[Congressional Record Volume 144, Number 66 (Thursday, May 21, 1998)]
[House]
[Page H3738]
From the Congressional Record Online through the Government Publishing Office [www.gpo.gov]




                              1990 CENSUS

  The SPEAKER pro tempore. Under a previous order of the House, the 
gentlewoman from New York (Mrs. Maloney) is recognized for 5 minutes.
  Mrs. MALONEY of New York. Mr. Speaker, last week the Subcommittee on 
the Census held a hearing on the 1990 census, and once again, the 
record is full of mistakes. Let me, once again, put the facts on the 
table so that Congress can make its decisions on what really happened.
  Some of the errors at the hearing are because most of the members and 
staff on the Republican side are new to the issue, and get confused 
about which facts apply to 1990 and which to previous censuses. Some of 
the errors occurred because two of the three statisticians who 
testified had no previous experience with the census undercount issue. 
It is often useful to get fresh minds to think about a problem, but in 
this case it also resulted in people making statements when they did 
not have the facts to support their position.
  At last week's hearing the statement was made that in 1990 50 percent 
of the undercount came from problems in the address list. That is 
wrong. The facts are that in 1990 70 percent of those missed were in 
households that were counted, and the address list was 97.5 percent 
accurate.
  One of the witnesses criticized the Post Enumeration Survey because 
it put more people into the census than other methods said were 
missing. That too is wrong. The problem with the Post Enumeration 
Survey in 1990 was that despite the Census Bureau's best efforts, it 
will missed people. In 1990 the Post-Enumeration Survey showed that the 
census net undercount was 1.6 percent, while the Census Bureau's 
Demographic Analysis, which they have done since 1940, showed an 
undercount rate of 1.8 percent.
  Finally, one witness said that after the 2000 census there would be 
no Demographic Analysis. That is just wrong.
  These are not all of the mistakes made at that hearing, but they do 
illustrate the point that new-comers to this issue are having a hard 
time understanding the facts. What I find more troubling is the 
intentional misrepresentation of information.
  At last weeks hearing the majority tried to suggest that the 1990 
census was actually better than the 1980 census. To do that they took 
the measure of the undercount of Blacks from Demographic analysis in 
1980 and compared it to the Post Enumeration Survey estimate of 
undercount for Blacks for 1990. I would hope that our Subcommittee 
Chairman is a good enough statistician to know that is wrong. In 1980, 
Demographic Analysis shows that the undercount of Blacks was 4.5 
percent. In 1990 it was 5.7 percent. The Post Enumeration Survey shows 
a lower undercount for Blacks because even after the Census Bureau's 
best efforts, the survey still misses some people.
  Unfortunately, it wasn't bad enough that the majority tried to 
minimize the fact that the census misses millions of poor and 
minorities. What they are really concerned about is that the Census 
Bureau may take out the millions of people who are counted twice. On 
the one hand they are saying that they don't care that millions of 
Blacks, and Hispanics and Asians and the poor are left out of the 
census. At the same time they are saying, don't you dare take out any 
of those white suburbanites who were counted twice in my district.

  Following the 1990 census, there was a broad and bipartisan consensus 
that we had to find a better way to conduct the census--to improve the 
accuracy of the counts and to control the cost. For several years, 
while experts toiled over alternative methods and the Census Bureau 
threw its energies into research, Republican in Congress paid little 
attention. In fact, the appropriators kept prodding the Census Bureau 
to move more quickly to develop a plan for a better census.
  It was not until consultants working for the Republican National 
Committee decided that the use of sampling methods to help fix the 
problem of undercounting might hurt Republicans in the redistricting 
process that the party leaders stood up and took notice. All of a 
sudden, scientific methods that the National Academy of Sciences, the 
General Accounting Office, and the Commerce Department's Inspector 
General had recommended a few years earlier, were no good. They were 
``unscientific'' according to a report pushed through by the majority 
of the Government Reform Committee. All of a sudden, the National 
Academy of Science was politically biased, and the Census Bureau 
incapable of conducting a census. Even the Speaker of the House changed 
his position on the issue. In 1991 he supported adjustment. In 1996 he 
did a 360 degree turn around.
  Now, I ask you: Is there any basis for the strong and sudden 
opposition to the use of scientific sampling methods in the 2000 census 
among Republicans, other than their concern that a more accurate count 
of African Americans and Hispanics and Asian Americans and poor people 
might somehow work to their disadvantage when political district 
boundaries are drawn.
  Let's not try to fool the American people with talk about the 
efficacy of choosing this post-stratification variable or that. All of 
this minutiae is meant to do one thing only: to confuse the American 
people, to make them think the Census Bureau isn't capable of honest, 
to undermine public confidence in the entire census process. All 
because Republican leaders believe that their hold on political power 
will slip if the census more accurately reflects the true composition 
of our diverse population.
  How utterly irresponsible! How utterly devoid of any shred of moral 
imperative. I ought to be angry or outraged. Instead I am genuinely 
saddened. Saddened because one of the most fundamental activities of 
our democratic system of governance is being belittled and diminished 
for partisan political advantage. The census and the Census Bureau may 
forever be tarnished by this organized effort to tear down the 
messenger because some people don't like the message.
  This is a sad day and a low point for this Congress. I hope my 
Republican colleagues will look within themselves before they continue 
on their campaign of terror against science in general, and the Census 
Bureau in particular. I hope they will decide if they really want to 
live with the consequences of their plan to ensure that the 2000 census 
will continue to miss millions of people and that the Census Bureau 
will be diminished in the eyes of the public.

                          ____________________