2000 CENSUS MAIL OUTREACH IMPROVEMENT ACT
|April 13, 1999- Committed to the Committee of the Whole House on the State of the Union and ordered to be printed|
|Mr. BURTON, from the Committee on Government Reform, submitted the following|
|[To accompany H.R. 928]|
|[Including cost estimate of the Congressional Budget Office]|
The Committee on Government Reform, to whom was referred the bill (H.R. 928) to require that the 2000 decennial census include either a general or targeted followup mailing of census questionnaires, whichever, in the judgment of the Secretary of Commerce, will be more effective in securing the return of census information from the greatest number of households possible, having considered the same, report favorably thereon without amendment and recommend that the bill do pass.
|I. Summary of Legislation||2|
|II. Background and the Need for Legislation||2|
|III. Legislative Hearings and Committee Actions||3|
|IV. Committee Hearings and Written Testimony||3|
|V. Explanation of the Bill||3|
|VI. Compliance with Rule XIII||4|
|VII. Budget Analysis and Projections||4|
|VIII. Cost Estimate of the Congressional Budget Office||4|
|IX. Specific Constitutional Authority for This Legislation||5|
|X. Committee Recommendation||5|
|XI. Congressional Accountability Act; Public Law 104-1||6|
|XII. Unfunded Mandates Reform Act; Public Law 104-4, Section 423||6|
|XIII. Federal Advisory Committee Act (5 U.S.C. App.) Section 5(b)||6|
H.R. 928, the 2000 Census Mail Outreach Improvement Act of 1999, requires the 2000 decennial census to include a second mailing of census questionnaires, either targeted (to those households who have not yet responded by mail) or general (to each household included in the original mailing). This bill also gives the Secretary of Commerce the authority to choose which method (targeted or general mailing) will achieve the highest number of responses possible, and will be most feasible for the Census Bureau to carry out.
Beginning in the early 1990s, the Bureau began planning for the 2000 Census and included the concept of a second mailing strategy. According to the original Bureau plan, the procedure for the mail survey during the first phase called for the use of a targeted second mailing, where all households who had not yet responded to the original census questionnaire would receive an additional form to complete and return by mail. The Census Bureau believed that using a `full mail implementation strategy' would increase mail response rates up to 6 percent.
Additionally, the Bureau's intent to use the second mailing procedure was endorsed by recommendations from the National Academy of Sciences (NAS), after Congress asked them to study and provide ways to improve the accuracy and cost-effectiveness of the census. They too supported the use of a second mailing strategy, citing the Bureau's own research which demonstrated that a replacement questionnaire (in a targeted mailing setting) would considerably increase response rates. The mail survey procedures were very cost-effective because the increase in response rates would provide savings in follow-up costs. Conducting a second mailing could save up to $150 million by reducing field data collection costs. These savings also outweighed the additional expenditures on
mail contacts, both direct and indirect costs such as postage and de-duplication of census forms from households who return two questionnaires. 1
[Footnote 1: National Research Council, Commission on Behavioral and Social Sciences and Education, letter to Dr. Martha Farnsworth Riche, November 10, 1997.]
After consultation with contracting vendors, the Bureau concluded that a second targeted mailing could not be conducted due to technological and time constraints. Rather, a second blanket mailing (a replacement questionnaire to all households, regardless of whether or not they have returned the original questionnaire) would be more feasible. The NAS agreed again that the benefit of increased mail response outweighs the cost of printing sufficient replacement forms and mailing one to each household.
In the 1998 Dress Rehearsals for the 2000 decennial census, the second blanket mailing strategy was tested at two locations, and met all of the standards for increasing response rates. According to the Census 2000 Dress Rehearsal Report Card, in Columbia, SC, the response rate increased 8.6 percent for the short form, and 6.2 percent for the long form. Likewise, at the Sacramento, CA site, the response rate increased 7.9 percent for the short form, and 5.2 percent for the long form. In May of 1998, the Bureau decided to eliminate the second mailing due to difficulties in handling the de-duplication of questionnaires. 2
[Footnote 2: Bureau of the Census, Census 2000 Decision Memorandum No. 54, May 11, 1998.]
In July of 1998, the General Accounting Office (GAO) testified before the Senate Committee on Governmental Affairs concerning the results of the second mailing from testing in the 1998 Dress Rehearsals. The GAO testified that, ` * * * the Bureau does not currently plan to use in 2000 a key ingredient of the response rate achieved during the dress rehearsal--a second mailing. According to a Bureau official, concerns about public confusion have contributed to the Bureau's decision not to use a second questionnaire mailing in 2000. The preliminary results of the dress rehearsal suggest that the Bureau may need to reconsider its decision. At both the South Carolina and Sacramento sites, the Bureau obtained approximately a 7 percentage point `bump' in response rates by sending a second questionnaire to all households located in mailout/mailback areas. According to a senior Bureau official, this 7 percentage point increase represents real additions to the court and does not include duplicate submissions from households that already had responded.' 3
[Footnote 3: Committee on Governmental Affairs, United States Senate, `Decennial Census: Preliminary Observations on the Results to Date of the Dress Rehearsal and the Census Bureau's Readiness for 2000,' July 30, 1998 (GAO/T-GGD-98-178).]
The Congress is concerned with obtaining the most accurate census possible. H.R. 928, The 2000 Census Mail Outreach Improvement Act, reestablishes the second mailing to achieve this goal. To best facilitate this endeavor, H.R. 928 grants authority to the Secretary of Commerce to choose which method, either targeted or blanket mailing, shall be used in the 2000 Census.
H.R. 928 was introduced on March 2, 1999 by the Honorable Dan Miller (R-FL). The bill was referred to the Committee on Government Reform on March 2, 1999, and it was referred to the Subcommittee on the Census March 10, 1999. The Subcommittee held a mark-up on March 11, 1999. No amendments were offered, and the measure was ordered favorably reported to the full Committee by a vote of 6-4.
The Committee did not hold any specific legislative or oversight hearings on H.R. 928. However, during an oversight hearing entitled `Oversight of the 2000 Census: Examining the America Counts Today (ACT) Initiatives To Enhance Traditional Enumeration Methods,' held on March 2, 1999, the issue addressed by this legislation was mentioned.
Section 1. Short Title. This Act may be cited as the `2000 Census Mail Outreach Improvement Act.'
Section 2. This section requires the Secretary of Commerce to ensure that a second mailing will occur after the original mailing of census questionnaires in the taking of the 2000 decennial census. The replacement questionnaire will be sent either to all households included in the original mailing, or to each household who does not respond to the original mailing. Whichever method is more effective in securing the return of a completed questionnaire from the greatest number of households will be left to the judgement of the Secretary. In making this decision, the Secretary of Commerce shall take into consideration factors such as the need for timely and accurate information, cost, and ease of administration.
Pursuant to rule XIII, clause 3(c)(1) of the Rules of the House of Representatives, under the authority of rule X, clause 2(b)(1) and clause 3(e), the results and findings from committee oversight activities are incorporated in the bill and this report.
The budget analysis and projections required by section 308(a) of the Congressional Budget Act of 1974 are contained in the estimate of the Congressional Budget Office.
Hon. DAN BURTON,
Chairman, Committee on Government Reform,
U.S. House of Representatives, Washington, DC.
DEAR MR. CHAIRMAN: The Congressional Budget Office has prepared the enclosed cost estimate for H.R. 928, the 2000 Census Mail Outreach Improvement Act.
If you wish further details on this estimate, we will be pleased to provide them. The CBO staff contact is Mark Hadley.
Barry B. Anderson
(For Dan L. Crippen).
H.R. 928--2000 Census Mail Outreach Improvement Act
In conducting the decennial census in 2000, the Bureau of the Census will mail questionnaires to nearly 95 million households. H.R. 928 would require the bureau to mail a replacement questionnaire either to each household that does not respond to the original mailing or to all households. The bill would give the Secretary of Commerce the authority to choose whichever option he believes would result in the greater response rate.
Based on information from the Bureau of the Census and subject to appropriation of the necessary amount, CBO estimates that implementing H.R. 928 would cost either $110 million--if replacement questionnaires are mailed just to nonrespondents--and $275 million--if replacement questionnaires are mailed to all households. Based on the response rate for the 1990 census and information from the bureau, CBO estimates that around 40 percent of all households will not respond to the original questionnaire. The bureau expects to spend about $160 million on postage for the original questionnaires and estimates that printing a second questionnaire for all households would cost $75 million. Based on the experience of the dress rehearsal of the 2000 census, CBO estimates that the bureau would spend $40 million to process questionnaires from a second mailing to all households. CBO cannot predict which of the two options the Secretary of Commerce would choose for replacement questionnaires. Sending a second mailing to all households could increase the overall response rate more than a targeted mailing because the bureau could reach nonrespondents more quickly. However, it is unclear whether the difference in the overall response rate would be significant.
In addition to the costs cited above, H.R. 928 could affect spending by the Bureau of the Census in two other ways, but CBO cannot estimate their effects. First, the bureau would likely incur additional costs to process more duplicate questionnaires if a complete second mailing is done. During the dress rehearsal, the bureau found that more than one-third of all second questionnaries that were returned were duplicates of original questionnaires. Second, because H.R. 928 could increase the rate of response by mail, it could reduce the costs for having temporary employees telephoning or visiting households that did not respond to the questionnaire.
H.R. 928 would not affect direct spending or receipts; therefore, pay-as-you-go procedures would not apply. H.R. 928 contains no intergovernmental or private-sector mandates as defined in the Unfunded Mandates Reform Act and would impose no costs on state, local, or tribal governments.
The CBO staff contact is Mark Hadley. This estimate was approved by Paul N. Van de Water, Assistant Director for Budget Analysis.
Clauses 1 and 18 of Article 1, Sec. 8, and Article 1, Sec. 2 of the Constitution grant Congress the power to enact this law.
On March 17, 1999, a quorum being present, the Committee on Government Reform ordered the bill favorably reported.
Date: March 17, 1999.
Summary: Final Passage of H.R. 928.
Offered by: Hon. Dan Miller.
Approved by Record Vote, 23 Ayes to 20 Nays.
Vote by Members: Mr. Burton--Aye; Mr. Gilman--Aye; Mrs. Morella--Aye; Mr. Shays--Aye; Ms. Ros-Lehtinen--Aye; Mr. McHugh--Aye; Mr. Horn--Aye; Mr. Mica--Aye; Mr. Davis of Virginia--Aye; Mr. McIntosh--Aye; Mr. Souder--Aye; Mr. Scarborough--Aye; Mr. LaTourette--Aye; Mr. Sanford--Aye; Mr. Barr--Aye; Mr. Miller--Aye; Mr. Hutchinson--Not Voting; Mr. Terry--Aye; Mrs. Biggert--Aye; Mr. Walden--Aye; Mr. Ose--Aye; Mr. Ryan--Aye; Mr. Doolittle--Aye; Mrs. Chenoweth--Aye; Mr. Waxman--Nay; Mr. Wise--Nay; Mr. Owens--Nay; Mr. Towns--Nay; Mr. Kanjorski--Nay; Mrs. Mink--Nay; Mr. Sanders--Nay; Mrs. Maloney--Nay; Ms. Norton--Nay; Mr. Fattah; Mr. Cummings--Nay; Mr. Kucinich--Nay; Mr. Blagojevich--Nay; Mr. Davis of Illinois--Nay; Mr. Tierney--Nay; Mr. Turner--Nay; Mr. Allen--Nay; Mr. Ford--Nay; Ms. Schakowsky--Nay.
H.R. 928 requires the Secretary of Commerce to include either a general or targeted followup mailing of census questionnaires in taking the 2000 Decennial Census. The Act does not apply to the House of Representatives or to the Senate, thus H.R. 928 does not apply to Congress.
The Committee finds that the legislation does not impose any Federal mandates within the meaning of section 423 of the Unfunded Mandates Reform Act (PL 1044).
The Committee finds that the legislation does not establish or authorize establishment of an advisory committee within the definition of 5 U.S.C. App., Section 5(b).
H.R. 928, which requires either a blanket or targeted second mailing of the census questionnaire, should not be supported. To get a fair and accurate census, we must allow the professionals at the Census Bureau to make the many critical decisions involved in taking a census, based on their expertise and experience. Those Census Bureau professionals have told us that conducting a second mailing is a bad idea.
Current law requires that the census start on April 1, 2000, that state population totals be reported to the President and Congress by December 31, 2000, and that final population counts be reported to the states by April 1, 2001. During that time, census takers must assign 275 million people to 120 million addresses on a fixed, single date, April 1, 2000--a task of mind boggling scope and complexity.
H.R. 928 compromises the ability of the Census Bureau to meet these deadlines and would have the effect of blocking the use of modern statistical methods in determining the final totals from the 2000 Census. The statutory deadline established by 13 U.S.C. for the release of those totals is April 1, 2001. Any second mailing would add a minimum of six weeks before nonresponse follow-up could begin, making it impossible for the Census Bureau to use modern statistical methods and meet the April 1, 2001, deadline. The Census Bureau would be forced to release less accurate numbers on April 1, 2001, which were not corrected for overcounts and undercounts.
The Census Bureau tested a blanket second mailing in a dress rehearsal and it didn't work. About 40% of the `second' forms returned during the dress rehearsal were duplicates. If that rate were repeated at the national level in 2000, there would be over 11 million duplicates. As Census Bureau Director Dr. Kenneth Prewitt explained, `For Census 2000, a work load of this magnitude would significantly delay data processing operations and potentially introduce significant errors into the data. In addition, our dress rehearsal experience indicated that the public was confused by the second mailing.'1
[Footnote] In addition, a National Academy of Sciences panel advises that a blanket second mailing could reduce the accuracy of the census.2
[Footnote] 1Kenneth Prewitt, Director of the Census Bureau, Memorandum for the Secretary, 3 (March 16, 1999).
[Footnote] 2National Research Council, Committee on National Statistics Panel on Alternative Census Methodologies, Measuring a Changing Nation, Modern Methods for the 2000 Census, 22 (February 1999).
There is also fault with a targeted second mailing. Dr. Prewitt stated, `printing vendors informed us they would require at least a month to send a second mailing targeted only to nonresponding housing units. A targeted second mailing would, thus, have significantly delayed the start of the nonresponse follow-up operation. Our experience and research indicate that the longer the delay between Census Day and the start of nonresponse follow-up, the more inaccuracies are introduced to the census data.'3
[Footnote] 3Memorandum for the Secretary, supra note 1, 3.
The value of a second mailing is substantially outweighed by the risks that it introduces in other census operations and the delays it would cause in achieving the most accurate census totals.
Henry A. Waxman.
R.E. Wise, Jr.
Major R. Owens.
Paul E. Kanjorski.
Patsy T. Mink.
Carolyn B. Maloney.
Eleanor H. Norton.
Elijah E. Cummings.
Dennis J. Kucinich.
Rod R. Blagojevich.
Danny K. Davis.
John F. Tierney.
Harold E. Ford, Jr.