
We've
had
627
cases
added,
as
the
work
has
progressed
completed
cases
is
47
947,
so
it
gives
us
a
76.7
percent
completion
rate
and
that's
as
of
the
15th.
So
this
is
a
couple
days
old,
so
we're
even
farther
along
than
that.
C
Another
operation
that
I
get
a
lot
of
questions
about
is
about
to
start
and
it's
called
service
based
enumeration.
This
is
how
we
capture
people
experiencing
homelessness.
It's
done
by
doing
a
rapid
canvassing
of
locations
where
services
are
provided,
shelters
and
soup
kitchens,
but
also
by
enumerating
locations
where
people
experience
experiencing
homeless
that
are
likely
to
congregate.
C
C
Now
the
main
operation
at
this
time
that
has
most
of
the
public's
attention
is
our
non-response
file
operation.
We
shorten
it
to
north
nerfu
and
it's
going
well
as
of
tuesday,
the
15th
we
were
82
complete,
which
exceeded
our
goal.
For
that
time,
which
was
at
77.8
percent,
the
numerators
are
working
hard,
they're
averaging
19.3
hours
per
week,
and
that's
measured
as
of
the
week
of
september
3rd
to
the
9th.
C
The
census
bureau
has
put
together
some
tools
to
help
the
public
track.
The
progress
of
norfolk,
a
response
rate
by
state
page,
has
the
self-response
rate,
the
enumeration
in
our
food
rate
and
then
the
total
enumeration
rate
for
each
state,
the
district
of
columbia
and
puerto
rico.
The
tables
on
this
slide
are
from
the
12th,
but
actually,
as
of
today,
with
more
updated
data,
we
have
two
states
that
have
crossed
our
99
mark,
idaho
and
west
virginia
and
99
mark
is
important
because
it's
the
census,
stated
quality
goal
for
each
state.
C
And
for
those
of
you
who
aren't
aware
the
census
conducts
our
our
field
work
through
our
area,
census
offices,
scattered
all
across
the
country,
and
we
to
create
a
sub
state
level
version
of
of
the
data
that
is
shown
in
that
state
table
that
I
just
showed.
We've
created
this
mapper,
where
you
can
monitor
each
individual,
aco
and
drill
in
to
see
what
what
their
rate
is
in
our
food
completion.
So
it's
just
another
tool
to
help.
You
identify
the
areas
that
are
progressing
through
the
norfolk
operation.
C
So
now
I'm
going
to
switch
gears
a
little
to
talk
about
what
happens
once
we
actually
have
those
responses
in-house
the
census.
Responses
go
through
a
series
of
transformations
before
they
get
turned
into
data
for
the
public,
and
this
is
a
very
overly
simplified
description
of
that
typical
flow.
But
those
responses
come
in.
They
become
part
of
the
decennial
response
file.
Of
course,
we
have
to
merge
paper
responses
that
come
in
through
our
paper,
operation
and
the
internet
and
from
the
phone
they
get
transformed
and
become
the
census
unedited
file.
C
This
is
the
the
final
census
count
of
people
in
living
quarters,
and
this
is
the
file
that
we
produce
apportionment
counts
from
once
we
have
that
file.
We
move
on
to
create
the
census
edited
file.
This
is
where
we
ensure
each
record
has
valid
values
for
the
major
characteristics,
and
these
are
the
counts
that
you're
used
to
seeing
that
are
used
for
redistricting
those
accounts
get
processed
through
the
disclosure
avoidance
system
before
that
redistricting
data
is
produced
and
then
come
out
of
our
tabulation
system
in
the
tables
that
you're
familiar
with.
C
C
So
when
transforming
these
responses
into
the
subsequent
formats
is
when
we
encounter
activities
or
concepts
of
imputation
and
editing,
these
show
up
as
essentially
two
types
of
actions.
We
have
count
imputation
and
we
have
characteristic
editing
and
imputation
count.
Imputation
is
used
to
replace
missing
count.
Information
from
housing
units
that
have
been
identified
and
verified
as
occupied
count
imputations
conducted
during
the
construction
of
our
unedited
file
and
the
complete
unedited
file
provides
us
with
the
first
official
accounts
from
the
decennial
census,
which
is
those
total
population
counts.
C
I
mentioned
for
apportionment,
so
this
would
be
something
like
a
record
where
an
enumerator
has
gone
to
the
house.
The
they've
gone
for
a
proxy
next
door
and
the
person
said
yes,
someone
lives
there,
that
house
is
occupied,
but
we
don't
have
the
actual
account
there,
so
that
will
go
through
the
count
imputation
process
now,
once
that
census
unedited
file
is
complete,
the
next
stage
is
to
construct
the
census
edited
file
or
the
ceph
as
part
of
creating
that
file.
C
We
use
the
edits
and
characteristic
imputation
to
ensure
that
every
record
has
a
valid
response
for
the
person
and
housing
responses
in
the
census.
So
it's
important
to
understand
the
purpose
of
edits
and
characteristic
imputation.
The
purpose
is
to
ensure
that
every
respondent
has
a
valid
response.
C
Edits
in
characteristic
imputation
do
not
add
additional
people
to
the
census.
As
I
mentioned
earlier,
the
total
population
counts
from
the
sensors
are
finalized
in
the
previous
step.
In
the
cuff,
these
edits
are
just
ensuring
consistency
among
characteristics
such
as
a
person
being
identified
as
a
parent,
nothing
younger
than
a
young
child.
C
The
characteristic
computation
ensures
that
each
person
in
a
housing
item
has
a
valid
response,
such
as
an
occupied
housing
unit
that
doesn't
have
tenure
status,
which
is
an
indication
of
whether
it's
rented
or
owned.
We
make
sure
that's
there
and
we
use
administrative
records
for
the
2020
census
or
being
used
to
to
help
improve
the
quality
of
this
characteristic
imputation
sort
of
the
the
flying
way.
C
This
has
been
explained
to
me
is
the
the
the
the
staff,
the
cuff,
can
be
a
little
like
swiss
cheese
in
a
sense
that
you
know
exactly
how
many
people
are
there,
but
you
may
not
know
everything
about
them.
They
may
have
misskipped
this
a
question
or
a
question
couldn't
be
coded,
and
the
creation
of
the
staff
takes
care
of
that
and
makes
sure
that
all
of
those
attributes
exist
and
that
they
are
consistent.
C
So
there's
three:
there
are
different
types
of
characteristic
imputation
assignments
when
a
response
is
missing
or
inconsistent
with
other
data
that
was
supplied
for
that
person
or
household,
but
the
supplied
information
that
is
available
can
be
used
for
that
missing
or
inconsistent
response,
and
the
simplest
of
these
is
when
either
the
birth
date
or
the
age
is
missing.
If
you
have
one,
then
you
can
calculate
the
other
allocation
is
when
you
have
the
same
situation,
but
you
cannot
make
a
determination
for
the
missing
or
inconsistent
response
based
on
the
already
supplied
person
or
household
data.
C
C
C
The
first
scenario
I
mentioned
earlier,
one
is
provided.
The
other
is
missing.
So
if
age
is
reported
but
not
the
date
of
birth,
then
we'll
generate
a
random
birth
date
for
that
reported
age.
If
the
date
of
birth
is
reported,
then
we'll
calculate
the
age.
So
we
use
that
to
create
a
consistency
in
the
file.
C
If
both
of
them
are
reported,
but
they're
inconsistent
by
two
or
more
years,
then
we
use
edits
to
determine
which
is
more
consistent
with
other
data
that's
been
reported
and
we
edit
the
other
one,
the
other
characteristic
we
may
need
to
edit
an
age
for
consistency
like
child
parent
relationships.
From
the
previous
slide
that
I
mentioned,
if
necessary,
an
allocation
from
a
hot
deck
can
be
used.
C
The
characteristics
in
the
hot
deck
vary,
depending
on
the
nature
of
the
unanswered
questions,
and
then,
once
all
of
this
review
and
editing
and
imputation
are
completed
and
remember,
the
majority
of
records
need
to
have
none
of
this
done.
This
is
just
a
very
small
percentage
of
the
overall
census
returns,
but
then
we
have
that
completed
census
edited
file.
This
is
the
full
census
records
with
characteristics.
C
C
C
This
was
so
that
we
could
show
that
we
could
run
at
the
scale
of
the
decennial
census
and
then
to
demonstrate
what
progress
we'd
made
on
the
system.
Overall.
To
that
point,
we
then
followed
this
with
what
we
call
a
ppmf.
It's
a
privacy
protected
micro
data
file.
This
looks
like
individual
census
records,
but
it's
actually
privacy
protected
records
that
have
been
processed
through
the
protection
system.
C
It's
a
little
difficult
for
most
people,
myself
included
to
work
with
a
300
million
plus
record
file,
so
our
colleagues
over
at
the
iplums
national
historic
gis,
converted
that
ppmf
into
tables
that
data
users
would
recognize
and
in
smaller,
broken
out
by
state
for
some
of
the
block
level.
Data
different
manipulations
like
that.
Just
to
make
it
easier
for
folks
to
work
with,
we
actually
have
a
newer
version
that
came
out
today
or
was
posted
actually
last
night
and
then
the
newsletter
that
came
out
today.
C
That
shows
not
only
the
progress
that
we've
made
on
these
through
this
process
of
development
of
the
system,
but
also
it
shows
it
shows
the
focus
now
that
we
have
turned
to
to
focus
solely
on
the
pl
94
171
redistricting
data.
For
this
version
of
the
ppmf.
C
I
do
expect
to
see
at
least
one
more
ppmf
to
come
out
prior
to
the
data
stewardship
executive
policy
committee,
making
its
final
decision
on
where
to
set
epsilon
epsilon
being
the
value
which
controls
the
trade-off
between
accuracy
and
privacy
in
the
data.
And
so
we
should
have
that
out
sometime
before
the
end
of
this
year.
C
So
now
we're
going
to
turn
to
the
part,
that's
near
and
dear
to
my
heart,
which
is
this
some
specifics
around
the
redistricting
data
products.
So
I
just
wanted
to
put
as
a
reminder
that
the
redistributing
data
products
that
we've
been
talking
about
for
the
last
several
years
have
not
changed:
we're
planning
to
produce
the
geographic
products
of
shape
files,
the
pdf
maps,
the
block
assignment
files,
the
blocked
block
relationship
files.
C
You
should
be
familiar
with
these
from
our
previous
discussions.
These
are
identical
to
what
we
produced
for
the
2018
end-to-end
test
as
the
prototype
products
which
are
available
on
the
redistributing
offices
website.
So
please,
my
contact
information
will
be
at
the
end.
So
if
you
need
access
to
those,
you
can
just
contact
me
and-
and
I
can
make
sure
that
we
get
you
the
links,
but
these
these
products
haven't
changed
from
that
prototype,
so
you
can
still
expect
the
same
products.
C
C
So
this
is
the
files
that
are
the
the
tables
that
we're
planning
to
produce
for
the
2020
census.
This
again
is
what
we
put
into
the
2018
and
test
prototype.
C
It
was
only
for
providence,
rhode,
island,
because
that's
where
the
test
was
conducted,
but
the
the
schema
and
the
content
of
the
files
is
an
example
of
exactly
what
we
intend
to
produce
for
the
2020
census.
The
tables
the
p1
through
p4
and
table
h1
are
exact
replicas
of
what
we
produced
in
2010
and
then,
of
course,
we've
added.
C
This
group
quarters
table
by
group
quarters
type
which
we've
reported
at
the
total
population
level
for
those
states
who
have
a
need
to
reallocate
group
quarters
populations
to
an
alternate
address
prior
to
doing
their
redistricting
the
the
dates
on
this.
You
know
we
we
actually
I'll
talk
about
it
here.
We
did
put
out
that
prototype
data
that
I
mentioned
that
went
out
in
february
of
2019
for
the
geography
and
then
in
march
of
2019
for
the
tabulations
it's
still
available
online.
C
So
if
people
need
that
our
original
plan
for
delivering
the
official
geography
was
november
through
february
1st,
as
I
mentioned,
we're
looking
like
the
end
of
february
beginning
of
march,
to
get
that
to
folks
and
then
the
redistricting
data
itself
we're
currently
working
with
a
goal
of
re-establishing
a
schedule
that
will
get
us
to
that
data
out
to
registering
officials
no
later
than
april
1st
of
2021,
which
is
our
statutory
deadline.
C
C
It
would
be
a
single
national
delivery
to
all
the
recipients
so
most
likely
if
we
are
able
to
meet
the
statutory
deadline
when
we
get
this
plan
finalized.
Part
of
what
that
plan
would
include
is
that
you
can
expect
a
single
delivery
at
the
end
of
the
delivery
window
for
everyone.
So
that
would
be
the
end
of
march
of
2021..
C
C
But
we
started
this
one
off
the
american
community
survey
in
2011
and
released
it
every
year
for
10
years
with
our
instruction
to
create
a
2020
census,
dvap
special
tabulation
off
of
the
2020
census,
using
administrative
records
and
creating
it
down
to
the
block
level.
We
initially
thought
that
we
would
be
canceling.
The
american
community
survey
version
of
this
tabulation
just
for
this
coming
year,
through
the
request
of
several
states,
with
some
very
clear
guidance
on
how
it
would
still
be
useful
to
receive
that
american
community
survey
version
of
the
cvap
data.
C
We
have
reinstated
that
and
we
still
expect
to
be
able
to
produce
that
in
february
or
by
the
first
week
of
february
of
2021.,
so
that
is
still
on
the
table,
although
I
may,
in
previous
gatherings,
have
said
that
it
wouldn't
be
produced.
We've
now
switched
course
on
that,
so
that'll
be
something
that
we
will
produce.
We
also
still
plan
to
produce
the
2020
census,
evaps
special
tabulation,
using
administrative
records
in
the
2020
census.
C
C
The
content
on
the
it
shows
you
all
the
different
categories
that
are
included
in
the
cvap
special
tab
in
order
to
make
the
2020
census
evaps
special
tabulation
from
administrative
records
correspond
better
with
the
pl
data
set,
we've
added
in
some
other
race
alone,
category,
which
is
not
typically
a
category
in
the
acs,
but
it
is
a
category
in
the
descending
census
and
then,
of
course,
the
reason
for
the
2020
census.
Evap
special
tap
is
to
be
able
to
get
this
data
at
the
block
level,
so
block
level.
C
So
how
this
is
going
to
be
created,
it
was
just
sort
of
a
ironic
or
unfortunate
twist
and
at
the
exact
same
time
as
we're
here
talking
today
at
our
census,
scientific
advisory
committee
meeting
one
of
the
leads
from
the
internal
expert
panel.
That's
developing
the
methodology
was
presenting
their
first
results
of
their
their
research
to
the
advisory
committee.
C
The
the
links
are
here
in
the
in
the
powerpoint
and
I'll
make
sure
that
this
powerpoint
goes
to
wendy
and
christy
so
that
they
can
distribute
it
to
the
attendees
there's,
essentially,
four
different
techniques
that
are
being
considered.
Three
of
them
rely
on
a
combination
of
business
rules
plus
modeling,
and
then
one
technique
relies
on
a
latent
class
model
for
the
full
population.
C
Essentially,
when
we
say
business
rules
and
modeling,
it's
using
the
administrative
records
and
when
there's
a
very
clear
indication
of
someone's
status
as
either
a
citizen
or
non-citizen,
then
using
that
information
to
I
tag
that
person
as
the
citizen
non-citizen
and
then
the
remaining.
What
turns
out
to
be
just
between
nine
and
ten
percent
of
the
population
that
we
can't
make
those
determinations
for
that
smaller
part
is
modeled.
It's
modeled
in
different
ways.
One
is
to
a
hot
deck.
C
So
if
you
want
to
get
a
hold
of
me,
the
easiest
way
to
do
so
is
probably
rdo
at
census.gov,
but
you
can
also
call
the
763-4039
and
leave
us
a
message
and
we'll
be
able
to
get
back
to
you
that
way
as
well,
and
thanks
again
to
ncs
for
giving
us
this
chance
to
talk
to
you.
A
James
is
always
wonderful
to
have
you
with
us
and
you
are
very
responsive,
so
anyone
who
does
send
you
an
email,
it's
very
kind
of
you
to
put
your
email
up
there.
A
I
know
you'll
be
responsive
and
thanks
for
sharing
the
slides,
I
think
we'll
distribute
them
to
the
people
who
are
on
this
call
and
then
christy
lets
us
also
link
it
on
to
the
page
that
we
have
for
this
call
as
well,
and
then
you
mentioned
our
redistricting
distribution
list
and
let
me
just
take
a
moment
to
say
that
for
people
who
are
interested
in
more
information
from
wherever
it
might
be,
from
the
bureau
and
from
any
redistricters
from
the
news
media
about
major
case,
we
do
have
a
distribution
list.
A
A
C
That
is
a
question
that
I
don't
think
I'm
prepared
to
answer
not
not
trying
to
be
to
avoid
the
question.
I
just
don't
have
a
good,
clear
answer
or
understanding
of
how
that
happened.
Terry
may
even
know-
and
she
can
probably
address
that
too,
but
I
hopefully
we
captured
the
response
back
in
april
first,
but
if
their
house
hasn't
responded,
I'm
not
sure
how
it
gets
picked
up.
A
I've
been
thinking
about
this
with
wildfires
living
out
west.
Okay,
with
that,
let's
say
thank
you
to
james
and
terry
anne.
Let's
bring
you
up.
D
Well,
thank
you
so
much.
Thank
you.
Wendy
christie,
james.
It's
always
a
privilege
to
join
you
and
all
of
your
members
for
a
conversation
about
the
census
and
as
if
we
didn't
all
have
enough
drama
in
our
lives
in
2020.
You
know
the
census
has
kept
us
on
our
toes
with
seemingly
unending
shifts
in
timelines
operations
and
policies,
so
never
a
dull
moment
in
census
land,
as
I
like
to
say
james.
D
Thank
you
for
the
great
context
and
important
information
about
the
status
of
census
operations
and
a
bit
of
a
crystal
ball
into
the
timing
and
quality
of
the
results
which
will
dominate
much
of
our
legislators
time
for
the
next
year
or
more,
but
also
inform
their
decision
making
on
resource
allocations
for
the
communities
this
audience
represents
for
the
next
decade.
D
So
I'm
gonna
try
to
put
the
update
that
james
shared
in
the
broader
political
context,
by
offering
my
perspective
on
the
census
bureau's
ability
to
produce
what
I'm
going
to
call
an
acceptably,
accurate
census
and
then
I'll
give
you
some
additional
information
on
congressional
efforts
to
lend
some
certainty
to
remaining
census
operations
and
the
data
release
schedules.
D
So
let
me
just
start
with
the
fundamental
question
of
whether
the
census
bureau
can
produce
acceptably
accurate
census
numbers
under
what
I'm
going
to
call
the
rushed.
That's
air
quotes
folks
rush
census
plan,
which
the
census
bureau
announced
on
august
3rd
now,
as
james
described,
the
census
bureau
is
publishing
two
sets
of
progress
indicators
every
day,
total
response
rates
by
state
and
then
the
proportion
of
the
door
knocking
operation
workload.
D
D
What
do
those
indicators
mean?
Is
the
census
bureau
on
track
to
finish
again?
Air
quotes
quote
unquote,
counting
all
households
by
the
end
of
september,
if
necessary
or
by
the
end
of
october.
If
the
courts
continue
to
prohibit
the
rush,
the
accelerated
schedule,
james,
I
think
you
cited
a
national
total
response
rate
of
close
to
94.
D
Today,
so
don't
forget
the
total
response
rate
combined
self-response
and
households
or
homes
enumerated
during
the
door
knocking
phase,
but
the
proportion
of
households
that
haven't
been
counted
varies
greatly
state
to
state
and
even
within
most
states.
D
So
as
of
yesterday,
for
example,
the
states
where
completion
of
the
door
knocking
and
self-response
operations
where
the
were
lagging,
those
are
mostly
along
the
gulf
coast
and
in
the
southeast
u.s
up
to
the
mid-atlantic,
can
anyone
say:
hurricane,
laura
hurricane
sally
and
the
rest
of
the
hurricane
alphabet,
that's
lurking
in
the
atlantic,
somewhere,
the
census
bureau
still
hasn't
completed,
the
non-response
follow
operation
hasn't
completed,
90
percent
or
even
80
or
70
of
it
in
large
swaths
of
texas,
arizona,
new
mexico,
colorado,
montana
and
calif
california,
much
of
iowa
and
kentucky
tennessee
and
illinois.
D
The
completion
rates
again
for
this
non-response
follow-up
operation
are
still
below
80.
But
I
want
to
lift
up
a
more
important
question.
Even
if
the
census
bureau
reaches
a
99
completion
rate,
which
is
james
said,
is
the
census
bureau's
goal.
D
D
You
know,
even
in
a
less
challenging
census,
environment
data
collected
during
of
the
door.
Knocking
operation
are
less
accurate
than
answers
that
you
get
on
a
self-response.
So
in
areas
where
you
had
lower
self-response
historically
under
counted
communities,
a
lot
of
rural
areas,
american
indian
reservations
and
so
forth,
there
are
more
households.
The
bureau
has
to
count
in
the
non-response
follow-up
operation,
where
we
know
accuracy
goes
down.
D
D
A
completed
census
case
doesn't
always
represent
a
household
that
was
counted
through
a
direct
interview.
The
case
could
have
been
I'm
using
again
in
air
quotes
resolved,
but
that's
the
word
the
census
bureau
uses
it
could
have
been
resolved
as
vacant
as
non-existent.
So
it
didn't
meet
the
definition
of
a
housing
unit.
An
enumerator
could
have
talked
with
a
neighbor
or
a
landlord
to
get
the
information
at
the
proxy
enumeration
in
2010.
D
22
of
all
of
the
field.
Cases
were
enumerated
by
proxy
and
a
quarter
of
all
of
those
didn't
yield
enough
information
to
be
useful,
so
they
required
the
statistical
imputation
methods
that
james
described
earlier,
and
we
aren't
only
concerned
about
enumerating
all
households
right.
We
are
interested
in
accurate
data
about
the
characteristics
of
those
households,
because
that
information
guides
the
policies
that
are
aimed
at
children
and
senior
citizens
right
and
ensuring
equality
of
access
by
race
and
ethnicity
and
gender
in
health
care,
education,
voting
rights
and
so
forth.
D
All
in
your
wheelhouses
and
finally,
please
remember
the
and
and
james
talked
about
these
other
operations,
the
enumeration
of
housing
units.
So
that's
what
they're,
talking
about
with
the
completion
rates
doesn't
cover
everyone
living
in
the
united
states
right.
The
non-household
counting
operations,
which
are
difficult
under
the
best
of
circumstances
which
we
don't
have
this
year,
have
been
delayed
considerably
and
they'll
be
more
difficult.
Judakova
19
concerns
and
population
displacement.
D
These
operations,
I
think,
are
now
more
likely
to
miss
people
or
to
count
people
twice
or
in
the
wrong
place,
or
fail
to
collect
accurate
information
on
characteristics
such
as
race
and
age.
So
I
think
the
bottom
line
is
in
the
wake
of
the
broad
pandemic,
related
disruption
and
delays
in
all
of
these
operations.
D
It's
going
to
be
more
difficult
to
enumerate
households
accurately
so
rushing
the
data
collection
will
certainly
leave
us
with
less
accurate
raw
data,
but
that's
only
half
the
threat
of
this
shortened
census
timeline.
D
The
data,
processing
and
quality
assurance
activities
normally
take
five
months
in
a
normal
census
year
and
that
will
be
compressed
down
to
three
months
at
best
if
the
current
rush
deadline
holds
according
to
the
government
accountability
office.
At
a
congressional
hearing
last
week,
the
census
bureau
hadn't
even
finished
testing.
All
of
the
data
processing
systems
that
increases
risk
and
less
time
for
data
processing
means.
The
bureau
may
not
have
enough
time
to
do
all
of
the
on
duplication,
which
I
think
is
going
to
be
higher
in
this
census.
D
There
were
eight
million
duplicates
in
2010.
It's
going
to
be
much
higher
this
time.
Think
about
the
college
students
who
should
have
been
counted
at
their
college,
residence
on
or
off
campus.
They
were
sent
home
just
as
the
census
was
starting
mom
and
dad
might
have
counted
them
by
mistake.
The
people
with
second
homes
who
fled
cities
or
states
with
early
covet
surges
and
so
forth.
D
So
I
think
the
chances
of
counting
people
in
the
right
of
not
counting
people
in
the
right
place
or
counting
them
twice
have
gone
way
up
and
if
the
census
bureau
doesn't
have
the
time
it
needs
to
untangle
this
mess
of
raw
data-
and
you
know
james
talked
about
a
lot
of
the
processes
and
to
try
and
fill
in
the
gaps
with
administrative
records,
other
statistical
methods
and
so
forth.
D
So
I
just
want
to
say
in
closing:
I
don't
want
to
end
on
a
sober
note.
Wendy.
If
I
can
have
one
more
minute.
I
think
there
is
a
way
out
of
this
ditch.
I
I
said
wendy
said
I
spent
much
of
my
professional
career
in
congress.
D
Congress
has
constitutional
responsibility
for
the
census,
so
I
think
congress
must
write
this
badly
listing
ship
and
not
it's
not
the
census.
Bureau's
fault
that
the
census
got
hit
by
a
pandemic.
The
census
bureau's
expert
professional
staff
said
that
counting
operations
right
should
continue
through
october
31st.
D
The
commerce
secretary
made
that
request
to
come
to
congress
and
the
house
subsequently
in
may
passed,
extended
deadlines,
pushing
back
the
apportionment
and
redistricting
reporting
deadlines
by
four
months.
Each
and
the
president
publicly
supported
the
request
in
april,
so
the
administration
reported
supported
the
extended
census
timelines
until
it
didn't
without
explanation
or
warning.
So
I
think
congress
has
to
step
in
to
extend
the
statutory
reporting
deadlines,
as
of
today,
as
christie
said,
that
temporary
restraining
order
of
preventing
the
census
bureau
from
winding
up
field
operations
on
september
30th
is
still
in
place.
D
There
is
broad
and
diverse
support,
I
think,
for
a
good
census,
because
we
all
have
to
live
with
those
results
for
the
next
10
years.
So
I
hope
we
can
count
on
your
support
on
in
these
closing
days
and
hours.
These
bills
are
not
going
to
move
on
their
own,
but
stakeholders
are
working
to
get
those
provisions
in
a
must
pass
bill.
D
You
know
what
those
are
before
congress
recesses
for
the
elections
and
a
must-pass
bill
would
be
either
a
covered
relief
bill
or
the
ubiquitous
continuing
resolution,
in
other
words,
the
temporary
short-term
funding
measure
for
the
beginning
of
fiscal
year,
2021
to
keep
the
government
from
shutting
down.
So
I
do
want
to
echo
james's
a
call
to
action
at
the
end,
it's
not
too
late
for
you
to
urge
your
constituents
to
respond
to
the
census,
whether
on
their
own
self-response
or
by
speaking
with
an
enumerator
at
the
door
and
with
that
wendy
christie.
A
Great
well,
thank
you
very
much,
terry
ann.
That
was
a
really
nice
wrap-up
of
the
things
that
we'd
asked
you
to
address.
There
was
one
question
in
the
chat
box
about
what
is
ncsl's
position
on
the
congressional
bills
and
and
jeff
responded
and
he's
absolutely
right.
Ncsl
is
on
record
with
a
general
generic
kind
of
policy,
supporting
a
full
and
accurate
count.
We
also
this
year
sent
forward
a
letter
to
both
chambers
and
also
to
the
census
bureau,
expressing
concerns
about
the
delays
and
concerns
about
differential
privacy.
A
Now
we
have
what
I'm
referring
to
as
the
undelay,
but
that
puts
us
in
a
different
situation
than
we
were
before
so
I'll
say
to
everybody
who's
on
the
line.
If
you
are
a
legislator
or
legislative
staff
person,
I
would
be
happy
to
hear
from
you
your
thoughts
about
what
ncso
could
do.
If
anything.
At
this
point,
it
does
strike
me
that
the
time
on
this
is
coming
right
to
an
end.
We
are
coming
to
a
stop
sign.
A
I
don't
know
many
of
you
know
some
of
you,
don't
that
my
husband
is
an
enumerator
and
he's
out
this
very
moment,
knocking
on
doors
and
he's
had
all
kinds
of
interesting
experiences
with
this
mostly
positive,
but
he
hasn't
had
a
single
word
from
his
boss.
Yet
on
whether
they're
extending
or
not
so
he
hasn't
gotten
any
word
beyond
the
september.
30Th
deadline.
A
Okay,
so
we
do
have
some
questions.
Jeff,
I'm
going
to
ask
you
to
come
on
and
let's
start
with
your
question
about
the
apportionment
state
population
account.
I
know
you
have
several,
but
let's
start
with
that,
one
jeff
weiss.
E
Okay,
good
afternoon
everybody
I'm
jeff
weiss
from
new
york.
I've
been
serving
as
the
staff
co-chair
of
ncsl's
redistricting
committee.
My
question
to
james
is
whether
there
will
be
any
impact
on
the
pl
94
171
file.
If
any
changes
are
made
to
the
congressional
apportionment
state
population
count
totals.
C
And
so
the
the
one
of
the
key
pieces
for
the
apportionment
population
is
something
that's
called
the
resident
population
and
that
resident
state
population
is
unaltered
by
differential
privacy.
It's
the
population
as
enumerated,
and
that
is
something
that,
when
we
move
to
the
creation
of
the
redistricting
data,
we're
going
to
work
to
ensure
that
the
state
population
totals
in
the
redistricting
data
are
directly
equal
to
what's
been
reported.
As
the
official
resident
population
for
the
states
in
the
abortion
account.
E
E
Do
we
have
any
idea
what
the
the
inability
of
the
census
bureau,
not
because
of
its
own
fault,
simply
weather
and
other
kinds
of
circumstances
that
doesn't
get
to
the
99
goal
might
mean
for
demographic
groups
within
states
in
the
pl
data.
In
other
words,
if
you
will
you
be
using
imputation
and
proxies
to
make
up
the
difference,
because
in
a
lot
of
these
states
with
especially
minority
populations,
an
incomplete
cat
will
have
an
impact
on
redistricting.
D
So
james
I
mean,
if
I
could
jump
in
quickly
here
so
because
I
really
want
to
emphasize
something.
First
of
all
the
completion
rate,
even
if
it's
99
or
100
already
include
proxies
and
again
in
2010
22
of
all
cases.
In
that
non-response
follow-up
workload
that
door
knocking
operation
were
completed
by
proxy
interviews.
I
think
it's
going
to
be
higher
this
time.
D
I'm
not
don't
have
any
data
yet
because
they
enumerates
were
not
allowed
to
start
talking
to
a
proxy
to
a
neighbor
and
so
forth
in
2010
until
they
had
six
unsuccessful
attempts
at
the
door.
Now
they're
allowed
to
go
to
proxy
after
three
unsuccessful
attempts.
So
that's
number
one
but
again
just
because
they
get
to
99
doesn't
mean
those
data
are
accurate
and
we
already
know
that
cases
that
are
completed
right.
Households
that
are
counted
during
the
non-response
follow-up
operation
are
twice
as
likely
to
miss
people
within
a
household.
D
That's
already
counted
than
a
case.
That's
been
done
through
a
self-response
right
and
a
significant
part
of
the
census
under
count,
which
already
is
disproportionately
people
of
color
young
children
renters,
which
is
a
proxy
for
low
income.
A
significant
portion
of
that
under
count
are
people
who
are
missed
in
households
that
were
counted
so
to
me,
just
focusing
on
the
99
doesn't
tell
us
everything.
We
really
need
to
know
about
how
accurate
the
data
are,
that
james
is
going
to
hand
you
in
the
redistricting
data
file,
but
I'll
get
off.
A
Okay,
I'll
say
that
we
had
another
comment
earlier
on.
That
was
in
regards
to
what
you
said
at
the
very
beginning.
James
about
the
data
would
be
coming
in
one
big
unit
not
rolled
out
over
the
course
of
six
weeks
and
new
jersey
points
out
that
it
has
a
february
15th
deadline
to
get
redistricting
done
in
this
coming
year.
As
you
probably
know,
they
have
a
constitutional
amendment
on
their
ballot
in
november.
So
I
guess
there's
a
lot
on
that
front.
C
I
did
want
to
make
a
comment
on
that,
because
the
even
before
anyone
had
heard
the
word
of
kovid
when
we
had
our
original
2020
census
schedule,
the
earliest
that
any
state
would
have
received
data
was
february
18th.
C
A
Got
it
got
it
okay
and
tyrion?
I
want
you
to
address
this.
Please
someone
asks.
Are
you
finding
congressional
support
depends
on
where
the
political
power
may
lay
given
current
versus
potential
count
and
kathy.
If,
if
we're
not
understanding
that
question-
and
you
want
to
pop
on
and
ask
us,
that's
fine
too,.
D
I
let
let
me
say
this
because,
and
let
me
also
say
I
mean
when
I
worked
in
congress
many
moons
ago.
You
know
things
were
more
cooperative.
There
was
a
lot
more
bipartisanship.
D
Yes,
I
was
staff
director
for
a
democratic
chairperson,
but
I
started
my
congressional
career
in
a
senate
republican
office.
So
I
you
know,
I
just
get
very
frustrated
by
a
lot
of
the
partisanship
right
now
and
I
think,
regrettably,
I
think
the
administration
very
unwisely
abandoned
its
support
for
letting
the
census
bureau
follow
the
adjusted
schedule
and
have
enough
time
to
do
all
the
counting
and
the
data
processing
and
the
quality
assurance
that
it
said.
D
You
know
it
needed
to
do
for
a
reason
that
just
was
sort
of
a
red
herring
right
and
that
has
to
do
with
the
you
know
whether
the
apportionment
counts
will
be
delivered
to
the
president.
While
this
president
conceivably
is
still
president
or
not
meaning
by
the
end
of
this
year,
as
opposed
to
next
spring.
But
what
we're
trying
to
remind
folks
in
congress
and
they're
finally
starting
to
get
this
is
normally,
and
this
is
the
first
time
with
this
presidential
memorandum
to
exclude
undocumented
residents
from
the
apportionment
counts.
D
The
president
doesn't
do
anything
to
the
apportionment
numbers
he
hasn't
he.
Hopefully
she
one
day
has
a
ministerial
role.
He
gets
the
numbers
from
the
commerce
secretary
and
the
resulting
apportionment,
and
he
turns
around
and
transmits
those
to
the
clerk
of
the
house
of
representative
congress
at
the
start
of
a
new
congress.
C
D
Presidency
or
a
biden
presidency
there
wasn't
going
to
be
anything.
A
president
biden
would
do
to
change
the
numbers
and
I
think,
a
lot
of
republican
lawmakers
just
didn't
weren't,
focusing
on
that
and
are
starting
to
come
around
to
the
understanding
that
every
single
state
is
really
at
risk
of
significant
undercounting
and
poor
quality
data.
If
the
census
bureau
is
forced
to
rush
the
remaining
operations
and
equally
important
the
crazy
complex
data
processing
on
the
you
know,
the
accelerated
schedule.
A
So
I'll
say
that
it
is
12
58
we
do
have
the
ability
to
stay
on.
So
any
of
you
who'd
like
to
stay
on.
Please
do
and
I
did
take
the
easier
questions.
First,
you
might
say,
or
the
more
general
ones,
so
I'm
just
going
to
keep
rolling.
I
hope
that
that's
okay
with
james
you
as
well.
Can
you
stay
on
longer?
A
Okay,
so
we
had
a
couple
of
questions
further
about
the
disclosure
avoidance
system,
and
maybe
you
can
just
kind
of
pick
up
where
you
left
off
or
or
maybe
back
it
up
a
little
bit.
One
is
what's
the
current
status
of
the
differential
privacy
program.
Is
it
still
in
play
under
the
current
type
time
frame?
I
guess
that
means.
A
How
can
you
do
this
new
process,
even
though
now
that
you've
got
fewer
months
and
while
you're
at
it
dan
asks
have
the
latest
results
of
the
tweet
disclosure
avoidance
system
processing
been
evaluated
outside
the
census
bureau,
and
I
think
you
answered
that
in
the
chat
that
the
latest
is
really
today
yeah.
But
what
do
you
know
so
far?
So
if
you
could
just
give
us
a
little
more
about
that.
C
Yeah
I
mean
the
the
census
bureau
is
legally
obligated
to
protect
the
data
that
we
publish
to
respect,
protect
respondent
data,
so
not
performing
some
form
of
disclosure
avoidance
on
the
data
that
we
produce
just
would
not
be
acceptable,
so
the
disclosure
avoidance
is
still
part
of
the
process.
C
C
I
think
when
people
see
the
metrics
that
come
out
from
this
latest
version,
which
we'll
be
following
in
about
another
week,
we've
revamped
the
metrics.
Some
of
the
metrics
are
now
very
focused
on
the
same
thing
characteristics
that
people
who
work
with
redistricting
data
would
be
concerned
about,
and
with
this
version
of
the
ppmf
from
the
disclosure
avoidance
system
being
tailored
directly
to
the
redistricting
data,
I
think
people
will
be
very
pleasantly
surprised
to
see
the
level
of
improvement
that
we've
been
able
to
have
and
we're
continuing
to
work
to
improve
the
data.
C
So
so,
yes,
we.
We
fully
admit
that
we
know
that
the
data
we
released
in
october
was
not
acceptable
for
for
use,
but
it
was
showing
that
we
could
do
the
do
this
on
on
something
at
the
scale
of
the
decennial
census.
So
we
and
then
we've
been
working
to
refine
it
to
make
that
data
more
fit
for
use
with
each
iteration,
so
that
is
still
in
the
in
the
pipeline.
C
We
do
believe
we'll
be
able
to
do
so
and
do
it
well,
and
the
review
of
the
external
data
has
not
happened
for
this
latest
one
because,
as
you
mentioned
wendy,
this
was
just
published
last
night.
I
think
kim
typed
in
there
that
he's
going
to
send
out
a
link
to
the
group
once
the
data
gets
processed
by
the
icons
and
hgis
folks,
who
will
create
nice
tables
and
break
things
down
so.
E
A
The
redistricting
data
use
above
many
other
data
uses,
which
probably
works
for
the
people
who
are
on
this
phone
call,
might
not
work
for
the
people
who
are
distributing
money
elsewhere
or
people
in
civil
rights
groups,
perhaps,
but
for
this
group
that
was
much
appreciated
and
we
do
have
statutory
standings
in
that
regard.
So
so
I
think
that
that's
part
of
the
reason
kim
I'm
going
to
ask
you
to
hop
on
here,
and
you
had
a
couple
of
questions.
Why
don't
you
ask
them
out
loud.
E
E
We
did
get
the
data
this
morning
and
it
is
being
processed
by
the
university
of
michigan's
site
and
they
told
me
just
a
little
while
ago
that
they
will
have
the
data
for
everybody's
use
tabulated
to
the
block
level
on
monday,
and
so
when
I
get
the
link
on
where
that
is,
I
will
send
that
to
christy
so
that
she
can
get
that
information
out
to
everybody.
I
would
certainly
encouraging
all
the
states
to
take
a
look
at
their
data.
E
So
that's
certainly
something
for
next
week
for
people
to
play
around
with
the
other
question
I
had
for
james
is
last
night's
extension
of
the
tro
by
the
court
saying
that
the
you'll
have,
until
september
24th
for
the
restraining
order
for
stopping
things
that
puts
you
much
closer
to
alphonse
date
of
the
end
of
the
month.
When
you
guys
say
you'll
no
longer
be
able
to
produce
the
apportionment
counts,
nor
the
pl
counts
on
the
remap
schedule.
E
C
The
schedule
that
we're
working
to
develop-
that's
still
an
unofficial
schedule,
is
tied
to
our
ability
to
get
through
the
process
of
creating
the
abortion
counts.
A
And
that's
a
good
lead
into
a
question
from
kathy
duvall.
She
asks
you
james.
Will
the
resident
population
numbers
between
the
pl
file
and
apportionment
still
equal,
even
if
the
apportionment
totals
subtract,
the
undocumented
residents
based
on
the
executive
memorandum,
did
that
in
catholic
yeah?
So
I.
C
Think,
in
order
to
probably
clear
this
up,
I
think
I
need
to
describe
how
we
would
typically
do
apportion
accounts.
We
we
create
for
the
the
apportionment
counts.
We
actually
have
two
sets
of
numbers.
Typically,
we
have
the
the
resident
population,
which
is
all
persons
residing
within
the
state,
and
then
we
have
the
federally
affiliated
account
overseas,
which
is
a
secondary
bucket
that
the
the
folks
that
were
countered
overseas
are
usually
typically
in
the
military.
C
99
of
them
are
military
and
that
number
gets
added
into
the
apportionment
total
and
that's
how
the
number
that's
used
for
allocating
seats
that
resident
population
count,
which
is
all
persons
residing
in
the
state,
is
the
number
that
will
be
held
equal
for
the
state
totals
for
the
redistricting
data
set.
Should
there
be
other
other
subpopulations
that
are
calculated
so
that
people
can
move
them
in
or
out?
A
Thank
you.
I
I
think
the
answer
to
kathy's
question
is
yes,
but
that
was
a
much
better
answer
than
just
yes,
and
if
I'm
wrong
about
that
being
a
yes,
then
we
might
want
to
back
up
another
question.
I'm
a
bit
confused
about
the
cvap
data.
A
Am
I
correct
that
there
will
be
two
c-vap
counts,
one
using
the
traditional
special
tabulation
off
of
the
acs
data
and
one
using
the
administration's
direction
on
using
administrative
records,
and-
and
did
I
understand
that
correctly-
that
the
early
february
special
tab
of
acs
data
will
be
produced
at
the
census
block
level,
or
is
that
just
the
administrative
record
so
just
separate
those
two
out
one
more
time.
C
Okay,
yes,
there
will
be
two
citizen
voting
age
population
data
products
produced
the
first
one
will
be
the
version
that
we've
been
creating
for
the
last
10
years
off
of
the
american
community
survey.
That'll
go
down
to
the
block
group
level,
not
the
block
level
and.
A
C
So
it's
the
normal
acs
production
cycle,
the
2020
cvap
special
tabulation,
which
is
the
one
that
we're
using
off
of
administrative
records,
will
be
produced
down
to
the
block
group
or
the
block
level,
the
the
individual
tabulation
block
and
that
will
be
based
off
of
2020
census.
Statistical
geography,
20
20
blocks,
20
20
block
groups
and
20
20
tracks.
A
Okay,
I've
got
it,
you
get
the
acs
based
ones
on
2010
geography,
you
get
the
administrative
record
ones
based
on
2020
geography.
Okay,
we've
only
got
one
or
two
more
questions.
Let
me
ask
you
this.
I
think
james
said
that
racial
and
ethnic
data
would
not
be
distributed.
For
group
quarters.
Is
that
data
being
used
under
disclosure
avoidance?
A
I
have
heard
that
it
has
an
effect
on
population
surrounding
prisons,
and
let
me
just
add
to
that
that
we've
heard
recently
that
households
in
the
same
census
block
with
prisons,
then
the
data
when
both
before
and
after
death,
it
looks
inaccurate
for
both
sets
of
people.
So
how
are
prison
households
connected
and
how
does
differential
privacy
relate
to
that.
C
The
group
quarters
the
saying
that
the
there
would
be
no
racial
or
ethnic
data
associated
with
the
group
quarters.
That's
in
the
pl
file.
So
only
in
the
registering
data
file
there
will
be
subsequent
data
products
that
the
census
bureau
has
planned
the
demographic
housing
and
characteristic
file,
which
is
the
replacement
for
the
summary
file,
one
if
you're
familiar
with
with
last
decades
and
previous
decades,
work
that
will
likely
have
group
quarters
tables
which
will
include
race
and
ethnicity,
probably
at
higher
levels
of
geography.
C
It
won't
be
at
the
block
level,
but
at
higher
levels
of
geography
they
would
likely
have
some
tables
like
that.
There
is
the
possibility
when
the
das
system
is
run,
that
people
who
are
in
housing
units
may
get
moved
into
group
quarters
or
people
who
are
in
group
quarters
make
it
moved
into
housing
units
so
that
that
is
something
that
has
been
seen
as
an
effect
of
applying
the
differential
privacy.
D
Wendy,
could
I
just
stand
quickly
to
this
whole
question
about
group
quarters
and
characteristics
data,
and
then
group
quarters
is,
you
know,
covers
one
part
of
college
university
student
populations,
but
not
the
ones
we're
living
in
private
housing
off
campus.
So
I
just
want
to
again
again
I
don't
mean
to
always
sound
like
a
naysayer,
but
this
gets
back
to
the
sausage
making
of
census.
Taking
we
are
dealing
with.
D
You
know
the
potential
for
a
lot
of
inaccuracy
in
the
data
and
especially
because
of
the
pandemic,
but
also,
I
think,
because
the
bureau
may
not
have
enough
time
to
clean
up
a
lot
of
this,
but
colleges
and
universities
have
largely
we
understand,
have
been
transferring
data
on
their
group
quarters,
so
kids,
living
in
dorms,
right
in
residence,
halls
and
brats
and
so
forth,
with
the
bare
minimum,
sometimes
just
first
name
and
address
of
the
dorm,
no
race,
no
ethnicity,
no
age,
and
then,
if
you
don't
have
a
last
name
or
a
birth
date,
it's
impossible
to
unduplicate
if
mom
and
dad
incorrectly
also
counted
that
student
at
home.
D
But
furthermore,
the
bigger
problem
is
that
there
are
really
low
self
response
rates
in
census,
tracts
where
there
were
a
lot
of
college
students
living
off
campus
right,
so
it
could
be
a
neighborhood
in
a
big
city
like
around
columbia,
university,
new
york,
I'm
in
stanford,
connecticut
fyi
shout
out.
I
saw
somebody
from
connecticut
here,
but
also
you
know,
the
smaller
college
stands
like
ithaca
new
york,
where
I
went
to
school
at
cornell.
You
know
their
self-response
rates
are
in
the
dump
and
they
can't
get
these
students.
D
You
know
to
respond
and
a
lot
of
them
have
not
come
back
to
the
area.
I
think
the
college
is
totally
remote
right
now
and
the
sense
is
the
the
census.
Bureau
is
pleading
with
colleges
to
supply
administrative
records
also
for
students
who
live
off
campus,
but
those
data
are
not
going
to
include
race
and
ethnicity.
D
A
I
remember
when
I
first
started
learning
about
the
census,
probably
from
unitarian,
and
I
thought.
Oh,
my
goodness.
I
always
thought
those
were
perfect
numbers
and
now
there's
never
been
a
perfect
census.
There's
never
been
to
count
every
single
person,
and
maybe
the
issues
relating
have
changed
from
time
to
time.
Randy.
D
A
Yes,
absolutely
and
one
last
question:
unless
something
else:
pops
up
and
that's
from
mike
fortner-
and
it's
probably
yours
james
if
the
enforcement
data
is
delayed,
perhaps
by
that
act
of
congress
that
terry
anne's
looking
for,
will
the
redistricting
data
go
back
to
a
staggered
rollout?
Do
you
know
the
answer
to
that
yet,
or
is
that
too
many
ifs
between
now
and
then.
C
I
think
that's
a
too
many
f's
strung
together
to
be
able
to
make
a
definitive
answer.
You
know
there's
we
would
just
have
to
see
how
things
play
out
and
we'll
we'll
work
to
do
what
was
best
for
the
quality
of
the
data
and
for
trying
to
meet
the
needs
of
the
stakeholders
and
balancing
those
two
so
but
yeah.
So
it's
a
few
too
many
ifs.
I
can't
make
a
a
guess
at
this
point.
A
All
right:
well,
I
want
you
to
know
I'm
going
to
stay
on
this
call
until
the
last
person
is
off,
so
anybody
wants
to
keep
talking
about
the
census,
I'm
here
and
james
and
tyrion.
If
you
want
to
stick
with
us,
that
would
be
even
better
because
you'll
have
way
more
good
stuff
to
say,
but
I
want
to
thank
you
both
very
much.
A
This
has
been
a
heck
of
a
week
and
the
fact
that
you're
here
with
us
on
friday
afternoon
is
just
really
warms
my
heart
and
I
appreciate
everything
you
do
and
I
love
the
fact
that
we're
all
in
communication
for
all
of
you
who
are
on
the
line.
Thank
you
very
much.
As
I
said
earlier,
ncsl
is
welcome,
welcoming
to
your
input.
I
don't
know
that
we
can
do
much
besides,
be
this
conduit.
A
The
role
we
currently
play
primarily,
but
do
let
us
know
what
you're
thinking
and
do
go
back
to
your
states
with
the
idea
that
place
to
be
paying
attention
is:
where
are
the
colleges
there
could
be
some
data
changes?
Where
might
be
some
data
changes
so
take
back
some
of
this
to
your
data.
Folks,
your
state
data
center
and
such
and
with
that
I'll
just
say,
stay
safe,
stay,
calm
and
stay
engaged
on
the
census.