[Congressional Record Volume 144, Number 108 (Tuesday, August 4, 1998)]
[House]
[Page H7175]
From the Congressional Record Online through the Government Publishing Office [www.gpo.gov]




                                 CENSUS

  (Mr. SAWYER asked and was given permission to address the House for 1 
minute and to revise and extend his remarks and include extraneous 
material.)
  Mr. SAWYER. Mr. Speaker, later on today we are going to take up an 
issue of enormous importance to the Nation, and that is how we count 
and measure ourselves. Last week in a debate that was largely 
constructive on the floor, we had a discussion that was thoughtful and 
well informed. However, insofar as one of our Members, the gentleman 
from Florida (Mr. Miller), suggested that there was a hand-picked 
nature of the scientific panels that recommended statistical sampling 
methods, I wanted to share with the Members the reply of the American 
Statistical Association, whose president wrote to me over the weekend 
and said that the members of the panel that made this recommendation 
are recognized by their peers as among the Nation's leading experts on 
sampling large human populations. It included Janet Norwood, who served 
three administrations, Carter and Reagan and Bush, with, as the New 
York Times put it, her near legendary reputation for nonpartisanship. 
Dr. Moore, the president of the American Statistical Association, went 
on to cite the extraordinary quality of the members of that panel.
  Mr. Speaker, I would like to insert into the Record at this point the 
substance of his letter.

                             American Statistical Association,

                                   Alexandria, VA, August 3, 1998.
     Congressman Thomas Sawyer,
     Longworth House Office Building,
     Washington, DC.
       Dear Congressman Sawyer: Thank you for sending me the 
     Congressional Record account of debate on H. Res. 508, 
     containing the remarks of several Members regarding the use 
     of statistical sampling methods in the 2000 Census. Despite 
     obvious differences in perspective, the discussion is 
     thoughtful and well-informed, the sole major exception being 
     the incorrect statement by Mr. Miller of California that the 
     Census Bureau plans to intentionally not count 10 percent of 
     the population. The overall level of the discussion does 
     credit to the House of Representatives.
       I do wish to respond on behalf of the American Statistical 
     Association to the remarks of Mr. Miller of Florida 
     concerning the ``hand-picked'' nature of the scientific 
     panels that have recommended consideration of statistical 
     sampling methods. I refer specifically to the Blue Ribbon 
     Panel of the American Statistical Association. The members of 
     this panel are recognized by their peers as among the 
     nation's leading experts on sampling large human populations. 
     They are certainly not identified with any political 
     interest.
       The ASA Blue Ribbon Panel included Janet Norwood, who 
     served three administrations as Commissioner of Labor 
     Statistics from 1979 to 1991. On her retirement, the New York 
     Times (December 31, 1991) spoke of her ``near-legendary 
     reputation for nonpartisanship.'' Dr. Norwood is a past 
     president of ASA, as is Dr. Neter of the University of 
     Georgia, another panel member. Like these, the other members 
     of the panel have been repeatedly elected by their peers to 
     posts of professional responsibility. For example, Dr. Rubin 
     of Harvard University is currently chair of ASA's Section on 
     Survey Research Methods, the statistical specialty directly 
     relevant to the census proposals. I assure you that this 
     panel was selected solely on the basis of their widely 
     recognized scientific expertise. Their judgment that 
     ``sampling has the potential to increase the quality and 
     accuracy of the count and to reduce costs'' is authoritative.
       Mr. Miller, in hearings before his committee, has indeed 
     produced reputable academics who disagree with the findings 
     of the ASA Blue Ribbon Panel and the several National 
     Research Council panels which reported similar conclusions. 
     Those whose names I have seen lack the expertise and 
     experience in sampling that characterize the panel members. 
     Statistics, like medicine, has specialties: one does not seek 
     out a proctologist for heart bypass surgery.
       I do wish to make it clear that the American Statistical 
     Association takes no position on the political or 
     constitutional issues surrounding the census. We also express 
     no opinion on details of the specific proposals put forth by 
     the Census Bureau for employing statistical sampling. As the 
     nation's primary professional association of statisticians 
     and users of statistics, we wish to make only two points in 
     this continuing debate:
        Estimation based on statistical sampling is a 
     valid and widely-based scientific method. The general attacks 
     on sampling that the census debate has called forth from some 
     quarters are uninformed and unjustified.
        The non-partisan professional status of government 
     statistical offices is a national asset that should be 
     carefully guarded. We depend on the statistical professionals 
     in these offices for information widely used in both 
     government and private sector decisions. Attacks on these 
     offices as ``politicized'' damage public confidence in vital 
     data.
       Thank you for the opportunity to make these comments.
           Sincerely yours,
                                                   David S. Moore,
     President.

                          ____________________