Equal Employment Opportunity: Data Shortcomings Hinder Assessment of
Conflicts in the Federal Workplace (Letter Report, 05/04/99,
GAO/GGD-99-75).

Pursuant to a congressional request, GAO provided information on the
nature and extent of workplace conflicts that underlie the rising number
of discrimination cases, focusing on: (1) the statutory bases (e.g.,
race, sex, or disability discrimination) under which employees filed
complaints; (2) the kinds of issues (e.g., nonselection for promotion,
harassment) that were cited in these complaints; and (3) why the data
collected and reported by the Equal Employment Opportunity Commission
(EEOC) were not helpful in answering the questions raised.

GAO noted that: (1) relevant and reliable data about the bases for
federal employee discrimination complaints and the specific issues
giving rise to these complaints would help decisionmakers and program
managers understand the nature and extent of conflict in the federal
workplace; (2) these data could also be used to help plan corrective
actions and measure the results of interventions; (3) however, EEOC does
not collect and report data about bases and issues in a way that would
help answer some fundamental questions about the nature and extent of
workplace conflicts, such as: (a) how many individuals filed complaints;
(b) in how many complaints each of the bases for discrimination was
alleged; and (c) the most frequently cited issues in employees'
discrimination complaints and in how many complaints was each of the
issues cited; (4) moreover, the reliability of the data that EEOC
collects from agencies and reports is questionable; (5) GAO found that
agencies reported basis and issue data to EEOC in an inconsistent
manner; (6) GAO also found that agencies did not report to EEOC some of
the data it requested and reported some other data incorrectly; and (7)
in addition, because EEOC did not have procedures that ensured the
reliability of the data it collected from agencies, it published some
unreliable data in its annual Federal Sector Report on Equal Employment
Opportunity Complaints Processing and Appeals.

--------------------------- Indexing Terms -----------------------------

 REPORTNUM:  GGD-99-75
     TITLE:  Equal Employment Opportunity: Data Shortcomings Hinder
	     Assessment of Conflicts in the Federal Workplace
      DATE:  05/04/99
   SUBJECT:  Data integrity
	     Statistical data
	     Postal service employees
	     Federal employees
	     Employment discrimination
	     Data collection
	     Reporting requirements
	     Internal controls
	     Fair employment programs
	     Agency reports

******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO report.  This text was extracted from a PDF file.        **
** Delineations within the text indicating chapter titles,      **
** headings, and bullets have not been preserved, and in some   **
** cases heading text has been incorrectly merged into          **
** body text in the adjacent column.  Graphic images have       **
** not been reproduced, but figure captions are included.       **
** Tables are included, but column deliniations have not been   **
** preserved.                                                   **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
** A printed copy of this report may be obtained from the GAO   **
** Document Distribution Center.  For further details, please   **
** send an e-mail message to:                                   **
**                                                              **
**                                            **
**                                                              **
** with the message 'info' in the body.                         **
******************************************************************

EQUAL EMPLOYMENT OPPORTUNITY 
Data Shortcomings Hinder Assessment of
Conflicts in the Federal Workplace

United States General Accounting OfficeGAO Report to Congressional
Requesters May 1999

GAO/GGD-99-75

United StatesGeneral Accounting Office Washington, D.C.  20548

General Government Division

B-282387

Page 1 GAO/GGD-99-75 EEO Complaint Data Shortcomings

GAO

May 4, 1999 The Honorable Elijah E. CummingsRanking Minority
Member Subcommittee on Civil ServiceCommittee on Government Reform
House of Representatives The Honorable Albert R. WynnHouse of
Representatives

In July 1998, we reported to you that the number of equal
employmentopportunity (EEO) complaints alleging discrimination
filed by federal employees had steadily increased, from 17,696 in
fiscal year 1991 to 28,947in fiscal year 1997.

1 In order to answer certain fundamental questions

about the nature and extent of the workplace conflicts that
underlie therising number of discrimination cases, you requested
that we develop

information about (1) the statutory bases (e.g., race, sex, or
disabilitydiscrimination) under which employees filed complaints
and (2) the kinds of issues (e.g., nonselection for promotion,
harassment) that were cited inthese complaints. In our attempt to
respond to your request, we analyzed data collected and reported
by the Equal Employment OpportunityCommission (EEOC) on the bases
for complaints filed with federal agencies, including the U.S.
Postal Service, and the issues raised in thesecomplaints.

2 On March 2, 1999, we informed you that these data would not

answer the fundamental questions you had asked about the nature
andextent of conflict in the federal workplace. You then asked for
our

assessment as to why the data collected and reported by EEOC were
nothelpful in answering these questions. This report discusses our
assessment. Relevant and reliable data about the bases for federal
employeediscrimination complaints and the specific issues giving
rise to these complaints would help decisionmakers and program
managers understandthe nature and extent of conflict in the
federal workplace. These data could also be used to help plan
corrective actions and measure the results
1 Equal Employment Opportunity: Rising Trends in Complaint
Caseloads in the Federal Sector

(GAO/GGD-98-157BR, July 24, 1998). The figure for fiscal year 1997
is revised from the preliminaryfigure contained in the report.

2 We refer to the U.S. Postal Service as a federal agency, even
though it is an independent governmental establishment, because it
is bound by most of the same discrimination complaint processes
that applyto most federal agencies.

Results in Brief

B-282387

Page 2 GAO/GGD-99-75 EEO Complaint Data Shortcomings of
interventions. However, EEOC does not collect and report data
aboutbases and issues in a way that would help answer some
fundamental questions about the nature and extent of workplace
conflicts, such as:

*  How many individuals filed complaints?*

In how many complaints was each of the bases for discrimination
alleged?*   What were the most frequently cited issues in
employees' discriminationcomplaints and in how many complaints was
each of the issues cited?

Moreover, the reliability of the data that EEOC collects from
agencies andreports is questionable. We found that agencies
reported basis and issue data to EEOC in an inconsistent manner.
We also found that agencies didnot report to EEOC some of the data
it requested and reported some other data incorrectly. In
addition, because EEOC did not have procedures thatensured the
reliability of the data it collected from agencies, it published
some unreliable data in its annual Federal Sector Report on
EEOComplaints Processing and Appeals.

We make recommendations to address these concerns near the end of
thisreport. Under a variety of statutes, federal employees,
including postal workers,can file a complaint alleging unlawful
employment discrimination.

3 Each

discrimination complaint contains two key elements that
provideinformation about the nature of the conflict.

The first of these two elements is the "basis" of the allegation
under federalantidiscrimination law. An employee can allege
discrimination on any of seven bases--race, color, national
origin, sex, religion, age, and disability.In addition, federal
employees can claim an eighth basis--reprisal--if they believe
that they have been retaliated against for having filed a
complaint,participated in an investigation of a complaint, or
opposed a prohibited

3 Federal employee discrimination complaints are processed in
accordance with regulations (29 C.F.R.

part 1614) promulgated by EEOC. Under these regulations, a
complaint is filed with the federal agencyaccused of
discrimination. The agency decides whether to dismiss or accept
the complaint, and

investigates the complaint if it is accepted. After the
investigation, a complainant can request a hearingbefore an EEOC
administrative judge, who may issue a recommended decision that
the agency is to consider in making its final decision on the
merits of the complaint. An employee dissatisfied with afinal
agency decision or its decision to dismiss a complaint may file an
appeal with EEOC. For a further discussion of the complaint
process, see Equal Employment Opportunity: Rising Trends in
ComplaintCaseloads in the Federal Sector (GAO/GGD-98-157BR, July
24, 1998).

Background

B-282387

Page 3 GAO/GGD-99-75 EEO Complaint Data Shortcomings personnel
practice. Depending upon the employee's situation, he or shecan
claim more than one basis when filing an EEO complaint.

4

The second of the two elements that help define the nature of the
conflictin a discrimination complaint is the "issue"--that is, the
specific condition or event that is the subject of the complaint.
Issues that employees can filecomplaints about include nonsexual
and sexual harassment, nonselection for promotion, performance
evaluations, duties that are assigned to them,and disciplinary
actions (e.g., demotion, reprimand, suspension, and termination).
(See app. I for a listing of categories of issues). As is truewith
respect to bases for complaints, an employee can raise multiple
issues in a single complaint. Agencies are required by regulations
(29 C.F.R. 1614.602) and the EEOCFederal Sector Complaint
Processing Manual, Equal Employment Opportunity Management
Directive (EEO MD)-110 to report annually toEEOC data about the
bases and issues cited in complaints, along with other complaint-
related statistics.5 EEOC compiles the data from theagencies for
publication in the annual Federal Sector Report on EEO Complaints
Processing and Appeals. According to the Management Directive,
"The analyses of the datacollected enable the EEOC to assist in
refining the efficiency and effectiveness of the Federal EEO
process." This objective conforms withone of the goals contained
in EEOC's Annual Performance Plans for fiscal years 1999 and 2000.
Likewise, as indicators of the nature and extent ofworkplace
conflict, these data could be important to EEOC as it carries out
its broader mission, which, as stated in the agency's Strategic
Plan, "isto promote equal opportunity in employment by enforcing
the federal civil rights employment laws through administrative
and judicial actions, andeducation and technical assistance."

6

4 For example, an employee complaining about nonselection for a
promotion can allege (1) race

discrimination if the person who was selected is of a different
race, (2) sex discrimination if the personwho was selected is of
the opposite sex, (3) age discrimination if the person who was
selected is under

40 years old while the complainant is 40 years of age or older,
and (4) disability discrimination if theperson who was selected is
not disabled but the complainant is.

5 Agencies with an allocation of 100 or more employees are to
complete the Annual Federal Equal Employment Opportunity
Statistical Report of Discrimination Complaints (EEOC Form 462). 6
The Strategic and Annual Performance Plans are required by the
Government Performance and Results Act.

B-282387 Page 4 GAO/GGD-99-75 EEO Complaint Data Shortcomings In
assessing why the data collected and reported by EEOC were
nothelpful in answering fundamental questions about the nature and
extent of conflict in the federal workplace, we examined several
sources. Wereviewed instructions for EEOC Form 462, Annual Federal
Equal Employment Opportunity Statistical Report of Discrimination
Complaints,the form that agencies use to report complaint basis
and issue data to EEOC, particularly part IV of the form, Summary
of Bases and Issues inComplaints Filed (see app. I for a copy of
part IV of EEOC Form 462.) We examined statistics on complaint
bases and issues published in EEOC'sFederal Sector Report on EEO
Complaints Processing and Appeals for fiscal years 1991 to 1997.7
Because postal workers accounted for about halfof the
discrimination complaints federal workers filed in fiscal year
1997, we obtained and analyzed forms 462 covering fiscal years
1991 to 1997 thatthe Postal Service submitted to EEOC in order to
compare statistics for the postal workforce with the nonpostal
workforce. In addition, the PostalService provided us additional
data on bases and issues generated by its complaint information
system. We did not examine forms 462 fornonpostal agencies as we
did for the Postal Service.

Although Form 462 data that each agency submits show the number
oftimes the different issues were raised in each basis category,
EEOC does not aggregate these data from all agencies to prepare a
consolidated Form462 (part IV). At our request, EEOC prepared a
consolidated Form 462 (part IV). Because EEOC does not routinely
compile data this way, werequested this information only for
fiscal year 1997. EEOC provided data for all federal agencies and,
by subtracting Postal Service data, alsoprovided data for
nonpostal agencies.

Further, we spoke with officials at EEOC and the Postal Service
andrepresentatives of the Council of Federal EEO and Civil Rights
Executives. These officials provided observations about trends in
the bases for andissues cited in complaints. Their comments, they
said, were based on their experiences, rather than on specific
studies. In addition, Council membersfrom the Departments of
Treasury and the Army provided information on how their respective
agencies report complaint basis and issue data. Finally, we
reviewed sections of EEOC's Strategic Plan and its
AnnualPerformance Plans for fiscal years 1999 and 2000 pertaining
to the agency's federal sector operations.
7 We worked with a draft EEOC provided of the fiscal year 1997
Federal Sector Report on EEO

Complaints Processing and Appeals. The fiscal year 1997 federal
sector report was published on April27, 1999.

Scope andMethodology

B-282387

Page 5 GAO/GGD-99-75 EEO Complaint Data Shortcomings We requested
comments on a draft of this report from the Chairwoman,EEOC, and
the Postmaster General. Their comments are discussed near the end
of this report. We did our work from October 1998 through
March1999 in accordance with generally accepted government
auditing standards. EEOC does not collect relevant data in a way
that would help answer somefundamental questions about the nature
and extent of workplace conflict alleged in federal employees'
discrimination complaints. Among the kindsof questions that cannot
be answered are:

*  How many individuals filed complaints?*

In how many complaints was each of the bases for discrimination
alleged?*   What were the most frequently cited issues in
employees' discriminationcomplaints and in how many complaints was
each of the issues cited?

Answers to such questions would help decisionmakers and
programmanagers understand the extent to which different
categories of employees are filing complaints and the conditions
or events that arecausing them to allege discrimination.

One fundamental question that cannot be answered is the number
ofindividual employees who have filed complaints. EEOC does not
collect data on the number of employees who file complaints, nor
on how oftenindividual employees file complaints. These numbers
would be crucial to an analysis of the extent to which the
increase in the number of complaintsin the 1990s (see p. 1) was
due to individuals filing first-time complaints or included
individuals who had filed other complaints in the past.
Withoutdata on the number of complainants and the frequency of
their complaints, decisionmakers do not have a clear picture of
the nature and extent ofalleged discrimination in the workplace
and the actions that may be necessary to deal with these
allegations. For example, a number of factors indicate that the
increase in the numberof discrimination complaints does not
necessarily signify an equivalent increase in the actual number of
individuals filing complainants. First, anundetermined number of
federal employees have filed multiple complaints. According to
EEOC and Postal Service officials and representatives of
theCouncil of EEO and Civil Rights Executives, while they could
not readily provide figures, they said it has been their
experience that a small numberof employees--often referred to as
"repeat filers"--account for a disproportionate share of
complaints. Additionally, an EEOC workgroupthat reviewed the
federal employee discrimination complaint process

Some Data That WouldAnswer Fundamental Questions Are NotCollected

Data on the Number ofComplainants and How Often Employees
FileComplaints Are Not Collected

B-282387

Page 6 GAO/GGD-99-75 EEO Complaint Data Shortcomings reported that
the number of cases in the system was "swollen" byemployees filing
"spin-off complaints"--new complaints challenging the processing
of existing complaints. Further, the work group found that
thenumber of complaints was "unnecessarily multiplied" by agencies
fragmenting some claims involving a number of different
allegations by thesame employee into separate complaints rather
than consolidating these claims into one complaint.8 In addition,
there has been an increase in thenumber of complaints alleging
reprisal, which, for the most part, involve claims of retaliation
by employees who have previously participated in thecomplaint
process.

Questions about the prevalence of bases and issues in the universe
ofcomplaints are not answerable because of the manner in which
EEOC collects these data. Accurate answers to such questions are
necessary tohelp decisionmakers and program managers discern
trends in workplace conflicts, understand the sources of conflict,
and plan corrective actions.These data could give managers a
clearer picture of the extent to which particular groups of
employees may feel aggrieved and the conditions orevents that
trigger their complaints. For example, managers would be able to
better discern trends in the numbers of black employees alleging
racialdiscrimination and the issues they have raised most
frequently.

EEOC prescribes a format for agencies to report complaint bases
andissues data (see app. I). The form is a matrix that, according
to EEOC instructions, requires agencies to associate the basis or
bases of anindividual complaint with the issue or issues raised in
that complaint. However, there are problems in counting bases and
issues this way.Complaints with two or more bases and/or issues
can result in the same basis and/or issue being counted more than
once. For example, suppose anemployee specifies that race, sex,
age, and disability discrimination were the bases for his or her
complaint, while nonselection for promotion, apoor performance
evaluation, and an assignment to noncareer-enhancing duties were
the issues. In preparing the report to EEOC, the agency
wouldrecord each of the three issues in the columns corresponding
with each of the four bases. Table 1 illustrates how this
complaint would fit into thepreparation of the overall report to
EEOC. The table is a matrix with excerpts of similar rows and
columns that appear on the form submitted toEEOC (see app. I).

8 EEOC has proposed regulatory changes to deal with "spin-off"
complaints and said it plans to amend

its policies to deal with the problem of fragmented complaints.

Prevalence of Bases andIssues Is Not Determined

B-282387

Page 7 GAO/GGD-99-75 EEO Complaint Data Shortcomings

Bases  Issues Race Sex Age Disability TotalNonselection x x x x 4

Evaluation x x x x 4Duties x x x x 4 Total 3 3 3 3 12 Legend: "X"
indicates that an issue and a basis were associated.

Source: GAO hypothetical illustration.

To determine the number of times each basis is alleged, EEOC
instructsagencies to add the number of times each issue was
recorded in each column of the matrix. In this illustration, the
agency would count eachbasis three times--once for each of the
three issues recorded in each of the columns. To determine the
number of times each issue is alleged,EEOC instructs agencies to
add each row of the matrix. In this illustration, the agency would
count each issue four times--once for each of the fourbases under
which they were recorded. Overall, the agency would report that 12
bases and 12 issues were alleged in this single
hypotheticalcomplaint rather than the 4 bases and 3 issues
actually cited.

EEOC uses these data from agencies to compile the number of times
eachbasis and each issue was alleged governmentwide, which it
publishes in the annual Federal Sector Report on EEO Complaints
Processing andAppeals. The figure reported for the number of times
that a particular basis was alleged, however, represents the sum
of the number of times that thevarious issues were recorded in the
column under that basis, not the actual number of complaints in
which that basis was alleged. Similarly, the figurereported for
the number of times that a particular issue was cited represents
the sum of the number of times the issue was recorded undereach of
the bases, not the actual number of complaints in which that issue
was cited. EEOC does not know the extent to which bases and issues
maybe counted more than once for the same complaint. EEOC's
Complaint Adjudication Division Director said that while the
reporting proceduresresult in overreporting of the number of times
the different bases and issues were alleged, he believes that the
data provide a "fairapproximation" of bases and issues included in
complaints. He agreed, however, that recording data in a way that
would establish the number oftimes the different bases and issues
are cited in the universe of complaints would make sense. The way
EEOC collects basis and issue data does, however, yield
someinsight into the importance of the different issues to the
different categories of complainants. The form that each agency is
to completeshows the issues raised under each basis and the number
of times that a

Table 1: Recording Bases and Issues ina Hypothetical Complaint

B-282387

Page 8 GAO/GGD-99-75 EEO Complaint Data Shortcomings particular
issue was raised. With these data, an agency manager candetermine,
for example, the issues that female employees alleging sex
discrimination complained about and the number of times each of
thoseissues was raised. The one essential statistic that is
missing, however, is the actual number of complaints made by women
alleging sexdiscrimination. Further, while EEOC collects
information showing the extent to which specific issues are
associated with specific bases at eachagency, it does not
aggregate this information for all federal agencies.

9

The discrimination complaint data that EEOC has collected and
reportedare of questionable reliability because (1) agencies did
not always report data consistently, completely, or accurately and
(2) EEOC did not haveprocedures that ensured the data were
reliable.

Federal agencies take varying approaches to reporting data on
complaintbases and issues to EEOC. We reviewed the Postal
Service's data submissions to EEOC, as well as the process to
prepare these submissions,and found that the agency did not follow
EEOC's instructions to associate the issue or issues raised in
each complaint with the basis or basesinvolved. For each
complaint, regardless of the number of issues raised by the
employee, the Postal Service identified and reported only
one"primary" issue. In commenting on a draft of this report, the
Postal Service's Manager, EEO Compliance and Appeals, said that
the PostalService adopted this approach to give the data more
focus by identifying the primary issues driving postal workers'
discrimination complaints. We did not review reports and reporting
practices among nonpostalagencies for consistency and attention to
completeness and accuracy. However, we spoke with officials from
two large nonpostal agencies whoindicated that they followed EEOC
instructions, which, as discussed above, can result in an
overcounting of bases and issues. EEOC'sComplaints Adjudication
Division Director said that agencies might be using different
approaches to reporting the data. However, he said that hedid not
know the extent to which such variation may exist because EEOC had
not examined how agencies complete their reports. The issue of
incomplete or inaccurate reporting of data was evident in
ouranalysis of the data that the Postal Service reported to EEOC
for fiscal years 1992 and 1995 through 1997. We analyzed Postal
Service statistics
9 At our request, EEOC aggregated this data governmentwide for
fiscal year 1997. While we believe the

aggregated information provided some insight into federal
workplace conflict, its immediate usefulnesswas compromised by the
reliability issues discussed in this report.

Data Are Not AlwaysReported Consistently, Completely, orAccurately

B-282387

Page 9 GAO/GGD-99-75 EEO Complaint Data Shortcomings because
postal workers accounted for about half of the
discriminationcomplaints filed by federal employees in fiscal year
1997. In addition to not completely reporting all issues raised in
complaints, we found that thePostal Service's statistical reports
to EEOC for fiscal years 1996 and 1997 did not include data for
certain categories of issues.10 Further, we foundcertain
underreporting of bases for complaints and issues by the Postal
Service in fiscal year 1995. Postal Service officials also told us
thatcomplaint statistics were incomplete for fiscal year 1992.

Another, especially significant, reporting error we identified
involved thenumber of race-based complaints. As a result of a
computer programming error, the number of complaints reported by
the Postal Service to containallegations by white postal workers
of discrimination based on race was overstated in fiscal years
1996 and 1997 by about 500 percent.11 After we brought these
errors to the attention of Postal Service officials,they provided
corrected data to us and EEOC for all errors except those relating
to the fiscal year 1992 data. Postal Service officials said
thatbecause EEO-related staff had been reassigned during
restructuring of the Postal Service that began in fiscal year
1992, not all complaints wereproperly accounted for that year. The
officials also said that the computer program used to generate
reports to EEOC had been modified to correctthe fault in the way
race-based complaints are to be counted.

Errors in data reported to or by EEOC were a recurring problem in
ourwork identifying trends in federal sector EEO complaints.

12 In addition to

the Postal Service data errors, during our prior work, we found
errors fornonpostal agencies' data. EEOC does not audit or verify
the data it

receives from agencies and publishes in the annual Federal Sector
Reporton EEO Complaints Processing and Appeals because of time

10 The Postal Service developed additional categories of issues to
further delineate the categories

established by the EEOC statistical report (see app. I). In
completing the EEOC form, however, thePostal Service failed to
include data about these categories of issues. These omitted data
involved both

the number of complaints in which the delineated issue was the
primary issue and the bases alleged inthose complaints.

11 In fiscal year 1996, the Postal Service originally reported to
EEOC that 9,044 (about 68 percent) of the 13,252 complaints filed
that year contained allegations by white postal workers of race
discrimination.The corrected figure was 1,505 (11.4 percent) of
the complaints filed that year. In fiscal year 1997, the

Postal Service originally reported that 10,040 (70 percent) of the
14,326 complaints filed that yearcontained such allegations. The
corrected figure was 1,654 (11.5 percent) of the complaints filed
that year. 12 Equal Employment Opportunity: Rising Trends in EEO
Complaint Caseloads in the Federal Sector (GAO/GGD-98-157BR, July
24, 1998) and Equal Employment Opportunity: Administrative
Judges'Recommended Decisions and Agencies' Actions (GAO/GGD-98-
122R, June 10, 1998).

B-282387 Page 10 GAO/GGD-99-75 EEO Complaint Data Shortcomings
considerations and staff limitations, according to the
ComplaintsAdjudication Division Director. He said, however, that
EEOC staff review agencies' data to identify figures that appear
unusual or inconsistent withother data reported. As we observed,
this procedure did not ensure the reliability of the data EEOC
collected and put in print. For example, inpreparing the
aggregated figures that it published in its federal sector report
for fiscal year 1996, EEOC used the Postal Service's
vastlyoverstated data on racial discrimination complaints by white
employees, thereby skewing the portrayal of discrimination
complaint trendsgovernmentwide.

13

Data about the bases for complaints and the issues giving rise to
them canbe valuable in gauging conflict in the federal workplace.
However, EEOC does not collect or report relevant agency data in a
way that would helpanswer fundamental questions about the number
of complainants and the prevalence of bases and issues in the
universe of complaints. In addition,some of the data collected and
reported by EEOC have lacked the necessary reliability because
agencies did not report their dataconsistently, completely, or
accurately, and because EEOC did not have procedures that ensured
the data were reliable. Consequently, the data donot provide a
sound basis for decisionmakers, program managers, and EEOC to
understand the nature and extent of workplace conflict,
developstrategies to deal with conflict, and measure the results
of interventions.

To help ensure that relevant and reliable data are available
todecisionmakers and program managers, we recommend that the
Chairwoman, EEOC, take steps to enable EEOC to

*  collect and publish data on complaint bases and issues in a
manner thatwould allow fundamental questions about the number of
complainants and

the prevalence of bases and issues in the universe of complaints
to beanswered, and

*  develop procedures to help ensure that agencies report data
consistently,completely, and accurately.

We received comments on a draft of this report from EEOC and the
PostalService. In its written comments (see app. II), EEOC agreed
that the data collected from federal agencies could be more
comprehensive andaccurate. EEOC said that it would expedite its
efforts to revise the
13 After the incorrect Postal Service data had been brought to
EEOC's attention, an EEOC official told

us that EEOC recalled the fiscal year 1997 federal sector report
before final publication in order tocorrect fiscal year 1997 data.
The fiscal year 1997 report would also contain corrected numbers
for

fiscal years 1995 and 1996 when it is published.

Conclusions Recommendations Agency Comments andOur Evaluation

B-282387

Page 11 GAO/GGD-99-75 EEO Complaint Data Shortcomings instructions
for data collection and that it would address the concerns
weraised in this report. EEOC further stated that, given the
required review and approval processes, including allowing time
for federal agencies tocomment, it would take about 8 months to
issue the changes and an additional 12 months for the agencies to
report complaint data to EEOC inaccordance with the new
instructions.

Under EEOC's timetable, it will be several years before EEOC's
annualfederal sector reports reflect the results of the agency's
efforts to revise instructions for data collection and to promote
more comprehensive andreliable reporting. EEOC's revised
instructions would be issued in the beginning of fiscal year 2000,
and the first complete fiscal year for whichthe instructions would
be applicable would be fiscal year 2001. Agencies' statistical
reports for the fiscal year ending 2001 would not be submitted
toEEOC until fiscal year 2002 for later publication in the Federal
Sector Report on EEO Complaints Processing and Appeals. EEOC did
notindicate, however, when the first federal sector report
containing these data would be published. EEOC also said that that
it would take action to address our concernsabout data
consistency, completeness, and accuracy. To deal with problems in
the reliability of the data collected from agencies, EEOC saidthat
it would urge agencies to give higher priority to the accuracy of
their data. EEOC said it will ask agencies to certify the
reliability of the datathey provide and to explain how they ensure
the quality of their data. In addition, EEOC said that if
additional resources it has requested for fiscalyear 2000 become
available, it would be able to conduct on-site reviews to assess
the reliability of agency data, more closely examine the nature
ofworkplace disputes, and work with agencies to improve their EEO
programs. We believe that the actions proposed by EEOC are
generally responsive toour recommendation and would add some
measure of reliability to the data it collects and reports. By
urging agencies to give higher priority todata reliability, EEOC
would be reiterating its current policy, as stated in Management
Directive 110, that "Every effort should be made to ensureaccurate
recordkeeping and reporting of federal EEO data and that all data
submissions are fully responsive and in compliance with
informationrequests." By proposing that agencies certify the
reliability of the their data and explain how they ensure data
quality, EEOC will be providing amechanism for holding agencies
more accountable for producing reliable and accurate data and, if
followed, would have some basis to assess theextent to which an
agency's processes may ensure the data's reliability and

B-282387 Page 12 GAO/GGD-99-75 EEO Complaint Data Shortcomings
accuracy. An assessment of agencies' quality control procedures
andconsideration of discrepancies contained in previous data
submissions, among other factors, would enable EEOC to select
agencies for any futureon-site reviews based on the estimated risk
of agencies submitting unreliable data. On April 9, 1999, the
Postal Service's Manager, EEO Compliance andAppeals, provided oral
comments on a draft of this report. He said that the report, in
general, accurately describes the data shortcomings and opensthe
door for dialogue on how data could be collected in a manner that
would better serve decisionmakers. He agreed with the
recommendationthat data be collected on the number of
complainants. In addition, he suggested that data be collected on
the number of repeat filers. The officialsaid it has been his
experience that between 60 and 70 individuals account for every
100 complaints in a fiscal year. He also suggested that
EEOCcollect data about the race and sex of complainants along
basis and issue lines. He further suggested that similar data be
collected for individualsseeking counseling. The official said
that the Postal Service's complaint information system is capable
of producing this kind of informationbecause it tracks individuals
by their Social Security number. For example, he said that his
office has been able to provide Postal Service managementwith
complaint data for each of the Service's 85 districts in order to
identify the extent of workplace conflicts at the different
locations and theprimary issues driving the conflicts. He said,
however, that the issues listed on EEOC Form 462 (see app. I) need
to be revised to make them morerelevant to the agencies reporting
to EEOC. He suggested that EEOC convene a working group of federal
agency representatives to deal withthis and other data issues.

We believe the Postal Service official's suggestion that EEOC
develop aworking group of federal agency representatives to
participate in revising data collection requirements would allow
stakeholders to be activepartners in the development of data
collection requirements that affect them. Although we did not
identify all of the data that would be useful todecisionmakers and
program managers, a working group would provide a forum for
developing a consensus on data needs. It might be appropriate
toinclude congressional stakeholders in any working group because
of their oversight and policymaking responsibilities. It should be
noted that otheragencies that deal with redress and human capital
issues--the Office of Personnel Management and the Merit Systems
Protection Board--haveworking groups or panels to assist them in
carrying out their missions.

B-282387

Page 13 GAO/GGD-99-75 EEO Complaint Data Shortcomings The Postal
Service official also said it would be helpful if EEOC revised
itssystem of collecting data to facilitate more timely collection
and publication of federal sector EEO complaint data. He noted
that thefederal sector reports are published nearly 2 years after
the fiscal year's end. More timely data, he said, would make data
more useful todecisionmakers.

We agree that more timely data are more likely to be useful
todecisionmakers. Although timeliness is not an issue we reviewed,
we did observe what appeared to be lengthy periods before data
were madeavailable. For example, EEOC published the fiscal year
1997 Federal Sector Report on EEO Complaints Processing and
Appeals on April 27,1999, 18 months after the end of fiscal year
1997. The working group proposed by the Postal Service official
could be a forum for furtherexploring this issue.

As agreed with your offices, we plan no further distribution of
this reportuntil 30 days after its issuance, unless you publicly
release its contents earlier. We will then send copies of this
report to Senators Daniel K.Akaka, Thad Cochran, Joseph I.
Lieberman, and Fred Thompson; and Representatives Robert E.
Andrews, John A. Boehner, Dan Burton, WilliamL. Clay, Chaka
Fattah, William F. Goodling, Steny H. Hoyer, Jim Kolbe, John M.
McHugh, David Obey, Harold Rogers, Joe Scarborough, Jose
E.Serrano, Henry A. Waxman, and C. W. Bill Young in their
capacities as Chair or Ranking Minority Members of Senate and
House Committees andSubcommittees. We will also send copies to The
Honorable Ida L. Castro, Chairwoman, EEOC; The Honorable William
J. Henderson, PostmasterGeneral; The Honorable Janice R. Lachance,
Director, Office of Personnel Management; The Honorable Jacob Lew,
Director, Office of Managementand Budget; and other interested
parties. We will make copies of this report available to others on
request. Major contributors to this report are listed in appendix
III. Please contactme on (202) 512-8676 if you or your staff have
any questions concerning this report.

Michael BrostekAssociate Director, Federal Management

and Workforce Issues

Page 14 GAO/GGD-99-75 EEO Complaint Data Shortcomings

Contents

Letter 1 Appendix IPart IV of EEOC Form

462, Summary of Basesand Issues in Complaints Filed

16

Appendix IIComments From the U.S. EqualEmployment
OpportunityCommission

17

Appendix IIIMajor Contributors to This Report

19

Abbreviations EEO equal employment opportunity EEOC Equal
Employment Opportunity Commission

Page 15 GAO/GGD-99-75 EEO Complaint Data Shortcomings

Appendix IPart IV of EEOC Form 462, Summary of Bases and Issues in
Complaints Filed

Page 16 GAO/GGD-99-75 EEO Complaint Data Shortcomings Figure I.1:
Form Used by Agencies to Report Statistics on Bases and Issues in
Complaints Filed Appendix IIComments From the U.S. Equal
Employment Opportunity Commission

Page 17 GAO/GGD-99-75 EEO Complaint Data Shortcomings Appendix II
Comments From the U.S. Equal Employment Opportunity Commission

Page 18 GAO/GGD-99-75 EEO Complaint Data Shortcomings Appendix
IIIMajor Contributors to This Report

Page 19 GAO/GGD-99-75 EEO Complaint Data Shortcomings Stephen E.
Altman, Assistant Director, Federal Management and  Workforce
Issues Anthony P. Lofaro, Evaluator-in-ChargeGary V. Lawson,
Senior Evaluator Sharon T. Hogan, Evaluator

General GovernmentDivision

Page 20 GAO/GGD-99-75 EEO Complaint Data Shortcomings Ordering
Information The first copy of each GAO report and testimony is
free. Additionalcopies are $2 each. Orders should be sent to the
following address, accompanied by a check or money order made out
to theSuperintendent of Documents, when necessary. VISA and
MasterCard credit cards are accepted, also. Orders for 100 or
morecopies to be mailed to a single address are discounted 25
percent.

Order by mail: U.S. General Accounting OfficeP.O. Box 37050
Washington, DC 20013 or visit: Room 1100700 4

th St. NW (corner of 4th and G Sts. NW)

U.S. General Accounting OfficeWashington, DC

Orders may also be placed by calling (202) 512-6000 or by using
faxnumber (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.
Arecorded 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:

[email protected] or visit GAO's World Wide Web Home Page at:
http://www.gao.gov

United StatesGeneral Accounting Office Washington, D.C. 20548-0001

Official BusinessPenalty for Private Use $300 Address Correction
Requested

Bulk RatePostage & Fees Paid

GAOPermit No. G100

(410435)

*** End of document. ***