some folks may be having trouble downloading the patch. it can be reaced from
http://www.mcs.com/~berwyned/census/workup/
if you are having difficulties with it drop me your phone number or call and
we can figure something out. (202-366-0412)
"Charles H. Kooshian" wrote:
> url doesn't work
>
The attached message is intended for agencies participating in the
Workplace Update Program. Others should ignore it. The attachment is
in Rich Text Format. If you cannot read it, please call Gloria or Clara
at the Journey to Work and Migration Branch (301-457-2454). Thank You.
--Phil Salopek
From: Census2000(a)ccmc.org
Final Mail Response of 65% Equals 1990 Rate:
Plus: Stakeholders Voice Support for Long Form Data;
Census Advisory Committees Meet; and more.
Sixty-five percent of households filled out their census forms during
the first phase of Census 2000, the Census Bureau reported at a news
conference on April 19. The mail response rate, which includes
households that returned a form by mail, or provided their answers over
the Internet or by telephone, equals the 1990 rate and exceeds the
bureau's projected response of 61 percent. Beginning on April 27, about
500,000 census enumerators will visit the more than 43 million
households that did not yet answer the census. The 'non-response
follow-up' phase is scheduled to run through early July.
Census Bureau Director Kenneth Prewitt called the response rate "a
serious achievement; it is news to celebrate." Until 2000, mail
response rates had declined every decade since the 1970 census, the
first to rely primarily on mailed questionnaires. The rate was 78
percent in 1970, 75 percent in 1980, and 65 percent in 1990. "The
country is saying that democracy is about obligation as well as
benefits, about responsibility as well as rights," the director
observed. He gave credit to more than 100,000 local government,
community, civic, religious, and business "partners" who worked to
promote Census 2000. The Census Bureau challenged local governments to
exceed their 1990 mail response rates by five percent. About 15 percent
met their 'Plus 5' targets, although no states did.
Dr. Prewitt also said the difference in the mail response rate for
households that received a short form and households that received a
long form was 12 percent. He declined to speculate on whether public
criticism of the long form questions contributed to the lower return
rate for those forms. 66.6 percent of short form households have
answered the census, compared to 54.1 percent of long form households.
The differential is about twice the gap at the same point in the 1990
census.
The mail response figure does not include 'Be Counted' forms that were
available in public places for people who did not receive or misplaced a
questionnaire at their home, or who believe they were left off the form
completed by someone else in their household. Census takers must visit
the address provided on a Be Counted form to verify its existence and
confirm the number of residents. The final mail-back rate also does not
reflect the count in remote areas, where the absence of accurate address
lists requires enumerators to record the location of housing units and
interview the residents in person. This 'list/enumerate' procedure is
scheduled to conclude at the end of April.
A spokesperson for Rep. Dan Miller, who heads the House Census
Subcommittee, said the chairman was pleased with the Census Bureau's
success in stemming the decline in the response rate. The congressman
also is concerned, his spokesperson said, that plans for the
door-to-door operation do not focus enough attention on historically
'hard-to-count' areas. Rep. Miller said the bureau should target any
savings it realizes from exceeding its projected mail response rate at
improving the count of population groups that have been missed at higher
rates in the past. Rep. Carolyn Maloney (D-NY), the Census
Subcommittee's senior Democrat, praised the Census Bureau for its
"outstanding achievement" and also noted that the mail-back rate for
Manhattan was five percent above the borough's 1990 rate. She urged
people who did not respond during the mail phase to cooperate with
enumerators during the "most critical and labor intensive phase" of the
census.
The mail response rate, which reflects the portion of all housing units
that return a questionnaire, differs from the mail return rate, which is
the percent of occupied housing units mailing back forms. However, it
is the response rate that affects both the cost estimates and the scope
of the fieldwork, since the Census Bureau does not know whether an
unresponsive housing unit is occupied or vacant until an enumerator
visits. The mail return rate, calculated after the census is finished,
is a truer measure of public cooperation. Like the overall response
rate, the return rate from occupied housing units has dropped
substantially since 1970, from 87 percent then to 74 percent in 1990.
Final mail response rates for states, cities, American Indian
reservations, and other governmental units are posted on the Census
Bureau's web site as www.census.gov <http://www.census.gov> (click on
"Initial Response Rates").
Stakeholders continue support for long form data: Civic and business
stakeholders moved to counter public criticism of the census long form
by some members of Congress and conservative activists. Last week, the
Census Bureau's five race and ethnic advisory committees, and a majority
of members of the Commerce Secretary's 2000 Census Advisory Committee,
issued a joint statement reaffirming their support for the collection of
demographic and socio-economic information on the long form. These
data, they said, are needed "to [assist] the vulnerable in our
population" and are "critical to the government's decision-making about
our economy and our work-life." Advisory committee members urged
residents who did not mail back a census form to cooperate fully when
enumerators visit and to answer all questions on the short and long
forms.
In Congress, legislators appointed to iron out differences between the
House and Senate versions of the fiscal year 2001 budget resolution,
agreed to drop from the final bill non-binding language on census data
collection approved earlier by the Senate. The Senate amendment,
offered by Sen. Robert Smith (R-NH), suggested that no one should be
prosecuted, fined, or "harassed" for failing to answer census questions
on "race, national origin, living conditions, personal habits or mental
and/or physical condition." Rep. Dan Miller had written to House Budget
Committee Chairman John Kasich, urging the conferees to strike the
census provision because it was not related to budget issues.
Rep. Carolyn Maloney and Sen. Daniel P. Moynihan (D-NY), joined by 52
Representatives and 13 Senators, sent a letter to Census Bureau Director
Prewitt expressing support for the collection of data on the long form.
The legislators, Rep. Maloney said in a written statement, would help
"spread the message... that the census long form is used to collect
vitally important information" about the needs of communities.
Census advisory panels hold joint meeting: The Census Bureau's
professional association and race/ethnic advisory committees, and the
2000 Census Advisory Committee to the Secretary of Commerce, met jointly
on April 14 to review the status of census operations. Senior bureau
officials discussed the progress of the mail phase, Telephone
Questionnaire Assistance and Be Counted activities, recruitment efforts,
and special enumeration operations, as well as plans for enumerator
visits to unresponsive households and the 'cooperation' phase of the
advertising campaign. Compared to the same point in the 1990 census,
they said, the 2000 census was more successful: they had achieved their
hiring goals on time, the public was more aware of the census, and mail
response rates were on target.
Bureau staff also reported that Questionnaire Assistance Centers would
close after April 14; the count of people in 'group quarters' (such as
college dormitories, nursing homes, and military barracks) would run
through April 23; the bureau had started telephone follow-up for
households whose completed forms did not provide demographic information
about all of the listed residents (called 'coverage edit'); the final
phase of the general advertising and promotion campaign would run until
mid-to-late May, with all new ads showing the identification badge
enumerators would wear; and after May, the bureau would develop radio
ads targeting communities where follow-up visits were proceeding
slowly. Census Bureau and federal agency staff also briefed advisory
committee members on plans to tabulate and report data on race and
Hispanic origin. (See the March 13 Census 2000 Initiative News Alert
for a detailed summary of the tabulation policy. New guidance issued by
the Office of Management and Budget can be found at
www.whitehouse.gov/OMB/bulletins/index.html
Several participants expressed concern about the inadequate supply of
foreign language Be Counted forms in many areas. Robert Shapiro,
Commerce Under Secretary for Economic Affairs, acknowledged that demand
for forms in certain languages exceeded the bureau's expectations. The
bureau would boost advertising in affected communities, he said, to
encourage telephone response for those who had not yet found an
opportunity to be counted. The telephone assistance lines
(1-800-471-9424) will remain open through early June. Be Counted forms
could not be photocopied because the data processing equipment could
only scan original questionnaires.
Dr. Prewitt, addressing committee members at the end of the day, called
Census 2000 the best to date operationally. "All big systems worked,"
he said, adding that some components "at the edge" had not worked as
well as the Bureau had hoped. He cautioned that the early operational
successes would not necessarily translate to a "more accurate" count.
Dr. Prewitt also shared his concerns about the lower mail-back rate for
long forms and the criticism of long form questions generated initially
by some radio talk show hosts. Many national leaders, the director
said, failed to capitalize on the opportunity to discuss two important
concepts: the role of information in today's society, and civic
engagement and responsibility at the dawn of a new century. He gave
local elected and civic leaders credit for reaffirming the need for long
form data in light of the public controversy.
Commerce Secretary William Daley has appointed four new organizations to
his Census 2000 advisory panel: the AFL-CIO, the American Foundation for
the Blind, the National Asian Pacific American Legal Consortium
(NAPALC), and the National Association of Latino Elected and Appointed
Officials Educational Fund (NALEO). The Census Bureau also has
increased the number of race and ethnic advisory committees from four to
five, replacing the original committee on the Asian and Pacific Islander
populations with the Advisory Committee on the Asian Population and the
Advisory Committee on the Native Hawaiian and Other Pacific Islander
Populations, to reflect changes in the standard federal race
categories. Another joint meeting of the advisory committees is
scheduled for July 28.
State legislative activities update: On April 9, Virginia Governor James
Gilmore III (R) signed into law a bill (House Bill 1486) that requires
the use of 'unadjusted' census numbers for redrawing congressional,
state legislative, and local political district boundaries. Because
Virginia is one of 16 states covered in whole or part by section 5 of
the 1965 Voting Rights Act, the new law (now called Chapter 884) must be
approved either by the U.S. Department of Justice or the U.S. District
Court for the District of Columbia before it can take effect. In
mid-April, Virginia's Attorney General announced that the state would
attempt to bypass the Department of Justice approval process through
legal action filed with the U.S. District Court. Opponents of
Virginia's law, including civil rights groups, indicated that they would
move to block the move in court.
Questions about the information contained in this News Alert may be
directed to TerriAnn Lowenthal at 202/484-2270 or, by e-mail at
terriann2k(a)aol.com. For copies of previous News Alerts and other
information, use our web site www.census2000.org . Please direct all
requests to receive News Alerts, and all changes in
address/phone/fax/e-mail, to the Census 2000 Initiative at
Census2000(a)ccmc.org or 202/326-8700. Please feel free to circulate this
information to colleagues and other interested individuals.
from: Decennial Media Relations Team, e-mail: 2000usa(a)census.gov
The Census Bureau announced today that "the initial response rate" for
Census 2000 the percentage of questionnaires returned before enumerators
begin knocking on the doors of nonrespondents later this month was 65
percent, matching the 1990 rate. Beginning on April 27, Bureau staff
will begin contacting all of those who did not return their forms.
About 120 million census questionnaires were mailed or hand-delivered to
homes across the country in March. As questionnaires were returned, the
Census Bureau posted daily, from March 27 to April 11, on its Internet
site http://rates.census.gov/ the rates of return for the country, the
states and about 38,000 local and tribal governments. (Tract rates are
also available http://rates.census.gov/tracts/ )
Under a promotional campaign called '90 Plus Five, the Census Bureau
challenged the nation, the states and local communities to do as well as
they did in 1990, plus 5 percent. By April 18, about 15 percent of the
entities had met their goals. By April 11, the Census Bureau reported
that the response rate the number of those who either mailed back,
transmitted via the Internet or had a telephone assistance operator take
their answers over the phone stood at 62 percent, a percentage point
better than what it had projected for planning and budgetary purposes.
In the next phase of the census, called "non-response follow-up," as
many as half a million temporary workers, with address lists and maps,
will visit housing units the Census Bureau did not hear from. They will
make up to six attempts to contact nonresponding households three
personal visits and three phone calls. This operation will extend
through July 7.
Last week, Chuck Purvis wrote a long, useful message to all who are working on
the Work-UP program to help improve the workplace coding of the Decennial
Census. We at the Census Bureau have some additional thoughts to share with
you.
Chuck wrote:
1. The TIGER files *can't* be corrected as part of this WORK-UP
project. So, if the TIGER/ArcView files are missing key streets, or
address ranges on streets, then I think the best solution is to code
to the closest available block-face. The idea is to get the
employer location accurately coded to the nearest census block.
Once the Census Bureau is able to geocode to the nearest block, then
they will be able to aggregate to the CTPP Travel Analysis Zones
created as part of TAZ-UP.
If, on the other hand, the employer is within a large new subdivision
that *isn't* included in the current set of TIGER/ArcView files, then
I don't know what happens. I haven't come across this problem, though
I have had problems with streets-without-address ranges.
Census Bureau Response:
Item #1: You are correct in saying that these TIGER files cannot be corrected.
The files that are being used as a part of Work-UP are Tiger Line 98. In the
past year, a lot of work has been done to improve TIGER. Most of these changes
are now in 'live' TIGER and will be reflected in Tiger Line 99 and 00. The
Census Bureau is not prepared to process any changes that result from the
Work-UP program.
The Work-UP software is designed to allow you to map spot employers, taking into
account insufficient address coverage and missing streets. As long as you keep
the correct address we can match respondents to that employer and then they will
be coded to whatever Tiger Line segment that you put them on. In cases where
the street is missing just place the employer on a nearby street. BUT YOU NEED
TO MAKE SURE THEY ARE STILL WITHIN THE SAME TAZ!!!!! As long as the line
segment you associate the employer with is in the same TAZ your work place data
and flows will be correct.
Chuck wrote:
2. The alias street name problem is one I've been battling with as
early as this morning. The TIGER file will probably have the older
name (e.g., North Main Street), but the "newcomers" to the area may
be used to the new street name (e.g., Last Chance Gulch). [This is
very common in the City of San Francisco where they rename streets
after famous labor leaders or writers, e.g., Army Street is now
Cesar Chavez Street, Ivy Street is now Lech Walesa St., Stark
St. is now Jack Kerouac Alley, etc.] In answering Census 2000,
respondents will use what they want (North Main or Last Chance
Gulch). Maybe the Census Bureau Geography Division has a "street name
alias file" that can be used for improved workplace coding???? (If
not, they should!)
My recommendation may be to add duplicate employer records that have
the alias street name, e.g., if the employer record provided by the
Census Bureau is Joe's Bar & Grill, 100 North Main Street, then I
would add a new record (at the same precise geocoded location) as:
Joe's Bar & Grill, 100 Last Chance Gulch. (Now, if there were 500
establishments along this Last Chance Gulch, then I'd have second
thoughts about adding alias employer records.)
Census Bureau Response:
Item #2: Alias street names are a problem. To the best of our knowledge there
is not a street name alias feature in TIGER. We are talking here at Census
about a possible list of street name aliases that can be used to figure out that
respondents may report "Last Chance Gulch" but it may be shown in TIGER as "Main
St. " We will put together a suggested format of such a list and send it out to
the list serve soon. If you have any questions or comments, please call or
email me.
(email: clara.a.reschovsky(a)ccmail.census.gov or phone: 301-457-2454)
In the meantime, I would encourage you to add a second record for large
employers on streets that have alternative names. I understand that you
probably cannot do this for a lot of smaller employers.
Chuck Wrote:
Strategy:
I've been working on WORK-UP for the past week, and I've completed
initial work on five of my nine counties. In the Bay Area we have 300
thousand geocoded employer locations, and 32,800 un-geocoded employer
locations. Of the 32,800 ungeocoded, 1,102 are for employers of 50+
employees. In my large counties, I'm only reviewing the ungeocoded
for the 50+ employees (10+ employees in Napa County). The goal in my
initial work is to "eliminate the big red dots" (ungeocoded
workplaces of 50+).
I'm also reviewing *all* of the geocoded records for large (250+ or
500+) employers to make sure they're in the right locations. They
mostly are, though there were some problems (UCBerkeley was located
in downtown Berkeley instead of on the campus; Marine World USA
amusement park was on the wrong side of the road....) And some of
the research is actually kind of fun (finding the correct location
for Domaine Chandon & Niebaum-Coppola wineries; "Skywalker Sound" in
Marin County....) Unfortunately, a lot of the ungeocoded records are
shopping centers (they're a real mess.)
So, my first phase strategy is to correct the ungeocoded large
employers (10+ or 50+ employees); and to review the existing
geocoding for very large employers (250+ or 500+). I haven't yet
devised a second phase strategy.
Census Bureau Response:
Strategy: Chuck, your strategy sounds good. It is what we had figured many
MPOs would be doing. The shopping centers are important. We often have
responses with the store name as the employer and the shopping center name as
the only address information. If the shopping center is geocoded to the correct
location, then we can successfully geocode these responses. Another special
type of record is public schools. Usually people who work for school districts
will put down the name of the school district as their employer. Sometimes they
put the school name where they actually work in the address field and sometimes
they give us partial address information. If you can add records for each
school in your area and a geocoded address it will help in coding these
respondents to the correct TAZ.
Chuck wrote:
3. My question: would it be useful to include the acronym of the
employer in the employer name alias field? For example, I doubt that
our friends at AASHTO will fill in American Association of State
Highway Transportation Officials, but rather just use AASHTO....
Census Bureau Response:
Item 3: On the issue of acronyms, YES PLEASE, PLEASE, PLEASE!!!! Adding in
acronyms and abbreviations as aliases or separate records is very useful,
because some of these acronyms are nationwide in scope while some are highly
localized. In what we have seen in testing our coding systems with Dress
Rehearsal Data and ACS data, we get a lot responses that are acronyms or local
abbreviations. We have a clerical staff that attempts to match these responses
with employers in the employer file. There is a learning curve on figuring out
what some of these abbreviations really are that often make it very difficult
to resolve these cases.
If you have any questions about any of the information above, please call or
email me.
Clara Reschovsky
301 457 2454
From: Census2000(a)ccmc.org
Census Long Form Comes Under Attack, As
Mail-back Phase Winds Down
Plus: Congress and GAO Keep Watch Over Count;
State Legislative Update; and more.
Census Bureau Director Kenneth Prewitt told a congressional oversight
panel last week that he is "cautiously optimistic that we will achieve
the 61% [mail response rate] on which we based budget and staffing
[assumptions]." But he also told members of the House Subcommittee on
the Census at the April 5th hearing that the "public atmosphere that has
trivialized and discredited the long form" has contributed to a
lower-than-expected long form mail-back rate and could affect the morale
of census takers who must visit unresponsive households to collect the
information. The Director called upon Congress to explain the important
uses of data collected on the long form to the public.
Dr. Prewitt reported that the difference so far between short form and
long form mail response was about 12 percent, twice the differential
rate at the same stage of the 1990 census. The final response
difference between the two forms in 1990 was 4.5 percent, according to a
1995 National Academy of Sciences report. Each five percentage point
difference in the response rate translates to a one percent drop in
overall response rates, since the long form is sent only to a sample
(one in six) of households.
Saying the "widespread attack on the long form could have serious
consequences," the Director said he also was concerned about the
reliability of some long form data if too many people refused to provide
certain information, such as their income. "High non-response to the
long form and high non-compliance with particular items on forms
returned," Dr. Prewitt warned, could "combine to push data below our
quality threshold," a situation that might prevent the bureau from
releasing some of the information. Census enumerators will not
follow-up with households that fail to answer some of the questions if
the returned questionnaire includes basic information about the number
of people living there.
Subcommittee Chairman Dan Miller (R-FL) said he was "very impressed by
the complexity of current ongoing operations" and praised the bureau for
reporting mail response rates publicly on its Web site. He described
the public's reaction to the Census 2000 long form as a symptom of
broader concerns about personal privacy in an age of advanced technology
and the Internet. People should not "lay the blame for [the long form]
controversy at the feet of Republicans," Rep. Miller said. The chairman
blamed "some of the recent scandals involving this administration,
particularly the misuse of the FBI files" for increasing distrust of
government, which he said also affected response to the census. The
subcommittee will hold hearings later in the year on the long form and
privacy issues, with an eye toward "eliminat[ing] the long form for the
2010 Census."
Rep. Carolyn Maloney (D-NY), the subcommittee's ranking Democratic
member, noted that "the hard work on the advertising campaign,
partnerships, and other promotional activities appears to be paying
off." She called statements about the perceived intrusiveness of the
long form by some Republican members of Congress and presidential
candidate George W. Bush (see below) "outrageous, irresponsible,
demagoguery - pandering to fringe groups and the radio talk show
circuit. They threaten the success of the Census by driving response
down."
'Long form' criticism dominates public debate: Public criticism of the
long form first surfaced after questionnaires were mailed to most
households in mid-March (or hand delivered to non-city style
addresses). Dr. Prewitt noted in his congressional testimony that
leaders of the Libertarian Party had encouraged people not to answer any
questions beyond the number of household residents, and that some radio
talk show hosts had fueled public concern about "intrusive" and
"unconstitutional" questions.
In addition, news reports widely quoted a spokesman for Senate Majority
Leader Trent Lott (R-MS) as saying that people should provide "the basic
census information" but that if they "feel their privacy is being
invaded by [some] questions, they can choose not to answer." Texas
Governor George W. Bush (R) also was quoted as saying, "We want as
accurate a count as possible, but I can understand why people don't want
to give over that information to the government. If I have the long
form, I'm not so sure I would do it either." Rep. Tom Coburn (R-OK)
issued a press release saying he was "appalled and outraged by the
intrusiveness of this questionnaire. ...It is ridiculous for the Census
Bureau to ask personal questions that have nothing to do with their
Constitutional mandate to count" the population.
In his weekly radio address on April 1, President Clinton urged
Americans to fill out the short and long forms completely and noted that
"every question on both forms was reviewed by Congress two years ago."
He cited the allocation of federal funds, apportionment and
redistricting, local decisions about where to build schools, roads, and
hospitals, and business investment decisions as reasons to answer the
census. "[I]f we want to make good decisions about where we need to go
as a nation, we first have to know where we are," the President said.
The U.S. Census Monitoring Board sent a letter to Director Prewitt,
expressing their "continued bipartisan support for the Census 2000 long
form." In the letter, a copy of which was sent to all members of
Congress, Republican Co-Chair J. Kenneth Blackwell and Democratic member
Everett M. Ehrlich said, "[A]t this late date, we do not believe it is
productive to renew the debate over census content."
The Census Act (title 13, United States Code) authorizes a fine up to
$100 for failure to answer all census questions; intentionally giving
false answers can draw a fine up to $500. Prosecution of individuals
who refuse to answer the census is very rare; Director Prewitt cited
only two cases since enactment of the law.
'Long form' debate continues in Congress: House Republicans, led by
Speaker Dennis Hastert (R-IL), held a press conference on April 6 to
address the controversy over the long form questions. Republican
Conference Chairman J.C. Watts (R-OK) issued an "appeal to the American
people to take part in the census," but also said, "[I]f you just fill
out the first six questions and send it in, that helps, because it
allows people to be counted." The short form asks six questions of all
household members; the person completing the form also tells if the home
is rented or owned and how many people live there.
Speaker Hastert also urged people to send in their forms. But when
asked by a reporter if people should complete the entire long form, he
said "every American ought to fill it out to the best of their ability.
If they have reservations about ... some things that they think are
private," they should "make their own decision." Rep. Carolyn Maloney
criticized the Speaker's statements, saying he "continued to put at risk
the ability of this nation to collect quality data."
The next day, the Senate, with no public debate, adopted an amendment to
its version of the fiscal year 2001 budget resolution (S. Con. Res.
101), expressing its belief that no one should be prosecuted, fined, or
"harassed" for failing to answer census questions on "race, national
origin, living conditions, personal habits or mental and/or physical
condition." The "sense of the Senate" amendment, offered by Sen. Robert
Smith (R-NH), also "encouraged [Americans] to send in their census
forms." The annual budget resolution, which sets broad fiscal policy
and spending limits, is a congressional blueprint that does not require
the president's signature.
Census Bureau Director Prewitt issued a statement in response to the
Senate action. "Census 2000 is not designed by law as a pick and choose
exercise," the director said. He questioned whether the bureau's policy
of making six attempts to collect information from households that did
not mail back a form "would constitute harassment" in the eyes of the
Senate.
Also last week, Rep. Michael "Mac" Collins (R-GA) introduced the "Common
Sense Census Enforcement Act" (H.R. 4188), to eliminate the penalty for
refusing to answer census questions that are not on the short form.
Mail phase of Census 2000 nears end: The mail-out/mail-back phase of
Census 2000 ends on April 11. After that date, the Census Bureau will
start to prepare the list of addresses to be visited by census takers in
the Nonresponse Follow-up (NRFU) operation, eliminating those households
in its Master Address File that mailed back a census form. However, the
bureau will continue to accept and process mailed questionnaires
received by April 17. It will then prepare a second list for Local
Census Offices of additional addresses that will not require a personal
visit.
The NRFU operation - the most time-consuming, difficult, and costly
phase of the census - will run from April 27 through approximately July
7. Enumerators must make six attempts to collect data from an
unresponsive household - three personal visits and three telephone
contacts - before turning to neighbors, landlords, or other
knowledgeable sources to obtain basic data about the residents.
GAO evaluates census progress: The U.S. General Accounting Office (GAO),
the nonpartisan congressional audit and investigative agency, also
testified at several recent House census subcommittee hearings. At the
April 5th hearing, GAO officials reported on mail response rates,
completed census operations (such as delivery of questionnaires to rural
areas and enumeration of people without a fixed address), and the
Questionnaire Assistance Centers. They noted that response rates varied
widely at the local census office level, ranging from 30 percent to 72
percent as of Census Day (April 1). GAO staff concluded that the
"update/leave" questionnaire drop-off operation in rural areas "appears
to have improved the quality of the address list," but they also
observed difficulties such as "finding 'hidden' housing units and
accessing gated properties," as well as late delivery of training and
other materials at most of the local offices they visited. The GAO
testimony, "2000 Census: Progress Report on the Mail Response Rate and
Key Operations (GAO/T-GGD/AIMD-00-136), also discusses the status of
questionnaire processing.
The GAO issued a new report in February, entitled "2000 Census: Actions
Taken to Improve the Be Counted and Questionnaire Assistance Center
Programs" (GAO/GGD-00-47, February 2000). To obtain GAO reports or
testimony, call 202/512-6000 (TDD: 202/512-2537) or visit GAO's Web
site, www.gao.gov <http://www.gao.gov>.
State legislative update: The U.S. Department of Justice (USDOJ) has
accepted a request from Arizona Governor Jane Hull (R) to withdraw a
state law barring the use of statistically corrected census numbers from
the Voting Rights Act pre-clearance process. The status of the
governor's request was uncertain because the state's Attorney General
originally submitted the law for pre-clearance. Arizona is one of 16
states required under section 5 of the Act to obtain USDOJ approval for
any change to state election laws. The law, enacted last spring, cannot
take effect without the department's approval.
In its March 24th letter to Governor Hull, the Justice Department also
said it would use statistically corrected census numbers in analyzing
post-census redistricting plans under federal civil rights laws. "[T]he
Attorney General's review and assessment of any redistricting plan will
not be restricted by the data that a particular jurisdiction elects to
use in its redistricting process or its submission to us," department
officials advised. The Census Bureau plans to correct undercounts and
overcounts in the initial census numbers on the basis of a quality-check
survey called the Accuracy and Coverage Evaluation, or ACE, survey, and
transmit the second set of numbers to state legislatures by April 1,
2001. Arizona is one of five states (the others are Alaska, Colorado,
Kansas, and Virginia) that have enacted laws prohibiting the use of
census figures derived with sampling methods for redistricting purposes.
Late last month, the Louisiana State Senate passed a bill requiring the
use of unadjusted census data for redistricting. However, the state's
House of Representatives previously rejected a similar measure (HB
184). Senate Bill 85 was sent back to the House, where it awaits
further action.
Census Monitoring Board activities: The U.S. Census Monitoring Board
issued a report to Congress on April 1, entitled "Field Observations of
the New York and Dallas Regional and Local Census Offices, Alaska
Enumeration, and Household Matching Training." The report was prepared
jointly by the four congressional appointees (Republicans) and four
presidential appointees (Democrats); both sides of the oversight panel
occasionally have issued separate findings in the past. To obtain
copies of the report, visit the Presidential members' Web site at
www.cmbp.gov <http://www.cmbp.gov> or the Congressional members' Web
site at www.cmbc.gov <http://www.cmbc.gov>, or call 202/457-9900 or
202/457-5080, respectively. A Spanish-language version also is
available.
Update on collection of race data: The guidance recently issued by the
Office of Management and Budget on tabulating multiple race data for the
purpose of monitoring and enforcement of civil rights law is available
on the Internet at www.whitehouse.gov/OMB/bulletins/index.html
<http://www.whitehouse.gov/OMB/bulletins/index.html>. The relevant
document, dated March 9, is OMB Bulletin 00-02.
Stakeholder activities: Several of the Census Bureau's advisory
committees will hold a joint meeting on April 14 to review the status of
Census 2000 operations, as well as plans for reporting data on race and
Hispanic origin for various purposes. The bureau's five advisory panels
on race and ethnicity, committee of professional associations, and the
Commerce Secretary's 2000 Census Advisory Committee will participate in
the discussions. The meeting, which is open to the public, will be held
at the Sheraton Crystal City Hotel, 1800 Jefferson Davis Highway,
Arlington, VA (tel. 703/486-1111), from 8:40 a.m. to 5:00 p.m.
New resource reminder: The Census Information Exchange is an electronic
resource designed to keep community-based stakeholders and members of
the press informed about fast-moving census operations. The Web site
also encourages sharing of effective outreach and promotion strategies,
as well as concerns, among communities at greatest risk of an
undercount. Go to www.censusnetwork.org
<http://www.censusnetwork.org>. The Web site is a joint project of the
Census 2000 Initiative, the Leadership Conference on Civil Rights, and
the Leadership Conference Education Fund.
Questions about the information contained in this News Alert may be
directed to TerriAnn Lowenthal at 202/484-2270 or, by e-mail at
terriann2k(a)aol.com. For copies of previous News Alerts and other
information, use our Web site www.census2000.org
<http://www.census2000.org>. Please direct all requests to receive News
Alerts, and all changes in address/phone/fax/e-mail, to the Census 2000
Initiative at Census2000(a)ccmc.org or 202/326-8700. Please feel free to
circulate this information to colleagues and other interested
individuals.
Ed -- The following is in response to the CTPP listserve question about
the request for Social Security Numbers on some decennial census
questionnaires.
Randall Neugebauer of the Planning, Research, and Evaluation Division,
U.S. Census Bureau, provided me with the following information on the
SSN, Privacy Attitudes, and Notification (SPAN) experiment.
The SPAN is one of three panel experiments taking place concurrently
with Census 2000. Each experiment looks at innovative methodologies
and/or technologies that might be considered for 2010. Use of
administrative records is one such notion being considered and tied to
it is the collection and use of the SSN as a "match" key. The SPAN
gathers data as to respondent behavior in providing the SSN. A total of
21,000 households of the mailout/mailback universe received a short form
requesting the SSN. The cover letter, however, explains that providing
it is voluntary. Attached is a current list of Frequently Asked
Questions (FAQs) for the SPAN experiment, which I believe will address
your immediate concerns. These FAQs are being used by our telephone
questionnaire assistance operators to field questions on the SSN and
other topics related to the SPAN experiment.
(Ed's note: I converted the file to an RTF format. if your have
problems with it let me know.)
Al:
I'm responding to you and the CTPP-News listserv, since these effect
many different folks. My hopes are that the Census Bureau or USDOT
staff can corroborate or correct me as needed.
1. The TIGER files *can't* be corrected as part of this WORK-UP
project. So, if the TIGER/ArcView files are missing key streets, or
address ranges on streets, then I think the best solution is to code
to the closest available block-face. The idea is to get the
employer location accurately coded to the nearest census block.
Once the Census Bureau is able to geocode to the nearest block, then
they will be able to aggregate to the CTPP Travel Analysis Zones
created as part of TAZ-UP.
If, on the other hand, the employer is within a large new subdivision
that *isn't* included in the current set of TIGER/ArcView files, then
I don't know what happens. I haven't come across this problem, though
I have had problems with streets-without-address ranges.
2. The alias street name problem is one I've been battling with as
early as this morning. The TIGER file will probably have the older
name (e.g., North Main Street), but the "newcomers" to the area may
be used to the new street name (e.g., Last Chance Gulch). [This is
very common in the City of San Francisco where they rename streets
after famous labor leaders or writers, e.g., Army Street is now
Cesar Chavez Street, Ivy Street is now Lech Walesa St., Stark
St. is now Jack Kerouac Alley, etc.] In answering Census 2000,
respondents will use what they want (North Main or Last Chance
Gulch). Maybe the Census Bureau Geography Division has a "street name
alias file" that can be used for improved workplace coding???? (If
not, they should!)
My recommendation may be to add duplicate employer records that have
the alias street name, e.g., if the employer record provided by the
Census Bureau is Joe's Bar & Grill, 100 North Main Street, then I
would add a new record (at the same precise geocoded location) as:
Joe's Bar & Grill, 100 Last Chance Gulch. (Now, if there were 500
establishments along this Last Chance Gulch, then I'd have second
thoughts about adding alias employer records.)
Strategy:
I've been working on WORK-UP for the past week, and I've completed
initial work on five of my nine counties. In the Bay Area we have 300
thousand geocoded employer locations, and 32,800 un-geocoded employer
locations. Of the 32,800 ungeocoded, 1,102 are for employers of 50+
employees. In my large counties, I'm only reviewing the ungeocoded
for the 50+ employees (10+ employees in Napa County). The goal in my
initial work is to "eliminate the big red dots" (ungeocoded
workplaces of 50+).
I'm also reviewing *all* of the geocoded records for large (250+ or
500+) employers to make sure they're in the right locations. They
mostly are, though there were some problems (UCBerkeley was located
in downtown Berkeley instead of on the campus; Marine World USA
amusement park was on the wrong side of the road....) And some of
the research is actually kind of fun (finding the correct location
for Domaine Chandon & Niebaum-Coppola wineries; "Skywalker Sound" in
Marin County....) Unfortunately, a lot of the ungeocoded records are
shopping centers (they're a real mess.)
So, my first phase strategy is to correct the ungeocoded large
employers (10+ or 50+ employees); and to review the existing
geocoding for very large employers (250+ or 500+). I haven't yet
devised a second phase strategy.
3. My question: would it be useful to include the acronym of the
employer in the employer name alias fielf? For example, I doubt that
our friends at AASHTO will fill in American Association of State
Highway Transportation Officials, but rather just use AASHTO....
Chuck Purvis, MTC
> Hi Chuck, We are working on the Work-Up assignment. We have run
> into a number of problems that we need guidance with. One problem
> is that we are finding road segments in the tiger line files that
> do not have a to and from address or zip code coded into them,
> thus they are unable to be located. Another problem is that some
> of the addresses are on streets that are not in the tiger line
> files as a segment. And a third problem is that some of the
> streets actually have two (2) different names, i.e. some people
> think they are N Main and others think it is Last Chance Gulch.
> What makes some of these problems persistent is that the ArcView
> file does not allow for editing. The buttons are turned off or
> greyed out/invisible. What is the silver bullet strategy that we
> need to follow on this assignment. Thank you.
>
> Al V.
*******************************************************
e-mail: cpurvis(a)mtc.ca.gov
Chuck Purvis, AICP
Senior Transportation Planner/Analyst, Planning Section
Metropolitan Transportation Commission
101 Eighth Street, Oakland, CA 94607-4700
(510) 464-7731 (voice) (510) 464-7848 (fax)
WWW: http://www.mtc.ca.gov/
MTC DataMart: http://www.mtc.ca.gov/datamart/
MTC FTP Site: ftp://ftp.abag.ca.gov/pub/mtc/planning/
*******************************************************
My apologies for a inference based on recent news plus an April 1 check of
the "scoreboard" which showed your state far behind not only the national
average but also the other confederate states. A more recent check
indicates this gap is now mostly erased. Is that bad news?
This'll teach me to try catching up to my e-mail on weekends - not only am I
even more unprofessional than usual then, but quite forgetful i.e. I
completely forgot that the j-to-work question is used for the formal
definition of MSA's. But beyond that it does concern me a bit that the
Census web site cited in other responses seems to have more "assists in" and
"used by" than specific Congressional directives tied to specific Census
questions. The former is something all us planner types understand, but not
by most of the people we serve. (That is not intended to demean the hard
work that Ed Christopher and others have put in keeping transportation
visible in the long form - perhaps if I were in a larger metro area and/or
CAAA non-compliant I would understand much better the needs.)
> -----Original Message-----
> From: Jim Wilkinson [SMTP:JDWilkin@grpc.com]
> Sent: Tuesday, April 04, 2000 9:52 AM
> To: Granato, Sam
> Subject: Re: [CTPP] Experimental Census Forms
>
> Sam Granato:
>
> My senator IS majority leader. Was your remark intended
> to stigmatize us Mississippians as even more lacking in desire
> to fill out questionnaires than the average American? If it was,
> wouldn't that really make us more American than the homogenized
> variety? The truth is that there is still some resistance to the idea
> of collaboration 130-plus years into the occupation.
> >
