oversight

2000 Census: Local Address Review Program Has Had Mixed Results to Date

Published by the Government Accountability Office on 1999-09-29.

Below is a raw (and likely hideous) rendition of the original report. (PDF)

                          United States General Accounting Office

GAO                       Testimony
                          Before the Subcommittee on the Census
                          Committee on Government Reform
                          House of Representatives


For Release on Delivery
10:00 a.m. EDT
Wednesday,                2000 CENSUS
September 29, 1999



                          Local Address Review
                          Program Has Had Mixed
                          Results to Date
                          Statement of
                          J. Christopher Mihm
                          Associate Director, Federal Management
                           and Workforce Issues
                          General Government Division




GAO/T-GGD-99-184
Statement

2000 Census: Local Address Review Program
Has Had Mixed Results to Date

              Mr. Chairman and Members of the Subcommittee:

              I am pleased to be here today to discuss the Census Bureau’s
              implementation of the Local Update of Census Addresses (LUCA)
              program. As you are aware, in 1994 Congress required the Bureau to
              develop a local address review program in order to give local and tribal
              governments greater input into the Bureau’s address list development
                        1
              process. LUCA gives local and tribal governments the opportunity to
              review the accuracy and completeness of the Bureau’s address information
              for their respective jurisdictions, and suggest corrections where
              warranted.

              The success of LUCA is important because a high-quality census begins
              with an accurate address list and precise maps. Together, they help ensure
              that (1) questionnaires are properly delivered; (2) unnecessary and costly
              follow-up efforts at vacant or nonexistent residences are reduced; and (3)
              the correct portions of the population are counted in their proper
              locations, which is the basis of congressional reapportionment and
              redistricting.

              LUCA’s overall impact on the accuracy and completeness of the address
              list will not be fully known until after the census, when the Bureau will
              have completed additional verification procedures and can evaluate the
              accuracy and completeness of the address list. Therefore, as agreed with
              the Subcommittee, my statement today focuses on two initial measures of
              how well LUCA is working: (1) the Bureau’s operational experience to date
              in implementing LUCA, and (2) local governments’ views of the adequacy
              of local resources to conduct LUCA and of the quality of materials and
              assistance the Bureau has provided.

              Our views are based on our examination of those jurisdictions with city-
              style address areas; that is, jurisdictions where the U.S. Postal Service uses
              house-number and street-name addresses for most mail delivery. Because
              this program was initiated in 1998, the Bureau refers to it as “LUCA 1998.“
              According to the Bureau, about 80 percent of the nation’s housing units are
              located within city-style jurisdictions.

              We obtained information on the Bureau’s implementation of LUCA by
              interviewing Bureau officials and analyzing Bureau data on local
              governments’ participation in the program and the number of address list
              changes they suggested. We obtained local governments’ views of LUCA by
              1
                  Census Address List Improvement Act, P.L. 103-430, Oct. 31, 1994.




              Page 1                                                                  GAO/T-GGD-99-184 2000 Census
             Statement
             2000 Census: Local Address Review Program Has Had Mixed Results to Date




             surveying a stratified random sample of 150 city-style jurisdictions of
             varying sizes. Of these, 128 jurisdictions responded, yielding a response
             rate of 85 percent. Our survey results are generalizable to the 8,248 local
             governments that were sent LUCA 1998 materials and were participating in
             LUCA at the time our sample was drawn in January 1999. It is important to
             keep in mind that the data we present on the perceptions of local
             governments are estimates that are based on the results of our survey. We
             describe our survey methodology more fully in attachment I.

             In brief, the Bureau invited nearly 16,675 local governments to participate
             in LUCA 1998. Of these, 6,673 (40 percent) returned material to the Bureau
             as of August 1999. The remaining jurisdictions did not provide the Bureau
             with any input. The Bureau does not know whether these jurisdictions (1)
             did not review the census address list, or (2) did review the list and
             decided not to respond to the Bureau. Local governments suggested about
             7.74 million changes to the Bureau’s address list, of which about 5.4 million
             were suggested additions. The results of our survey suggest that many
             local governments appeared to be satisfied with the quality of the materials
             and assistance the Bureau provided. However, a number of jurisdictions
             rated these items, and the availability of local resources to review LUCA
             materials, less favorably.

             To develop the address list for the 2000 Census, the Bureau is using a
Background   series of operations, including LUCA. LUCA is one of only two components
             of the Bureau’s address list development program that gives local
             governments direct input into the Bureau’s address database.

             For LUCA 1998, the Bureau sent participating governments address list
             information and associated maps for their jurisdictions. Local governments
             had the option of reviewing either paper address lists or electronic address
             lists. Maps were available only in paper format. The local governments
             were to review this information for accuracy and completeness, and
             suggest additions, deletions, corrections, and other changes, where
             appropriate, to the Bureau. Bureau personnel were then to verify these
             changes—generally by visiting each address in question—and provide
             feedback to LUCA participants on the Bureau’s actions.

             Should local governments disagree with the Bureau’s final decisions, they
             can appeal the Bureau’s actions to an independent office established by
             the Office of Management and Budget. Appeals officers—who can be
             detailed or temporary federal employees, as well as contractors that have
             received special training—are to base their decisions on the validity of the
             map or address references supplied by the Bureau and local governments.



             Page 2                                                  GAO/T-GGD-99-184 2000 Census
                              Statement
                              2000 Census: Local Address Review Program Has Had Mixed Results to Date




                              The appeals process for LUCA 1998 has not yet begun. All appeals are to
                              be resolved no later than January 14, 2000.

                              LUCA was tested during the dress rehearsal for the 2000 Census that was
                              conducted in Sacramento, CA; Columbia, SC, and 11 surrounding counties;
                              and Menominee County in Wisconsin, including the Menominee American
                              Indian Reservation. In our testimony on the status of the dress rehearsal
                              before the Subcommittee in March 1998, we reported that LUCA
                              encountered various implementation problems that adversely affected
                                                                 2
                              local governments’ review efforts. They included problems with the
                              accuracy and completeness of the Bureau’s address list and maps, as well
                              as with the level of Bureau assistance. Following the dress rehearsal,
                              Bureau officials said the Bureau took steps to address some of these
                              shortcomings in time for LUCA 1998.

                              Although the Bureau invited all 16,675 jurisdictions with city-style
The Bureau’s                  addresses to participate in LUCA 1998, as shown in table 1, most did not
Operational                   provide the Bureau with any input. Indeed, according to Bureau data, of
Experience in                 the 16,675 jurisdictions eligible for LUCA 1998, 9,796 (about 59 percent)
                              volunteered to participate in the program and signed the Bureau’s
Implementing LUCA             confidentiality agreement, which is a prerequisite for reviewing the
1998                          Bureau’s address lists. However, 1,017 initial participants decided to drop
                              out of the program, and the Bureau ultimately mailed LUCA material to
                              8,779 local governments. Of these, 5,791 reviewed and annotated at least a
                              portion of the address material, and 882 returned unannotated material.
                              Thus, of the original universe of 16,675 jurisdictions, 6,673 (40 percent)
                              returned something to the Bureau.

Table 1: Local Governments’
Participation in LUCA 1998    Eligible jurisdictions                                                                    16,675
                              Jurisdictions that returned confidentiality agreements                                     9,796
                              Jurisdictions that were shipped materials                                                  8,779
                              Jurisdictions that returned annotated materials                                            5,791
                              Jurisdictions that returned unannotated materials                                            882
                              Source: U.S. Census Bureau.


                              Bureau officials said that they were generally pleased with LUCA 1998
                              participation rates, in part because they exceeded figures from a similar
                              program during the 1990 Census. Still, the Bureau cannot at this time
                              determine the extent to which LUCA 1998 has contributed to the accuracy
                              of the Bureau’s address list. This is because important data, such as the

                              2
                              Decennial Census: Preparations for Dress Rehearsal Underscore the Challenges for 2000 (GAO/T-
                              GGD-98-84, Mar. 26, 1998).




                              Page 3                                                         GAO/T-GGD-99-184 2000 Census
                                      Statement
                                      2000 Census: Local Address Review Program Has Had Mixed Results to Date




                                      number of housing unit addresses they actually reviewed, are unavailable.
                                      Also, 2,106 local governments were shipped LUCA materials but did not
                                      provide the Bureau with any input. The Bureau does not know whether
                                      these jurisdictions (1) did not review the census address list, or (2) did
                                      review the list and decided not to respond to the Bureau.

The Number of Changes                 The LUCA 1998 participants that furnished the Bureau with updated
                                      address list information suggested a total of about 7.74 million changes to
Suggested by Local                    the census address list, according to Bureau data (see table 2).
Governments
Table 2: Address List Changes
Suggested by LUCA 1998 Participants   Type of suggested change                                                   Number
                                      Address additions                                                         5,384,864
                                      Address deletions                                                           427,626
                                      Other (e.g., corrections, geographic changes, etc.)                       1,931,559
                                      Total                                                                     7,744,049
                                      Source: U.S. Census Bureau.


                                      As originally planned, the Bureau was to confirm the validity of these
                                      changes as part of a subsequent address list development operation called
                                      block canvassing, which took place between January and May 1999. In this
                                      operation, temporary Bureau employees were to verify all city-style
                                      addresses by systematically traveling each street in an assigned area,
                                      comparing their on-site inspections with the address list, and thus
                                      identifying incorrect, missing, nonexistent, or duplicate addresses.
                                      Addresses that were not accepted by the Bureau during block canvassing
                                      were to be rechecked during a subsequent on-site inspection operation
                                      called “reconciliation.”

                                      However, according to Bureau officials, the Bureau underestimated the
                                      amount of time it would take to get agreements from local governments to
                                      participate in the program and then to prepare address lists for local
                                      governments’ review. As a result, LUCA 1998 took several months longer
                                      to complete than the Bureau initially expected. In turn, a number of
                                      changes were submitted too late to be included in block canvassing and
                                      instead will be verified during the reconciliation process.

                                      For example, of the approximately 5.4 million suggested additions, the
                                      Bureau determined that about 2.76 million were valid—2.0 million during
                                      block canvassing and another 760,000 from matching LUCA 1998
                                      submissions with block canvassing results. The Bureau has not accepted
                                      about 2.2 million of the suggested address additions. This figure includes
                                      about 400,000 suggested additions checked during block canvassing. It also
                                      includes about 1.8 million additions that were not submitted in time to be



                                      Page 4                                                  GAO/T-GGD-99-184 2000 Census
                           Statement
                           2000 Census: Local Address Review Program Has Had Mixed Results to Date




                           included in block canvassing, and that were not independently found by
                           the Bureau during block canvassing. According to Bureau officials, these
                           2.2 million addresses are to be included in its reconciliation operation.

                           In addition to the Bureau’s experience in implementing LUCA 1998, the
LUCA Participants Had      degree to which local governments were satisfied with the LUCA process
Mixed Perceptions of       is another early indicator of how well the program is working. We
the Program                surveyed LUCA 1998 participants on their views of the (1) availability of
                           local resources to review LUCA materials; (2) adequacy of LUCA time
                           frames; (3) user-friendliness of LUCA address lists and maps, and the
                           media on which they were provided; (4) adequacy of Bureau support; and
                           (5) overall completeness and accuracy of LUCA address lists and maps.
                           While many local governments generally gave favorable reviews to the
                           materials and assistance the Bureau provided to them, a number of
                           jurisdictions held less positive views. Further, the availability of human
                           resources to review LUCA materials appeared to be particularly
                           problematic for many local governments.

Availability of Local      As shown in figure 1, about 44 percent of local governments participating
                           in LUCA rated the human resources in their jurisdiction as not at all
Resources to Review LUCA   sufficient or only sufficient to a small extent. In contrast, only about 23
Material                   percent indicated that their human resources were sufficient to a very
                           great or great extent. About 27 percent indicated that they were sufficient
                           to a moderate extent.

                           On the other hand, our survey suggests that the local staff that were
                           available had sufficient skills to review the LUCA material. Indeed, about
                           63 percent reported that the local staff assigned to LUCA had, to a great or
                           very great extent, the skills needed for LUCA reviews, while 19 percent
                           replied “to a moderate extent.”

                           In terms of the sufficiency of local governments’ technological resources
                           to review addresses, the results were mixed. Just over a third of local
                           governments indicated that, to a great or very great extent, their
                           technological resources were sufficient, while about a quarter reported
                           that their resources were sufficient to a small extent or not at all.




                           Page 5                                                  GAO/T-GGD-99-184 2000 Census
                                        Statement
                                        2000 Census: Local Address Review Program Has Had Mixed Results to Date




Figure 1: Extent to Which Local Governments Had Sufficient Resources to Review LUCA Materials




                                        Note: Percentages in figure have confidence intervals of less than + 12 percent.
                                        Source: GAO survey of LUCA 1998 participants.


                                        The availability of local resources appeared to be a factor in whether or
                                        not local governments fully reviewed the addresses and maps for their
                                        jurisdictions. About 14 percent of the local governments reported
                                        conducting a partial review of the Bureau’s address list for their
                                        jurisdiction, while about 6 percent reported partially reviewing the
                                        Bureau’s maps. Of those local governments that partially reviewed address
                                        lists, most cited limited resources for doing so. About 9 percent reported
                                        that they did not review the address lists, while about 16 percent did not
                                        review the maps.

                                        Local governments that reported partial reviews typically focused their
                                        efforts on those areas that were new or changed since 1990, or had
                                        experienced high growth. Some local governments also limited their


                                        Page 6                                                            GAO/T-GGD-99-184 2000 Census
                                      Statement
                                      2000 Census: Local Address Review Program Has Had Mixed Results to Date




                                      reviews to those areas where the local government’s count of housing
                                      units differed from the Bureau’s housing count, or to areas with multiple-
                                      family dwellings. According to Bureau officials, during training provided to
                                      local governments, the Bureau recommended that local jurisdictions
                                      conduct their reviews along these lines if they could not conduct full
                                      reviews.

                                      Overall, the survey results suggest that LUCA can be a burdensome
                                      process for local governments. Indeed, about 71 percent of respondents
                                      indicated that the LUCA workload was much or somewhat more than they
                                      had expected, while 24 percent said it was about what they expected, and
                                      5 percent reported that the workload was somewhat or much less than
                                      expected.

Adequacy of LUCA Time                 The Bureau gave local governments 3 months to review LUCA material. As
                                      shown in figure 2, about 38 percent of local governments reported that this
Frames                                time frame was adequate to a very great or great extent, while about 25
                                      percent indicated that it was adequate to a small extent or not at all.

Figure 2: Extent to Which Local
Governments Found the Time frame to
Complete Their LUCA Reviews
Adequate




                                      Note: Percentages in figure have confidence intervals of less than + 12 percent.
                                      Source: GAO survey of LUCA 1998 participants.




                                      Page 7                                                            GAO/T-GGD-99-184 2000 Census
                                       Statement
                                       2000 Census: Local Address Review Program Has Had Mixed Results to Date




User-friendliness of LUCA              Many local governments had positive views of the user-friendliness of
                                       LUCA address lists and maps, and the paper or electronic media on which
Media                                  they were provided. As shown in figure 3, roughly half of the local
                                       governments reported that both the Bureau’s address lists and maps were
                                       easy to work with to a great or very great extent. Likewise, most local
                                       governments reported that they encountered few or no problems with
                                       either the electronic or paper media on which the information was
                                       provided, nor the media for returning the information.

Figure 3: Extent to Which Local Governments Found LUCA Media Easy to Work With




                                       Note: Percentages in figure have confidence intervals ranging from + 6 percent to + 18 percent.
                                       Source: GAO survey of LUCA 1998 participants.




                                       Page 8                                                           GAO/T-GGD-99-184 2000 Census
                                     Statement
                                     2000 Census: Local Address Review Program Has Had Mixed Results to Date




Adequacy of Bureau                   To help local governments in their LUCA reviews, the Bureau provided
                                     various forms of support, such as formal training. Help was also available
Support                              through different sources, such as reference manuals and direct Bureau
                                     assistance.

                                     With regard to training, figure 4 shows that about 28 percent of local
                                     governments were satisfied with the extent of LUCA training to a very
                                     great or great extent, while about 13 percent were satisfied to a small
                                     extent or not at all. Local governments gave more favorable reviews to
                                     their ability to schedule LUCA training.

Figure 4: Extent to Which Local
Governments Found LUCA Training to
Be Adequate




                                     Note: Percentages in figure have confidence intervals of less than + 12 percent.
                                     Source: GAO survey of LUCA 1998 participants.


                                     The Bureau made available a variety of resources that local governments
                                     could turn to for help in completing their reviews. The sources of
                                     information ranged from formal workshops and reference manuals, to a
                                     video. As shown in figure 5, it appears that LUCA reference manuals were


                                     Page 9                                                            GAO/T-GGD-99-184 2000 Census
                                         Statement
                                         2000 Census: Local Address Review Program Has Had Mixed Results to Date




                                         used more than any other source of information. Moreover, 45 percent of
                                         localities found the manuals to be of great use. Local governments appear
                                         to have turned to the other sources of information, such as the Bureau’s
                                         Internet web site and e-mail contact with the Bureau far less frequently,
                                         and reported finding them to be less useful.

Figure 5: Local Governments’ Views of the Usefulness of Bureau Assistance




                                         Note: Percentages in figure have confidence intervals of less than + 12 percent.
                                         Source: GAO survey of LUCA 1998 participants.


                                         Overall, as shown in figure 6, about half of the localities appeared to be
                                         satisfied with various aspects of the Bureau’s assistance. For example,
                                         about 48 percent indicated that they were satisfied with the extent of



                                         Page 10                                                           GAO/T-GGD-99-184 2000 Census
                                         Statement
                                         2000 Census: Local Address Review Program Has Had Mixed Results to Date




                                         Bureau assistance to a great or very great extent, while about 52 percent
                                         reported that the Bureau’s responses to their questions met their needs to
                                         a great or very great extent. Local governments gave similar reviews to the
                                         timeliness of the Bureau’s response to their questions.

Figure 6: Extent to Which Local Governments Were Satisfied With the Bureau’s Assistance Overall




                                         Note: Percentages in figure have confidence intervals of less than + 12 percent.
                                         Source: GAO survey of LUCA 1998 participants.


Overall Completeness and                 Local governments gave the accuracy of the Bureau’s address lists and
                                         maps mixed reviews. For example, as can be seen in figure 7, about 43
Accuracy of LUCA Address                 percent indicated they encountered few problems with the accuracy and
Lists and Maps                           completeness of the address lists, while, about 18 percent reported
                                         encountering problems to a very great or great extent. Somewhat less
                                         problematic was the perceived completeness and accuracy of the Bureau’s



                                         Page 11                                                           GAO/T-GGD-99-184 2000 Census
                                        Statement
                                        2000 Census: Local Address Review Program Has Had Mixed Results to Date




                                        maps, where about 50 percent said that they encountered problems to a
                                        small extent or not at all, compared with 16 percent who indicated
                                        encountering problems to a very great or great extent.

Figure 7: Local Governments’ Views of
the Accuracy and Completeness of
Bureau Address Lists and Maps




                                        Note: Percentages in figure have confidence intervals of less than + 12 percent.
                                        Source: GAO survey of LUCA 1998 participants.


                                        In summary, Mr. Chairman, the overall results to date of LUCA 1998 appear
                                        to be mixed. On the one hand, many local governments said they were
                                        satisfied with specific aspects of the materials and assistance the Bureau
                                        provided to them. On the other hand, other components, such as training,
                                        received less favorable reviews. Moreover, LUCA may have stretched the
                                        resources of local governments, and overall, the LUCA 1998 workload was
                                        greater than most local governments had expected. In terms of
                                        participation rates and suggested changes, the full impact that these
                                        indicators had on the overall completeness and accuracy of the Bureau’s
                                        address list will not be known until well after Census Day.




                                        Page 12                                                           GAO/T-GGD-99-184 2000 Census
Statement
2000 Census: Local Address Review Program Has Had Mixed Results to Date




We will continue tracking the LUCA program as part of our overall review
of the 2000 Census, and will keep Congress informed of the results of our
work.

Mr. Chairman, this concludes my prepared statement. I would be pleased
to respond to any questions you or other Members of the Subcommittee
may have.

Contact and Acknowledgement

For further information regarding the testimony, please contact J.
Christopher Mihm at (202) 512-8676. Individuals making key contributions
to this testimony included Tom Beall, Robert Goldenkoff, Marcia
McWreath, Anne Rhodes-Kline, and Lynn Wasielewski.




Page 13                                                 GAO/T-GGD-99-184 2000 Census
Attachment I

Scope and Methodology


               To meet our objective of obtaining information on the Bureau’s experience
               to date in implementing the Local Update of Census Addresses (LUCA)
               program, we interviewed officials in the Bureau’s Directorate for
               Decennial Census, including officials in the Geography Division, and in the
               Bureau’s Regional Census Center in Dallas, TX. We asked these officials
               about the Bureau’s time frames for providing local governments with
               address lists and maps for review, reasons why the Bureau’s schedule was
               revised for completing the program, difficulties encountered, and lessons
               learned from the Bureau’s dress rehearsal. We also asked Geography
               Division officials about lessons they had learned for improving the
               program for future Bureau survey work.

               We also requested and obtained data from Bureau Geography Division
               officials on participation rates in the program, including data on the
               number of eligible entities to participate in the program, the number of
               entities that signed confidentiality agreements to participate in the
               program, and the number that were provided address lists and/or maps for
               review. We also requested and obtained data on the number of entities that
               returned annotated and unannotated address lists and/or maps.

               To meet our objective of obtaining information on local governments’
               LUCA participation experience, we faxed a questionnaire to a stratified,
               random probability sample of 150 local governments that were
               participating in the Census Bureau’s 1998 LUCA city-style address
               operation (LUCA 1998).

               We drew our sample from a database, prepared for us by the Bureau, of
               local governments participating in LUCA 1998. We defined participants as
               those local governments that had signed confidentiality agreements, had
               been shipped at least some of the material needed to perform their review,
               and had not subsequently indicated to the Bureau that they had decided to
               drop out of the review process.

               This database identified 8,248 local governments that were participating in
               LUCA 1998 in January 1999. This constituted the population from which
               we drew our sample. Each responding local government was subsequently
               weighted in the analysis to account statistically for all the members of this
               population, including those that were not selected.

               As shown in table I.1, the population was stratified into eight groups based
               on the number of geocoded residential address records taken from the
               Bureau’s master address file and delivered to the local government for its
               review. Also shown in table I.1 is the sample allocation for each stratum



               Page 14                                           GAO/T-GGD-99-184 2000 Census
                                          Attachment I
                                          Scope and Methodology




                                          and the number of local governments that provided usable, completed
                                          questionnaires.

Table I.1: Sample Allocation and Usable
Returns, by Strata                        Number of address              Number of local   Number of local governments
                                          records                  governments in sample      providing usable returns
                                          500,000 or more                             17                            17
                                          100,000–499,999                             31                            28
                                          50,000–99,999                               10                             7
                                          25,000–49,999                               11                            10
                                          10,000–24,999                               17                            13
                                          5,000–9,999                                 14                            12
                                          1,000–4,999                                 32                            28
                                          0–999                                       18                            13
                                          Total                                      150                           128


                                          The majority of items on the questionnaire were closed-ended, meaning
                                          that, depending on the particular item, respondents could choose one or
                                          more response categories or rate the strength of their perception on a 5-
                                          point extent scale. The remaining items were in an open-ended format; i.e.,
                                          the respondent writes in the answer. We analyzed the open-ended
                                          responses by sorting them into categories based on the content of the
                                          responses.

                                          After designing our questionnaire, we pretested it with local governments
                                          in the Washington, D.C., and Dallas, TX, areas. For each local government
                                          in our sample, we contacted the individual identified on the Bureau’s
                                          database as the local liaison for the LUCA 1998 review. Based on our
                                          conversation with this contact person, we sent our questionnaire to this
                                          individual or a person designated by this individual as being the most
                                          appropriate person to respond to our questionnaire for the local
                                          government. We sent out our questionnaires between February 22, 1999,
                                          and March 23, 1999.

                                          We received usable returns from 85 percent of the total eligible sample.
                                          Although we did not test the validity of the local governments’ responses
                                          or the comments they made, we took several steps to check the quality of
                                          our survey data. Specifically, we (1) reviewed and edited completed
                                          questionnaires, (2) made internal consistency checks on selected items,
                                          and (3) checked the accuracy of data entry on returned questionnaires.

                                          The overall survey results are generalizable to the 8,248 local governments
                                          that were participating in LUCA 1998 as of January 1999. Because we
                                          sampled a portion of local governments, all results are estimates and
                                          subject to some uncertainty or sampling error, as well as nonsampling


                                          Page 15                                          GAO/T-GGD-99-184 2000 Census
Attachment I
Scope and Methodology




error. Depending on the particular analysis being performed, percentages
reported for the entire sample have confidence intervals generally ranging
from + 2 to + 18 at the 95 percent confidence level. In other words, if we
had surveyed all the local governments, we are 95 percent confident that
the result obtained would not differ from our sample estimate, in the most
extreme case, by more than + 18 percent.

Our choice of sample size was adequate to support our objective of
obtaining simple, overall estimates of participating local governments’
views of their LUCA 1998 experience. However, this sample size is
generally not large enough to provide the degree of statistical sensitivity
that would be preferable for engaging in more detailed analyses of
differences between various groupings of local governments or
relationships between responses to two or more questionnaire items.

We conducted our work between September 1998 and September 1999 in
accordance with generally accepted government auditing standards.




Page 16                                           GAO/T-GGD-99-184 2000 Census
Page 17   GAO/T-GGD-99-184 2000 Census
Page 18   GAO/T-GGD-99-184 2000 Census
Page 19   GAO/T-GGD-99-184 2000 Census
Ordering Information

The first copy of each GAO report and testimony is free.
Additional copies are $2 each. Orders should be sent to the
following address, accompanied by a check or money order made
out to the Superintendent of Documents, when necessary. VISA
and MasterCard credit cards are accepted, also. Orders for 100 or
more copies to be mailed to a single address are discounted 25
percent.

Order by mail:

U.S. General Accounting Office
P.O. Box 37050
Washington, DC 20013

or visit:

Room 1100
700 4th St. NW (corner of 4th and G Sts. NW)
U.S. General Accounting Office
Washington, DC

Orders may also be placed by calling (202) 512-6000 or by using
fax number (202) 512-6061, or TDD (202) 512-2537.

Each day, GAO issues a list of newly available reports and
testimony. To receive facsimile copies of the daily list or any list
from the past 30 days, please call (202) 512-6000 using a touch-
tone phone. A recorded menu will provide information on how to
obtain these lists.

For information on how to access GAO reports on the INTERNET,
send e-mail message with “info” in the body to:

info@www.gao.gov

or visit GAO’s World Wide Web Home Page at:

http://www.gao.gov
United States                       Bulk Rate
General Accounting Office      Postage & Fees Paid
Washington, D.C. 20548-0001           GAO
                                Permit No. G100
Official Business
Penalty for Private Use $300

Address Correction Requested




(410508)