This is the accessible text file for GAO report number GAO-11-486T 
entitled 'Davis-Bacon Act: Methodological Expertise Critical for 
Improving Survey Quality' which was released on April 14, 2011. 

This text file was formatted by the U.S. Government Accountability 
Office (GAO) to be accessible to users with visual impairments, as 
part of a longer term project to improve GAO products' accessibility. 
Every attempt has been made to maintain the structural and data 
integrity of the original printed product. Accessibility features, 
such as text descriptions of tables, consecutively numbered footnotes 
placed at the end of the file, and the text of agency comment letters, 
are provided but may not exactly duplicate the presentation or format 
of the printed version. The portable document format (PDF) file is an 
exact electronic replica of the printed version. We welcome your 
feedback. Please E-mail your comments regarding the contents or 
accessibility features of this document to 

This is a work of the U.S. government and is not subject to copyright 
protection in the United States. It may be reproduced and distributed 
in its entirety without further permission from GAO. Because this work 
may contain copyrighted images or other material, permission from the 
copyright holder may be necessary if you wish to reproduce this 
material separately. 

United States Government Accountability Office: 


Before the Subcommittee on Workforce Protections, Committee on 
Education and the Workforce, House of Representatives: 

For Release on Delivery: 
Expected at 10:00 a.m. EDT:
Thursday, April 14, 2011: 

Davis-Bacon Act: 

Methodological Expertise Critical for Improving Survey Quality: 

Statement of Andrew Sherrill, Director:
Education, Workforce, and Income Security: 


Chairman Walberg, Ranking Member Woolsey, and Members of the 

I am pleased to be here today to discuss the Department of Labor's 
(Labor) procedures for determining prevailing wage rates under the 
Davis-Bacon Act.[Footnote 1] Davis-Bacon wages must be paid to workers 
on certain federally funded construction projects, and their 
vulnerability to the use of inaccurate data has long been an issue for 
Congress, employers, and workers. More recently, the passage of the 
American Recovery and Reinvestment Act of 2009,[Footnote 2] focused 
attention on the need for accurate and timely wage determinations, 
with more than $300 billion estimated to provide substantial funding 
for, among other things, federally funded building and infrastructure 
work potentially subject to Davis-Bacon wage rates.[Footnote 3] In the 
1990s, we issued two reports that found process changes were needed to 
increase confidence that wage rates were based on accurate data. 
[Footnote 4] A third report found that changes then planned by Labor, 
if successfully implemented, had the potential to improve the wage 
determination process.[Footnote 5] However, in 2004, Labor's Office of 
Inspector General (OIG) found that wage data errors and the timeliness 
of surveys used to gather wage information from contractors and 
others, continued to be issues.[Footnote 6] 

Today I will discuss (1) the extent to which Labor has addressed 
concerns regarding the quality of the Davis-Bacon wage determination 
process and (2) additional issues identified by stakeholders regarding 
the wage determination process. My remarks are based on our recently 
issued report, titled Davis-Bacon Act: Methodological Changes Needed 
to Improve Wage Survey.[Footnote 7] To evaluate the extent to which 
Labor has addressed concerns regarding the quality of the wage 
determination process, we interviewed Labor officials, reviewed 
relevant federal laws and regulations, and compared agency documents 
on current survey practices with guidance on data quality and survey 
design from the Office of Management and Budget (OMB) and Labor. 
[Footnote 8] We conducted site visits to three of Labor's five 
regional offices that process Davis-Bacon wage surveys, as well as to 
the Construction Industry Research and Policy Center (CIRPC), a Labor 
survey contractor.[Footnote 9] We compared 12 surveys performed under 
Labor's new processes against its revised timelines to assess whether 
the surveys were on schedule. We also analyzed the results of four 
surveys published in 2009 or 2010 and reviewed their verification 
reports prepared by Labor's contracted auditor.[Footnote 10] Further, 
we analyzed currently published wage rates to determine their age and 
the proportion of union-prevailing to nonunion-prevailing rates. 
[Footnote 11] To assess what additional issues were concerns for 
stakeholders, we conducted approximately 30 interviews with 
contractors and representatives from academia, contractor 
associations, and unions, and performed a content analysis of their 
responses. A more detailed explanation of our methodology is available 
in Appendix I of our full report. 

This testimony is based on work performed between September 2009 and 
March 2011 in accordance with generally accepted government auditing 
standards. Those standards require that we plan and perform the audit 
to obtain sufficient, appropriate evidence to provide a reasonable 
basis for our findings and conclusions based on our audit objectives. 
We believe that the evidence obtained provides a reasonable basis for 
our findings and conclusions based on our audit objectives. 

In summary, we found that recent efforts to improve the Davis-Bacon 
wage survey have not yet addressed key issues with survey quality, 
such as the representativeness and sufficiency of survey data 
collected. Labor has made some data collection and processing changes; 
however, we found some surveys initiated under these changes were 
behind Labor's processing schedule. Stakeholders said contractors may 
not participate in the survey because they do not understand its 
purpose or do not believe the resultant prevailing wages are fully 
accurate. In addition, they said addressing a lack of transparency in 
how the published wage rates are set could result in a better 
understanding of the process and greater participation in the survey. 
We suggest Congress consider amending its requirement that Labor issue 
wage rates by civil subdivision to allow more flexibility. To improve 
the quality and timeliness of the Davis-Bacon wage surveys, we 
recommend Labor obtain objective expert advice on its survey design 
and methodology. We also recommend Labor take steps to improve the 
transparency of its wage determinations. 


The Davis-Bacon Act was enacted in 1931, in part, to protect 
communities and workers from the economic disruption caused by 
contractors hiring lower-wage workers from outside their local area, 
thus obtaining federal construction contracts by underbidding 
competitors who pay local wage rates. Labor administers the act 
through its Wage and Hour Division, which conducts voluntary surveys 
of construction contractors and interested third parties[Footnote 12] 
on both federal and nonfederal projects to obtain wages paid to 
workers in each construction job by locality.[Footnote 13] It then 
uses the data submitted on these survey forms to determine locally 
prevailing wage and fringe benefit rates for its four construction 
types: building, heavy, highway, and residential.[Footnote 14] 

To determine a prevailing wage for a specific job classification, 
Labor considers sufficient information to be the receipt of wage data 
on at least three workers from two different employers in its 
designated survey area. Then, in accordance with its regulations, 
Labor calculates the prevailing wage by determining if the same wage 
rate is paid to the majority (more than 50 percent) of workers 
employed in a specific job classification on similar projects in the 
area.[Footnote 15] If the same rate is not paid to the majority of 
workers in a job classification, the prevailing wage is the average 
wage rate weighted by the number of employees for which that rate was 
reported. In cases where the prevailing wage is also a collectively 
bargained, or union, rate, the rate is determined to be "union-
prevailing." To issue a wage determination--a compilation of 
prevailing wage rates for multiple job classifications in a given 
area--Labor must, according to its procedures, also have sufficient 
data to determine prevailing wages for at least 50 percent of key job 
classifications. Key job classifications are those determined 
necessary for one or more of the four construction types.[Footnote 16] 

By statute, Labor must issue wage determinations based on similar 
projects in the "civil subdivision of the state" in which the federal 
work is to be performed.[Footnote 17] Labor's regulations state the 
civil subdivision will be the county, unless there are insufficient 
wage data.[Footnote 18] When data from a county are insufficient to 
issue a wage rate for a job classification, a group of counties is 
created. When data are still insufficient, Labor includes data from 
contiguous counties, combined in "groups" or "supergroups" of 
counties, until sufficient data are available to meet threshold 
guidelines to make a prevailing wage determination. Expansion to 
include other counties, if necessary, may continue until data from all 
counties in the state are combined. Counties are combined based on 
whether they are metropolitan or rural, and cannot be mixed. 

Recent Efforts to Improve Data Collection and Processing Have Not Yet 
Addressed Key Issues with Survey Quality: 

Labor has taken several steps over the last few years to address 
issues with its Davis-Bacon wage surveys. For example, it finished 22 
open surveys that had accumulated since the agency started conducting 
statewide surveys in 2002. Officials said completing these surveys 
will allow them to focus on more recent surveys. Labor also changed 
how it collects and processes information for its four construction 
types by surveying some construction types separately rather than 
simultaneously, using other available sources of wage data, adjusting 
survey time frames, and processing survey data as it is received 
rather than waiting until a survey closes. For highway surveys, Labor 
officials said they began using certified payrolls as the primary data 
source because certified payrolls provide accurate and reliable wage 
information and eliminate the need for Labor to verify wage data 
reported in surveys. Labor officials estimated these changes will 
reduce the processing time for highway surveys by more than 80 
percent, or from about 42 months to 8 months. For building and heavy 
surveys, Labor began a five-survey pilot in 2009, adjusting survey 
time frames--with shorter time frames for areas in which there are 
many active projects--to allow Labor to better manage the quantity of 
data received. In addition, Labor officials said their regional office 
staff have begun processing survey data as they are received rather 
than waiting until a survey closes, which, they said, will improve 
timeliness and accuracy because survey respondents will be better able 
to recall submitted information when contacted by regional office 
staff for clarification and verification. Labor expects these changes 
to reduce the time needed to process building and heavy surveys by 
approximately 54 percent, or from about 37 months to 17 months. 

However, while it is too early to fully assess the effects of Labor's 
2009 actions, our review found that changes to data collection and 
processing may not achieve expected results. We were able to analyze 
the timeliness of 12 of the 16 surveys conducted under Labor's new 
processes at the time of our review.[Footnote 19] Of those 12 surveys--
8 highway and 4 building and heavy--which we assessed against Labor's 
revised timelines, we found 10 behind schedule, 1 on schedule, and 1 
not started as of September 10, 2010.[Footnote 20] A challenge to 
survey timeliness is the fact that Labor conducts a "universe" or 
"census" survey of all active construction projects within a 
designated time frame and geographic area. As a result, the number of 
returned survey forms and the time required for the regional offices 
to process the data can vary widely. For example, for 14 surveys 
conducted prior to Labor's 2009 changes, the number of forms returned 
per survey ranged from less than 2,000 to more than 8,000, and the 
average processing time per survey for data clarification and analysis 
ranged from 10 months to more than 40. Moreover, Labor cannot entirely 
control when it receives survey forms. Some regional office officials 
said the bulk of the forms are returned on the last day of a survey 
limiting officials' ability to gain time by processing forms while the 
survey is ongoing as planned under the 2009 changes. To address these 
challenges, OMB guidance suggests agencies consider the cost and 
benefits of conducting a sample survey (versus a census survey) 
because it can often ensure data quality in a more efficient and 
economical way.[Footnote 21] 

The fact that Labor is behind schedule on surveys begun under the new 
processes may affect its ability to update the many published nonunion-
prevailing wage rates which are several years old. Labor's fiscal year 
2010 performance goal was for 90 percent of published wage rates for 
building, heavy, and highway construction types to be no more than 3 
years old. Our analysis found that 61 percent of published rates for 
these construction types were 3 years old or less.[Footnote 22] 
However, this figure can be somewhat misleading because of the 
difference in how union-and nonunion-prevailing wage rates are 
updated. Union-prevailing rates account for almost two-thirds of the 
more than 650,000 published building, heavy, and highway rates and, 
according to Labor's policy, can be updated when there is a new 
collective bargaining agreement without Labor conducting a new survey. 
We found almost 75 percent of those rates were 3 years old or less. 
However, 36 percent of the nonunion-prevailing wage rates were 3 years 
old or less and almost 46 percent were 10 or more years old. These 
rates are not updated until Labor conducts a new survey. Several of 
the union and contractor association representatives we interviewed 
said the age of the Davis-Bacon nonunion-prevailing rates means they 
often do not reflect actual prevailing wages, which can make it 
difficult for contractors to successfully bid on federal projects. 

Beyond concerns with processes and timelines, we also found that 
critical problems with Labor's wage survey methodology continue to 
hinder its survey quality. OMB guidance states that agencies need to 
consider the potential impact of response rate and nonresponse on the 
quality of information obtained through a survey. A low response rate 
may mean the results are misleading or inaccurate if those who respond 
differ substantially and systematically from those who do not respond. 
However, Labor cannot determine whether its Davis-Bacon survey results 
are representative of prevailing wages because it has not calculated 
survey response rates since 2002, and, other than a second letter 
automatically sent to nonrespondents, does not currently have a 
program to systematically follow up with or analyze nonrespondents. 
While a senior Labor official said the agency is taking steps to again 
calculate response rates, these changes have not been fully 
implemented and it is unclear if they will result in improved survey 

The utility of issuing wage determinations at the county level is also 
questionable. Labor's regulations state the county will normally be 
the civil subdivision for which a prevailing wage is determined; 
[Footnote 23] however, Labor is often unable to issue wage rates for 
job classifications at the county level because it does not collect 
enough data to meet its current sufficiency standard of wage 
information on at least three workers from two employers. In the 
results from the four surveys we reviewed, Labor issued about 11 
percent of wage rates for key job classifications using data from a 
single county (see figure 1).[Footnote 24],[Footnote 25] 

Figure 1: Percentage of Key Job Classification Wage Rates Issued at 
Each Geographic Level, for Four Surveys Reviewed: 

[Refer to PDF for image: pie-chart] 

State: 40%; 
Group: 22%; 
Supergroup: 20%; 
County: 11%; 
Geographic level not available: 7%. 

Source: GAO analysis of Labor data from Florida, Maryland, Tennessee, 
and West Texas Metropolitan surveys published in either 2009 or 2010. 

[End of figure] 

Moreover, in 1997, Labor's OIG reported that issuing rates by county 
may cause wage decisions to be based on an inadequate number of 
responses. In the four surveys we reviewed, more than one-quarter of 
the wage rates were based on data reported for six or fewer workers 
(see figure 2). 

Figure 2: Percentage of Key Job Classification Wage Rates Issued Based 
on Number of Workers, for Four Surveys Reviewed: 

[Refer to PDF for image: pie-chart] 

3 workers: 6%; 
4-6 workers: 20%; 
7-12 workers: 23%; 
13-28 workers: 26%; 
29 or more workers: 25%. 

Source: GAO analysis of Labor data from Florida, Maryland, Tennessee, 
and West Texas Metropolitan surveys published in either 2009 or 2010. 

[End of figure] 

The statutory requirement to issue Davis-Bacon wages based on similar 
projects in the "civil subdivision of the state" limits Labor's 
options to address inadequate data because it cannot use data from 
other sources if those data draw from geographic areas, such as 
metropolitan statistical areas, which are not the same as civil 
subdivisions. Officials from CIRPC, Labor's survey contractor, said 
one way to improve accuracy is to survey areas other than counties 
because counties are arbitrary geographic divisions whereas other 
geographic groupings, such as the economic areas used by the Bureau of 
Economic Analysis, are based on regional markets that frequently cross 
county and state lines.[Footnote 26] Some stakeholders said the focus 
on county-level wage rates results in the publication of illogical 
rates--for example, a contractor paving a road that crossed a county 
line had to pay workers different wage rates based on which side of 
the line they worked. 

Little Incentive to Participate and Lack of Transparency Remain Key 
Issues for Stakeholders: 

In our interviews with stakeholders, concerns about the survey process 
and accuracy of the published wage determinations were cited as 
disincentives to participate. Contractors may lack the necessary 
resources, may not understand the purpose of the survey, or may not 
see the point in responding because they believe the prevailing wages 
issued by Labor are inaccurate, stakeholders told us. Officials we 
interviewed in Labor regional offices echoed many of these same 
concerns about contractor participation. 

While 19 of the 27 contractors and interested parties we interviewed 
said the survey form was generally easy to understand, some identified 
challenges with completing specific sections, such as how to apply the 
correct job classification.[Footnote 27] Labor officials said they did 
not pretest the current survey form with respondents, and our review 
of reports by Labor's contracted auditor for four published surveys 
found most survey forms, which are verified against payroll data, had 
errors in areas such as number of employees and hourly and fringe 
benefit rates.[Footnote 28] Labor officials said they have plans to 
address portions of the form that confuse respondents, but could not 
provide specifics on how they intend to solicit input from 
respondents--a step recommended by OMB to reduce error. 

Fifteen stakeholders we interviewed said there is a lack of 
transparency in wage determinations because key information is not 
available or hard to find. Both contractor associations and union 
officials said improving transparency in how the published wage rates 
are set could enhance understanding of the process and result in 
greater participation in the survey. A senior Labor official said the 
agency is considering posting information used to determine wage rates 

Finally, while the pre-survey briefing is one of Labor's primary 
outreach efforts to inform stakeholders about upcoming surveys, 
awareness of these briefings was mixed. In three states that were 
surveyed for building and heavy construction in 2009 or 2010--Arizona, 
North Carolina, and West Virginia--all the union representatives we 
interviewed said they were aware of the pre-survey briefing and 
representatives from four of the six state contractor associations we 
interviewed said they were aware a briefing had been conducted. 
However, in Florida and New York--last surveyed in 2005 and 2006 
respectively--none of the 12 contractors we interviewed were aware 
that a briefing had been conducted prior to the survey. Seven of 27 
stakeholders indicated that alternative approaches, such as webinars 
or audioconferences, might be helpful ways to reach additional 
contractors.[Footnote 29] 

Conclusions and Recommendations: 

While Labor has made some changes to improve the wage determination 
process, further steps are needed to address longstanding issues with 
the quality of wage determinations and enhance their transparency. In 
our report, we suggested that Congress consider amending its 
requirement that Labor issue wage rates by civil subdivision to 
provide the agency with more flexibility. To improve the quality and 
timeliness of the wage surveys, we recommended that Labor enlist an 
independent statistical organization to evaluate and provide objective 
advice on the survey, including its methods and design; the potential 
for conducting a sample survey instead of a census survey; the 
collection, processing, tracking and analysis of data; and the 
promotion of survey awareness. We also recommended that Labor take 
steps to improve the transparency of its wage determinations, which 
could encourage greater participation in its survey. After reviewing 
the draft report, Labor agreed with our recommendation to improve 
transparency, but said obtaining expert survey advice may be 
premature, given current and planned changes. We believe a time of 
change is exactly when the agency should obtain expert advice to 
ensure their efforts improve the quality of the wage determination 
process. A complete discussion of our recommendations, Labor's 
comments, and our response are provided in our report. 

Chairman Walberg, Ranking Member Woolsey, and Members of the 
Subcommittee, this concludes my prepared remarks. I would be happy to 
answer any questions you may have. 

GAO Contact and Staff Acknowledgments: 

For further information regarding this statement, please contact 
Andrew Sherrill at (202) 512-7215 or Contact points 
for our Office of Congressional Relations and Public Affairs may be 
found on the last page of this statement. Individuals who made key 
contributions to this testimony include Gretta L. Goodwin (Assistant 
Director), Amy Anderson, Brenna Guarneros, Susan Aschoff, Walter 
Vance, Ronald Fecso (Chief Statistician), Melinda Cordero, Mimi 
Nguyen, and Alexander Galuten. 

[End of section] 


[1] Pub. L. No. 71-798, 46 Stat. 1494 (1931), as amended; codified at 
40 U.S.C. § 3141 et. seq. A prevailing wage rate is the wage that 
Labor determines to be prevailing for the corresponding class of 
laborers or mechanics employed on projects of a character similar to 
the contract work in the civil subdivision of the state in which the 
work is to be performed. 

[2] Pub. L. No. 111-5, 123 Stat. 115. 

[3] The Congressional Budget Office estimated in early 2009 that the 
combined spending and tax provisions of the American Recovery and 
Reinvestment Act of 2009 would cost $787 billion from 2009 through 
2019. In April 2009, the Congressional Research Service estimated the 
budget authority for Division A of the act to be more than $300 
billion for the same time period. Division A consists primarily of 
discretionary spending with some exceptions and includes federally 
funded building and infrastructure work. 

[4] See GAO, Davis-Bacon Act: Labor Now Verifies Wage Data, but 
Verification Process Needs Improvement, [hyperlink,] (Washington, D.C.: Jan. 
11, 1999) and Davis-Bacon Act: Process Changes Could Raise Confidence 
That Wage Rates Are Based on Accurate Data, [hyperlink,] (Washington, D.C.: May 
31, 1996). 

[5] See GAO, Davis-Bacon Act: Labor's Actions Have Potential to 
Improve Wage Determinations, [hyperlink,] (Washington, D.C.: May 28, 

[6] Department of Labor, Office of Inspector General, Concerns Persist 
with the Integrity of Davis-Bacon Prevailing Wage Determinations, 04- 
04-003-04-420 (Washington, D.C., Mar. 30, 2004). 

[7] See GAO, Davis-Bacon Act: Methodological Changes Needed to Improve 
Wage Survey, [hyperlink,] 
(Washington, D.C.: Mar. 22, 2011). 

[8] See Office of Management and Budget, Guidelines for Ensuring and 
Maximizing the Quality, Objectivity, Utility, and Integrity of 
Information Disseminated by Federal Agencies (effective date Jan. 3, 
2002). For additional OMB guidance on agency surveys see also Office 
of Management and Budget, Standards and Guidelines for Statistical 
Surveys (September 2006) and Questions and Answers When Designing 
Surveys for Information Collections (January 2006). For Labor 
guidance, see Department of Labor, Guidelines for Ensuring and 
Maximizing the Quality, Objectivity, Utility, and Integrity of 
Information Disseminated by the Department of Labor (Oct. 1, 2002). 

[9] We conducted site visits to the following Labor regional offices: 
Northeast region (Philadelphia, Pennsylvania), Southeast region 
(Atlanta, Georgia), and Southwest region (Dallas, Texas). The other 
two regional offices are the Midwest region (Chicago, Illinois) and 
the West region (San Francisco, California). 

[10] The four surveys were Florida 2005, Maryland 2005, Tennessee 
2006, and West Texas Metropolitan 2006. 

[11] We reviewed the wage rates that were published as of November 12, 

[12] Interested third parties include contractor associations; labor 
unions; federal, state, and local agencies; and members of Congress. 

[13] Labor conducts statewide surveys except in large states, such as 
California and Texas. 

[14] Highway construction includes the construction, alteration, or 
repair of roads, streets, highways, runways, parking areas, and other 
similar projects that are not associated with building or heavy 
construction. Residential construction includes single-family homes 
and apartment buildings that are not more than four stories. If a 
structure that houses people is more than four stories or if it houses 
machinery, equipment, or supplies it is considered building 
construction. Heavy construction generally includes any project that 
does not fall into the other three categories--for example, dam and 
sewer projects. 

[15] 29 C.F.R. § 1.2(a)(1). 

[16] Key job classifications across all four construction types 
include bricklayer, boilermaker, carpenter, cement mason, electrician, 
heat and frost insulator/asbestos worker/pipe insulator, iron worker, 
laborer-common, painter, pipefitter, plumber, power equipment 
operator, roofer, sheet metal worker, tile setter, and truck driver. 

[17] 40 U.S.C. § 3142(b). 

[18] 29 C.F.R. § 1.7(a). 

[19] We were unable to analyze four surveys because of unclear dates 
in Labor's data. 

[20] The highway surveys were Florida 2009, New Mexico 2009, North 
Carolina 2009, Oklahoma 2009, South Carolina 2009, Louisiana 2010, 
Nebraska 2010, and New Hampshire 2010. The building and heavy surveys 
were Montana 2009, North Carolina 2009, West Virginia 2009, and 
Wyoming 2009. 

[21] While a census survey attempts to collect data from the entire 
population, a sample survey collects data from a subset or sample of 
the population. When the sample is selected by a probability sampling 
method such that each member of the population has a known chance of 
being selected and that information is used with proper estimation 
techniques, the results are generalizable to the entire population 
with a known level of confidence in the precision of the estimates. 
Further, by reducing the data collection effort, more can be done to 
assure other aspects of data quality. 

[22] As of November 12, 2010. 

[23] 29 C.F.R. § 1.7(a). 

[24] We analyzed wage rates for key job classifications because wage 
rates for nonkey job classifications can only be issued at the county 
or group level, but not at the supergroup or state level. 

[25] Regional office officials said they may combine rates from 
counties with the exact same wage and fringe benefit data in their 
final wage compilation report, the WD-22. However, the rates being 
combined may have been calculated at different geographic levels--for 
example, one county's rates may have been calculated at the group 
level while another county's rates my have been calculated at the 
supergroup level. The geographic level at which rates for combined 
counties were calculated is not reported on the WD-22; therefore, we 
reported the percentage of these rates separately. 

[26] The Bureau of Economic Analysis is an agency within the 
Department of Commerce. It collects source data, conducts research and 
analysis, develops and implements estimation methodologies, and 
disseminates economic statistics to the public. 

[27] We did not ask the representatives from academia about the form 
because they generally would not be asked to fill it out as a survey 

[28] The four surveys we reviewed--Florida 2005, Maryland 2005, 
Tennessee 2006, and West Texas Metropolitan 2006--were conducted prior 
to new survey processes being implemented. No verification reports for 
surveys conducted under the new processes were available in time for 
our review. 

[29] We did not ask the representatives from academia about pre-survey 
briefings because they would generally not be one of the groups Labor 
would notify about an upcoming survey. 

[End of section] 

GAO's Mission: 

The Government Accountability Office, the audit, evaluation and 
investigative arm of Congress, exists to support Congress in meeting 
its constitutional responsibilities and to help improve the performance 
and accountability of the federal government for the American people. 
GAO examines the use of public funds; evaluates federal programs and 
policies; and provides analyses, recommendations, and other assistance 
to help Congress make informed oversight, policy, and funding 
decisions. GAO's commitment to good government is reflected in its core 
values of accountability, integrity, and reliability. 

Obtaining Copies of GAO Reports and Testimony: 

The fastest and easiest way to obtain copies of GAO documents at no 
cost is through GAO's Web site [hyperlink,]. Each 
weekday, GAO posts newly released reports, testimony, and 
correspondence on its Web site. To have GAO e-mail you a list of newly 
posted products every afternoon, go to [hyperlink,] 
and select "E-mail Updates." 

Order by Phone: 

The price of each GAO publication reflects GAO’s actual cost of
production and distribution and depends on the number of pages in the
publication and whether the publication is printed in color or black and
white. Pricing and ordering information is posted on GAO’s Web site, 

Place orders by calling (202) 512-6000, toll free (866) 801-7077, or
TDD (202) 512-2537. 

Orders may be paid for using American Express, Discover Card,
MasterCard, Visa, check, or money order. Call for additional 

To Report Fraud, Waste, and Abuse in Federal Programs: 


Web site: [hyperlink,]: 
Automated answering system: (800) 424-5454 or (202) 512-7470: 

Congressional Relations: 

Ralph Dawn, Managing Director, 
(202) 512-4400: 
U.S. Government Accountability Office: 
441 G Street NW, Room 7125: 
Washington, D.C. 20548: 

Public Affairs: 

Chuck Young, Managing Director, 
(202) 512-4800: 
U.S. Government Accountability Office: 
441 G Street NW, Room 7149: 
Washington, D.C. 20548: