PeopleBusinessGeographyNewsroomSubjects A to Z Search@Census

Public Information Office                                   CB01-CS.06
301-457-3691/301-457-3620 (fax)
301-457-1037 (TDD)
e-mail: 2000usa@census.gov

Statement by William G. Barron Jr. on the Current Status of Results of
      Census 2000 Accuracy and Coverage Evaluation Survey

  Recently, there have been a number of documents circulated or published
that might cause data users to understand or conclude that the statistical
adjustment based on the Census 2000 Accuracy and Coverage Evaluation
(A.C.E.) would add 6.4 million missed individuals to the results of Census
2000.  This conclusion is incorrect.

  The Census Bureau does have initial estimates from the Census 2000
A.C.E. that indicate there were 4.3 million persons undercounted by Census
2000 and 1.0 million persons overcounted, for a net measured undercount of
3.3 million persons.  These estimates which are still preliminary, and
which are being examined this summer in light of the serious concerns
identified by the Executive Steering Committee on Adjustment Policy (see
the report of the E.S.C.A.P. dated March 1, 2001)  are based on a
methodology known as Dual System Estimation that the Census Bureau has
designed and extensively tested.  This methodology is designed to adjust
for net coverage errors in the census, and used in that fashion, it is a
very robust estimator of net undercounts and overcounts.  Further, by using 
data obtained from the matching operations of the Census 2000 A.C. E. and
applying several simplifying assumptions, we can obtain estimates of the
erroneous enumerations.  These estimates of erroneous enumerations used in
conjunction with the measures of net coverage can be used to estimate
gross omissions, but both estimates are highly dependent upon the
assumptions made.  The Census Bureau's 6.4 million gross omissions (missed
persons) estimate and the corresponding 3.1 million erroneous enumerations
estimate provide general information about the quality of the census and
the quality of the coverage measurement. Unfortunately, the 6.4 million
estimate for gross omissions has been misinterpreted as appropriate to
adjust Census 2000 calculations.  In fact, the estimate for gross
omissions will not be used by the Census Bureau in the adjustment
calculations for Census 2000.

  To summarize, it is the net coverage estimates based on the dual system
methodology that are being evaluated for possible use in adjusting the
Census 2000 data. To clarify this matter, we are reissuing three tables
based on data already available on the U.S. Census Bureau's Web site, that
document the preliminary net coverage adjustments, estimated by race and
ethnicity. Note that these estimates are subject to revision potentially
substantial as the Census Bureau continues its work to determine if
intercensal population estimates and the population controls for household
surveys to be used in the next decade would be improved by statistical
adjustment.

  Table 1 shows the percent net undercount for major groups calculated in
the Census 2000 A.C.E. and the 1990 Post Enumeration Survey (PES).  Table
2 provides the distribution of net undercount estimates for the Census
2000 A.C.E. and the 1990 PES.  Table 3 shows the distribution of the
undercount and the overcount for the Census 2000 A.C.E.

Table 1: Percent Net Undercount for Major Groups: 2000 A.C.E. and 1990 PES

1. 2000 net undercount is for household population.

1. 1990 net undercount is for the PES universe which included noninstitutional Group Quarters in addition to the household population. As a result, the 1990 estimates may differ from the Committee on Adjustment of Postcensal Estimates (CAPE) results. See Bryant et al. (1992) and Thompson (1992).
2. The race and Hispanic categories shown on this table represent estimation groupings used in developing estimates based on the A.C.E. Survey and do not conform with race and Hispanic categories that appear in the redistricting (P.L. 94-171) files and other Census 2000 data products. In developing the estimation groupings used to evaluate the coverage of Census 2000, the principal consideration was to combine people who were expected to have the same probability of being counted in Census 2000. Consequently, the race and Hispanic origin groupings used to create the A.C.E. estimates of coverage are exceedingly complex. For a complete description of the estimation groups, see DSSD Memorandum Q-48, which will be provided on request.
3. In general, American Indians and Alaska Natives (AIAN) are included in that category, regardless of whether they marked another race or are Hispanic. A few exceptions apply, especially for those who do not live on a reservation, on trust lands, or in an AIAN statistical area.
4. Similarly, Native Hawaiians and Other Pacific Islanders (NHPI) generally are included in that category, unless they lived outside of Hawaii and marked more than one race or marked Hispanic.
5. Hispanics are mostly in that category, unless they marked AIAN and lived on reservation, on trust lands, or in an AIAN statistical area, or marked NHPI and lived in Hawaii.
6. People who marked Black or African American are generally in that category unless they fell in the categories described above; similarly those who marked Asian are generally in that category, unless they fell in categories described above.
7. The final category includes most people who marked only White or only Some Other Race or marked three or more races but did not fall into the categories described above.
8. The data in this table contain sampling and non-sampling error; a minus sign denotes a net overcount.

Table 2: Distribution of Net Undercount Estimates for the 2000 A.C.E. and the 1990 PES

1. Net undercount estimates are in thousands. Because of rounding, the distribution percentages may not add to 100.
2. 2000 net undercount is for household population.

1. 1990 net undercount is for the PES universe which included noninstitutional Group Quarters in addition to the household population. As a result, the 1990 estimates may differ from the Committee on Adjustment of Postcensal Estimates (CAPE) results. See Bryant et al. (1992) and Thompson (1992).
2. The race and Hispanic categories shown on this table represent estimation groupings used in developing estimates based on the A.C.E. Survey and do not conform with race and Hispanic categories that appear in the redistricting (P.L. 94-171) files and other Census 2000 data products. In developing the estimation groupings used to evaluate the coverage of Census 2000, the principal consideration was to combine people who were expected to have the same probability of being counted in Census 2000. Consequently, the race and Hispanic origin groupings used to create the A.C.E. estimates of coverage are exceedingly complex. For a complete description of the estimation groups, see DSSD Memorandum Q-48, which will be provided on request.
3. In general, American Indians and Alaska Natives (AIAN) are included in that category, regardless of whether they marked another race or are Hispanic. A few exceptions apply, especially for those who do not live on a reservation, on trust lands, or in an AIAN statistical area.
4. Similarly, Native Hawaiians and Other Pacific Islanders (NHPI) generally are included in that category, unless they lived outside of Hawaii and marked more than one race or marked Hispanic.
5. Hispanics are mostly in that category, unless they marked AIAN and lived on reservation, on trust lands, or in an AIAN statistical area, or marked NHPI and lived in Hawaii.
6. People who marked Black or African American are generally in that category unless they fell in the categories described above; similarly those who marked Asian are generally in that category, unless they fell in categories described above.
7. The final category includes most people who marked only White or only Some Other Race or marked three or more races but did not fall into the categories described above.
8. The data in this table contain sampling and non-sampling error; a minus sign denotes a net overcount.

Table 3: Distribution of Undercount and Overcount for the Census 2000 A.C.E.

1. Estimates of total persons are in thousands. Because of rounding, the distribution percentages may not add to 100.
2. The race and Hispanic categories shown on this table represent estimation groupings used in developing estimates based on the A.C.E. Survey and do not conform with race and Hispanic categories that appear in the redistricting (P.L. 94-171) files and other Census 2000 data products. In developing the estimation groupings used to evaluate the coverage of Census 2000, the principal consideration was to combine people who were expected to have the same probability of being counted in Census 2000. Consequently, the race and Hispanic origin groupings used to create the A.C.E. estimates of coverage are exceedingly complex. For a complete description of the estimation groups, see DSSD Memorandum Q-48, which will be provided on request.
3. In general, American Indians and Alaska Natives (AIAN) are included in that category, regardless of whether they marked another race or are Hispanic. A few exceptions apply, especially for those who do not live on a reservation, on trust lands, or in an AIAN statistical area.
4. Similarly, Native Hawaiians and Other Pacific Islanders (NHPI) generally are included in that category, unless they lived outside of Hawaii and marked more than one race or marked Hispanic.
5. Hispanics are mostly in that category, unless they marked AIAN and lived on reservation, on trust lands, or in an AIAN statistical area, or marked NHPI and lived in Hawaii.
6. People who marked Black or African American are generally in that category unless they fell in the categories described above; similarly those who marked Asian are generally in that category, unless they fell in categories described above.
7. The final category includes most people who marked only White or only Some Other Race or marked three or more races but did not fall into the categories described above.
8. The data in this table contain sampling and non-sampling error; a minus sign denotes a net overcount.

Data ToolsCatalogCensus 2000QualityPrivacy PolicyContact UsHome