Elections: All Levels of Government Are Needed to Address	 
Electronic Voting System Challenges (07-MAR-07, GAO-07-576T).	 
                                                                 
Since the 2000 national elections, concerns have been raised by  
various groups regarding the election process, including voting  
technologies. Beginning in 2001, GAO published a series of	 
reports examining virtually every aspect of the elections	 
process. GAO's complement of reports were used by Congress in	 
framing the Help America Vote Act of 2002, which, among other	 
things, provided for replacement of older voting equipment with  
more modern electronic voting systems and established the	 
Election Assistance Commission (EAC) to lead the nation's	 
election reform efforts. GAO's later reports have raised concerns
about the security and reliability of these electronic voting	 
systems, examined the EAC's efforts to address these concerns,	 
and surveyed state and local officials about practices used	 
during the 2004 election, as well as plans for their systems for 
the 2006 election. Using its published work on electronic voting 
systems, GAO was asked to testify on (1) the contextual role and 
characteristics of electronic voting systems, (2) the range of	 
security and reliability concerns that have been reported about  
these systems, (3) the experiences and management practices of	 
states and local jurisdictions regarding these systems, and (4)  
the longstanding and emerging challenges facing all levels of	 
government in using these systems.				 
-------------------------Indexing Terms------------------------- 
REPORTNUM:   GAO-07-576T					        
    ACCNO:   A66549						        
  TITLE:     Elections: All Levels of Government Are Needed to Address
Electronic Voting System Challenges				 
     DATE:   03/07/2007 
  SUBJECT:   Elections						 
	     Electronic equipment				 
	     Optical scanners					 
	     Performance appraisal				 
	     Standards						 
	     Strategic planning 				 
	     Systems design					 
	     Systems development life cycle			 
	     Systems management 				 
	     Systems testing					 
	     Testing						 
	     Voting						 
	     Policies and procedures				 

******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO Product.                                                 **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
******************************************************************
GAO-07-576T

   

     * [1]Background

          * [2]HAVA Was Enacted to Strengthen the Overall Election Process
          * [3]Electronic Voting Systems Fall into Two Primary Categories

               * [4]Optical Scan Systems
               * [5]Direct Recording Electronic Systems

     * [6]The Contextual Role and Performance Characteristics of Elect

          * [7]Despite Their Vital Role, Voting Systems Are Only One Aspect
          * [8]Management of Electronic Voting System Performance Is a Cont
          * [9]Electronic Voting System Performance Can Be Judged on Severa

     * [10]Widespread Concerns about Electronic Voting Systems Have Bee

          * [11]Inadequate National Standards
          * [12]Inadequate System Design and Development
          * [13]Inadequate System Operation and Management Activities
          * [14]Inadequate Testing

     * [15]Variability and Weaknesses in State and Local Approaches to

          * [16]States' Use of Standards Varied
          * [17]Jurisdictions Varied Widely in Applying Security Practices
          * [18]Some Types of Testing Were Not Commonly Performed
          * [19]Jurisdictions Did Not Consistently Monitor Voting System Per

     * [20]Addressing Voting System Challenges Requires the Combined Ef

          * [21]Establishing and Applying Current and Comprehensive Standard
          * [22]Ensuring that Necessary Security, Testing, and Operational A
          * [23]Managing the People, Processes, and Technology as Components
          * [24]Gathering and Using Reliable System Performance Measures and

     * [25]Contact and Acknowledgements
     * [26]PDF6-Ordering Information.pdf

          * [27]Order by Mail or Phone

Mr. Chairman and Members of the Subcommittee:

I appreciate the opportunity to participate in today's hearing on the
Election Assistance Commission (EAC) and the integrity of the elections
process. As requested, my testimony will focus on the challenges
associated with the use of electronic voting systems during elections,^1
including the EAC's efforts to address these challenges.

During the 2000 national elections, concerns were raised about "hanging
chads" and "butterfly ballots." In the 2004 and 2006 elections, concerns
shifted to "software bugs" and "voter verifiable paper trails." In light
of these and other election concerns, we produced a series of reports
between 2001 and 2006 in which we examined virtually every aspect of the
election process, including types of voting technology. We reported that
the particular technology used to cast and count votes is a critical part
of how elections are conducted, but it is only one facet of a multifaceted
election process that involves the interplay of people, processes, and
technology. Accordingly, we have long held the position that no voting
technology, however well designed, can be a magic bullet that will solve
all election problems.

My testimony today addresses four perspectives on the voting system
environment: (1) the contextual role and characteristics of electronic
voting systems, (2) the range of security and reliability concerns that
have been reported about these systems, (3) the experiences and management
practices of states and local jurisdictions regarding these systems, and
(4) longstanding and emerging intergovernmental challenges in using these
systems.

^1In this testimony, the term electronic voting system is used generically
to refer to both optical scan systems and direct recording electronic
systems, both of which depend on electronic technology. Each type of
system is described more fully in the background section of this
testimony.

In preparing this testimony, we drew extensively from our published work
on the election process.^2 In addition, we reviewed recent studies of
electronic voting systems, EAC publications, and other relevant documents.
All the work on which this testimony is based was performed in accordance
with generally accepted government auditing standards.

In summary, the integrity of voting systems--which are but one variable in
the successful election process equation--depends on effective system life
cycle management, which includes systems definition, development,
acquisition, operations, testing, and management. It also depends on
measuring actual voting system performance in terms of security,
reliability, ease of use, and cost effectiveness, so that any needed
corrective actions can be taken. Unless voting systems are properly
managed throughout their life cycle, this one facet of the election
process can significantly undermine the integrity of the whole.

Election officials, computer security experts, citizen advocacy groups,
and others have raised significant concerns about the security and
reliability of electronic voting systems, citing vague or incomplete
standards, weak security controls, system design flaws, incorrect system
configuration, poor security management, and inadequate security testing,
among other issues. These security and reliability concerns are legitimate
and thus merit the combined and focused attention of federal, state, and
local authorities responsible for election administration.

For the 2004 general election, states and local jurisdictions' responses
to our surveys showed that they did not always use the most current voting
system standards. Responses also showed that voting system practices were
implemented to varying degrees and that certain types of system testing
were not widely performed. Moreover, responses indicated that
jurisdictions did not consistently monitor the performance of their
systems. Such monitoring is important for determining where performance
needs, requirements, and expectations are not being met so that corrective
actions are taken.

^2For example, GAO, Elections: Perspectives on Activities and Challenges
across the Nation, GAO-02-3 (Washington, D.C.: Oct. 15, 2001); Elections:
Electronic Voting Offers Opportunities and Presents Challenges,
GAO-04-975T (Washington, D.C.: July 20, 2004); Elections: Federal Efforts
to Improve Security and Reliability of Electronic Voting Systems Are Under
Way, but Key Activities Need to Be Completed, GAO-05-956 (Washington,
D.C.: Sept. 21, 2005); and Elections: The Nation's Evolving Election
System as Reflected in the November 2004 General Election, GAO-06-450
(Washington, D.C.: June 6, 2006).

The challenges confronting all levels of government in acquiring and
operating voting systems for future elections are not unlike those faced
by any technology user: adoption and consistent application of standards
for system capabilities and performance; successful management and
integration of the people, process, and technology components; rigorous
and disciplined performance of testing and security activities; reliable
measurement to determine whether the systems are performing as intended;
and analytical and economically justified basis for making informed
decisions about voting system investment options. These challenges are
heightened by other conditions common to both the national elections
community and other information technology environments: the distribution
of responsibilities among various organizations, technology changes,
funding opportunities and constraints, emerging requirements and guidance,
and public attention.

Given the diffused and decentralized allocation of voting system roles and
responsibilities across all levels of government, addressing these
challenges will require the combined efforts of all levels of government,
under the leadership of the EAC. To assist the EAC in executing its
leadership role, we previously made recommendations to the commission
aimed at better planning its ongoing and future activities relative to,
for example, system standards and information sharing. While the EAC
agreed with the recommendations, it told us that its ability to
effectively execute its role is resource constrained. In our view,
critical to the commission's ability to perform its leadership role will
be the adequacy of resources at its disposal and the degree of cooperation
it receives from entities spanning all levels of government.

Background

Following the 2000 national elections, we produced a comprehensive series
of reports covering our nation's election process that culminated with a
capping report and framework for the Congress to use to enact reforms for
election administration.^3 Our reports were among the resources that the
Congress drew on in enacting the Help America Vote Act (HAVA) of 2002,^4
which provided guidance for fundamental election administration reform and
created the Election Assistance Commission (EAC) to oversee the election
administration reform process. HAVA also provided funding to replace older
voting equipment, specifically punch card and mechanical lever voting
equipment and encouraged adoption of other technology.^5 Subsequently,
jurisdictions have increased their use of electronic voting methods, of
which there are two commonly-used types: optical scan and direct recording
electronic (DRE).

HAVA Was Enacted to Strengthen the Overall Election Process

Enacted by Congress in October 2002, HAVA affects nearly every aspect of
the voting process, from voting technology to provisional ballots and from
voter registration to poll worker training. In particular, the act
authorized $3.86 billion in funding over several fiscal years for programs
to replace punch card and mechanical lever voting equipment, improve
election administration and accessibility, train poll workers, and perform
research and pilot studies. HAVA also established the EAC to assist in the
administration of federal elections and provide assistance with the
administration of certain federal election laws and programs. HAVA also
established minimum election administration standards for the states and
units of local government that are responsible for the administration of
federal elections. The act specifically tasked the EAC to serve as a
national clearinghouse and resource for compiling election information and
reviewing election procedures; for example, it is to conduct periodic
studies of election administration issues, including electronic voting
system performance, to promote methods of voting and administration that
are most convenient, accessible, and easy to use for all voters. Other
examples of EAC responsibilities include

^3See, for example, GAO, Elections: A Framework for Evaluating Reform
Proposals, [28]GAO-02-90 (Washington, D.C.: Oct. 15, 2001).

^4Help America Vote Act of 2002, Pub. L. No. 107-252 (Oct. 29, 2002).

^5The General Services Administration (GSA) is responsible for
administering grants to the states to replace punch card systems and lever
machines in qualifying states, including providing payments for general
election administration improvements to states that apply for funds to
replace voting equipment.

           o developing and adopting voluntary voting system guidelines and
           maintaining information on the experiences of states in
           implementing the guidelines and operating voting systems;
           o testing, certifying, decertifying, and recertifying voting
           system hardware and software through accredited laboratories;
           o making payments to states to help them improve elections in the
           areas of voting systems standards, provisional voting and voting
           information requirements, and computerized statewide voter
           registration lists; and
           o making grants for research on voting technology improvements.

The act also established the Technical Guidelines Development Committee to
support the EAC, making it responsible for recommending voluntary voting
system guidelines to the EAC. The act assigned the National Institute of
Standards and Technology (NIST) responsibility for providing technical
support to the development committee and made the NIST Director the
committee chair.

The EAC began operations in January 2004, initially focusing on the
distribution of funds to help states meet HAVA's Title III requirements
for uniform and nondiscriminatory election technology and administration,
including the act's requirements pertaining to voting system standards,
provisional voting, voting information, a computerized statewide voter
registration list, and identification for first-time voters who register
to vote by mail. Actions EAC has taken since 2004 to improve voting
systems include

           o publishing the Best Practices Toolkit and specialized management
           guides to assist states and local jurisdictions with managing
           election-related activities and equipment;
           o issuing voting system standards in 2005, referred to as the
           Voluntary Voting System Guidelines;
           o establishing procedures for certifying voting systems;
           o establishing a program for accreditation of independent testing
           laboratories, with support from NIST's National Voluntary
           Laboratory Accreditation Program;
           o disbursing to states approximately $2.3 billion in
           appropriations for the replacement of older voting equipment and
           election administration improvements under Title III of HAVA; and
           o conducting national surveys of the 2004 general election,
           uniformed and overseas voters, and other studies.

For fiscal year 2006, EAC's appropriation totaled $14.1 million. EAC
reported that this included $3.8 million (27 percent) for activities
related to development and adoption of the voting system standards and the
voting system certification program; $3.5 million (25 percent) for
research and study and to establish the EAC as a national clearinghouse of
election administration information; and $2.8 million (20 percent) to
manage HAVA funds distributed to the states. The remaining funds went to
various administrative expenses, including funding various advisory board
meetings. EAC's budget for fiscal year 2007 is $16.91 million, of which
$4.95 million (29 percent) is to be transferred to NIST for its work on
voting system standards and research performed under HAVA. EAC's requested
budget for fiscal year 2008 is $15.5 million, of which $3.25 million (21
percent) is to be transferred to NIST.

Electronic Voting Systems Fall into Two Primary Categories

In the United States today, most votes are cast and counted by one of two
types of electronic voting systems: optical scan and direct recording
electronic. For the November 2004 general election, optical scan was the
predominant voting method for more than half of local jurisdictions
nationwide. In contrast, DREs were used as the predominant voting method
by an estimated 7 percent of jurisdictions, although they were the
predominant voting method for large jurisdictions.^6 Figure 1 shows the
estimated use of predominant voting methods for small, medium, and large
jurisdictions in the 2004 general election.

^6 To obtain national information from local election officials on changes
to election systems since 2000, election administration, and their
experiences in the 2004 general election, we conducted a mail survey of a
stratified random probability sample of 788 local election jurisdictions
nationwide. Unless otherwise noted, the maximum sampling error at the 95
percent level of statistical confidence for estimates from this survey of
all jurisdictions is plus or minus 5 percentage points, plus or minus 7
percentage points for large jurisdictions, plus or minus 7 percentage
points for medium population size jurisdiction, and plus or minus 5
percentage points for small population size jurisdictions. For more
details about this survey, see Appendix V in GAO-06-450.

Figure 1: Estimated Percentage of Jurisdictions Using Predominant Voting
Methods in 2004, by Jurisdiction Size

Note: Percentages for predominant voting methods within each jurisdiction
size may not add to 100 because of rounding.

^aThe differences between small jurisdictions and both medium and large
jurisdictions are statistically significant.

^bThe differences between both small and medium jurisdictions and large
jurisdictions are statistically significant.

^cThe differences between both small and medium jurisdictions and large
jurisdictions are statistically significant.

^dThe difference between small jurisdictions and large jurisdictions is
statistically significant.

^eThe differences between small jurisdictions and both medium and large
jurisdictions are statistically significant.

  Optical Scan Systems

Optical scan voting systems use electronic technology to tabulate paper
ballots. For the 2004 general election, we estimated that about 51 percent
of all local jurisdictions predominantly used optical scan voting
equipment.

An optical scan voting system is made up of computer-readable ballots,
appropriate marking devices, privacy booths, and a computerized tabulation
device. The ballot, which can be of various sizes, lists the names of the
candidates and the issues. Voters record their choices using an
appropriate writing instrument to fill in boxes or ovals or to complete an
arrow next to the candidate's name or the issue. The ballot includes a
space for write-ins to be placed directly on the ballot.

Optical scan ballots are tabulated by optical-mark-recognition equipment
(see figure 2), which counts the ballots by sensing or reading the marks
on the ballot. Ballots can be counted at the polling place--this is
referred to as precinct-count optical scan^7--or at a central location. If
ballots are counted at the polling place, voters or election officials put
the ballots into the tabulation equipment, which tallies the votes; these
tallies can be captured in removable storage media that are transported to
a central tally location, or they can be electronically transmitted from
the polling place to the central tally location. If ballots are centrally
counted, voters drop ballots into sealed boxes and election officials
transfer the sealed boxes to the central location after the polls close,
where election officials run the ballots through the tabulation equipment.

^7Precinct-count optical scan equipment sits on a ballot box with two
compartments for scanned ballots--one for accepted ballots (i.e., those
that are properly filled out) and one for rejected ballots (i.e., blank
ballots, ballots with write-ins, or those accepted because of a forced
override). In addition, an auxiliary compartment in the ballot box is used
for storing ballots if an emergency arises (e.g., loss of power or machine
failure) that prevents the ballots from being scanned.

Figure 2: Precinct-Count Optical Scan Tabulator and Central-Count Optical
Scan Tabulator

Software instructs the tabulation equipment to assign each vote (i.e., to
assign valid marks on the ballot to the proper candidate or issue). In
addition to identifying the particular contests and candidates, the
software can be configured to capture, for example, straight party voting
and vote-for-no-more-than-N contests. Precinct-based optical scanners can
also be programmed to detect overvotes (where the voter votes for two
candidates for one office, for example, invalidating the vote) and
undervotes (where the voter does not vote for all contests or issues on
the ballot) and to take some action in response (rejecting the ballot, for
instance), so that voters can fix their mistakes before leaving the
polling place. If ballots are tabulated centrally, voters do not have the
opportunity to detect and correct mistakes that may have been made. In
addition, optical scan systems often use vote-tally software to tally the
vote totals from one or more vote tabulation devices.

Optical scan systems were widely used as the predominant voting method for
jurisdictions in the 2004 general election and we reported last year that
jurisdictions planned to acquire more of these systems for the 2006
general election. We estimated that 30 percent of jurisdictions nationwide
used precinct count optical scan voting equipment as their predominant
voting method for the 2004 general election, while an estimated 21 percent
used central count optical scan predominantly. Figure 1 shows the
percentages of jurisdictions using optical scan equipment as the
predominant voting method by jurisdiction size. While all sizes of
jurisdictions had plans to acquire both precinct count and central count
optical scan systems for the 2006 general election, small jurisdictions
showed a strong preference for acquiring precinct count optical scan
systems (estimated at 28 percent of small jurisdictions) compared with
DREs (13 percent) and central count optical scan (4 percent).

  Direct Recording Electronic Systems

DREs capture votes electronically without the use of paper ballots. For
the 2004 general election, we estimated that about 7 percent of all local
jurisdictions used DREs predominantly, although 30 percent of all large
jurisdictions used them as the predominant voting method.

DREs come in two basic types: pushbutton or touchscreen, with pushbutton
being the older technology. The two types vary considerably in appearance,
as shown in figure 3.

Figure 3: DRE Pushbutton and DRE Touchscreen

Pushbutton and touchscreen units differ significantly in the way they
present ballots to the voter. With the pushbutton type, all ballot
information is presented on a single "full-face" ballot. For example, a
ballot may have 50 buttons on a 3 by 3 foot ballot, with a candidate or
issue next to each button. In contrast, touchscreen DREs display ballot
information on an electronic display screen. For both pushbutton and
touchscreen types, the ballot information is programmed onto an electronic
storage medium, which is then uploaded to the machine. For touchscreens,
ballot information can be displayed in color and can incorporate pictures
of the candidates. Because the ballot space on a touchscreen is much
smaller than on a pushbutton machine, voters who use touchscreens must
page through the ballot information. Both touchscreen and pushbutton DREs
can accommodate multilingual ballots.

Despite differences between pushbutton and touchscreen DREs, the two types
have some similarities, such as how the voter interacts with the voting
equipment. To make a ballot selection, voters press a button or the screen
next to the candidate or issue, and the button or screen then lights up to
indicate the selection. When voters are finished making their selections,
they cast their votes by pressing a final "vote" button or screen. Until
they hit this final button or screen, voters can change their selections.
DREs are designed to not allow overvotes. Both types allow voters to write
in candidates. While most DREs allow voters to type write-ins on a
keyboard, some pushbutton types require voters to write the name on paper
tape that is part of the device. In addition, different types of DREs
offer a variety of options that jurisdictions may choose to purchase, such
as printed receipts or audio interfaces for voters with disabilities.

Although DREs do not receive paper ballots, they can retain permanent
electronic images of all the ballots, which can be stored on various
media, including internal hard disk drives, flash cards, or memory
cartridges. According to vendors, these ballot images, which can be
printed, can be used for auditing and recounts.

Like optical scan devices, DREs require the use of software to program the
various ballot styles and tally the votes, which is generally done through
the use of memory cartridges or other media. Some of the newer DREs use
smart card technology as a security feature. Smart cards are plastic
devices--about the size of a credit card--that use integrated circuit
chips to store and process data, much like a computer. Smart cards are
generally used as a means to open polls and to authorize voter access to
ballots. For instance, smart cards on some DREs store program data on the
election and are used to help set up the equipment; during setup, election
workers verify that the card received is for the proper election. Other
DREs are programmed to automatically activate when the voter inserts a
smart card; the card brings up the correct ballot onto the screen.

DREs offer various configurations for tallying the votes. Some contain
removable storage media that can be taken from the voting device and
transported to a central location to be tallied. Others can be configured
to electronically transmit the vote totals from the polling place to a
central tally location. Vote tally software is often used to tally the
vote totals from one or more units.

DREs were chosen as the predominant voting method by a relatively small
overall proportion of local jurisdictions for the 2004 general election (7
percent overall). However, as previously shown in figure 1, large and
medium jurisdictions identified DREs as their predominant voting method
(estimated at 30 percent and 20 percent of jurisdictions, respectively)
more often than small jurisdictions (estimated at 1 percent). DREs were
the leading choice among voting methods for both large and medium
jurisdictions that planned to acquire voting systems before the 2006
general election (an estimated 34 percent of jurisdictions in both size
groups).

The Contextual Role and Performance Characteristics of Electronic Voting Systems
Are Important to Understanding Their Use

Voting systems are one facet of a multifaceted, continuous elections
process that involves the interplay of people, processes, and technology.
All levels of government--federal, state, and local--share
responsibilities for aspects of elections and voting systems. Moreover,
effective performance of these systems is a product of effective system
life cycle management, which includes systems definition, development,
acquisition, operations, testing, and management. Such performance can be
viewed in terms of several characteristics, such as security, reliability,
ease of use, and cost effectiveness.

Despite Their Vital Role, Voting Systems Are Only One Aspect of the Larger
Election Process

Voting systems represent one of many important components in the overall
election process. This process involves all levels of government and is
made up of several stages, with each stage consisting of the interplay of
people, processes, and technology.

At the federal level, Congress has authority under the Constitution to
regulate the administration of presidential and congressional elections
and to enforce prohibitions against specific discriminatory practices in
all elections--federal, state, and local.^8 It has passed legislation
affecting the administration of state elections that addresses voter
registration,^9 absentee voting,^10 accessibility provisions for the
elderly and handicapped,^11 and prohibitions against discriminatory
practices.^12 Congress does not have general constitutional authority over
the administration of state and local elections.

At the state level, the states are responsible for the administration of
both their own elections and federal elections. States regulate the
election process, including, for example, adoption of voting system
standards, testing of voting systems, ballot access, registration
procedures, absentee voting requirements, establishment of voting
locations, provision of Election Day workers, and counting and
certification of the vote. As we have reported, the U.S. election process
can be seen as an assemblage of 51 somewhat distinct election
systems--those of the 50 states and of the District of Columbia.

^8For more information on the role of the federal government in the
administration of
elections, see GAO, Elections: The Scope of Congressional Authority in
Election Administration, GAO- 01-470 (Washington, D.C.: Mar. 13, 2001).

^9National Voter Registration Act of 1993, commonly known as the "Motor
Voter" Act; 42 U.S.C. 1973gg to 1973gg-10.

^10Uniformed and Overseas Citizens Absentee Voting Act (1986); 42 U.S.C.
1973ff to 1973ff-6.

^11Voting Accessibility for the Elderly and Handicapped Act (1984); 42
U.S.C. 1973ee to 1973ee-6.

^12Voting Rights Act of 1965, 42 U.S.C. 1973 to 1973bb-1.

Further, although election policy and procedures are legislated primarily
at the state level, states typically have decentralized this process so
that the details of administering elections are carried out at the city or
county levels, and voting is done at the local level. This is important
because local election jurisdictions number more than 10,000 and their
size varies enormously--from a rural county with about 200 voters to a
large urban county such as Los Angeles County, where the total number of
registered voters for the 2000 elections exceeded the registered voter
totals in 41 states.

The size and demographics of a voting jurisdiction significantly affects
the complexity of planning and conducting the election, as does the method
used to cast and count votes. For example, jurisdictions using DRE systems
may need to manage the electronic transmission of votes or vote counts,
while jurisdictions using optical scan technology need to manage the
transfer of the paper ballots this technology reads and tabulates.
Jurisdictions using optical scan technology may also need to manage
electronic transmissions if votes are counted at various locations and
totals are electronically transmitted to a central tally point. No matter
what technology is used, jurisdictions may need to provide ballot
translations; however, the logistics of printing paper materials in a
range of languages, as would be required for optical scan technology, is
different from the logistics of programming translations into DRE units.

Some states do have statewide election systems so that every voting
jurisdiction uses similar processes and equipment, but others do not. For
instance, we reported in 2001 that in Pennsylvania, local election
officials told us that there were 67 counties and consequently 67
different ways of handling elections.^13 In some states, such as Georgia,
state law prescribes the use of common voting technology throughout the
state while in other states, local election officials generally choose the
voting technology to be used in their precincts, often from a list of
state-certified options.

Regardless of levels of government, however, election administration is a
year-round activity, involving varying sets of people performing the
activities of each stage of the election process. These stages generally
consist of the following:

^13 [29]GAO-02-3.

           o Voter registration. Among other things, local election officials
           register eligible voters and maintain voter registration lists,
           including updates to registrants' information and deletions of the
           names of registrants who are no longer eligible to vote.
           o Absentee and early voting. This type of voting allows eligible
           persons to vote in person or by mail before Election Day. Election
           officials must design ballots and other systems to permit this
           type of voting and educate voters on how to vote by these methods.
           o Election Day vote casting. Election administration includes
           preparation before Election Day, such as local election officials
           arranging for polling places, recruiting and training poll
           workers, designing ballots, and preparing and testing voting
           equipment for use in casting and tabulating votes, as well as
           Election Day activities, such as opening and closing polling
           places and assisting voters in casting their votes.
           o Vote counting. At this stage, election officials tabulate the
           cast ballots, determine whether and how to count ballots that
           cannot be read by the vote counting equipment, certify the final
           vote counts, and perform recounts, if required.

As shown in figure 4, each stage of an election involves people,
processes, and technology.

Figure 4: Stages of Election Process

Electronic voting systems are primarily involved in the last three stages,
during which votes are recorded, cast, and counted. However, the type of
system that a jurisdiction uses may affect earlier stages. For example, in
a jurisdiction that uses optical scan systems, paper ballots like those
used on Election Day may be mailed in the absentee voting stage. On the
other hand, a jurisdiction that uses DRE technology would have to make a
different provision for absentee voting.

Management of Electronic Voting System Performance Is a Continuous Process

The performance of any information technology system, including electronic
voting systems, is heavily influenced by a number of factors, including
how well the system is defined, developed, acquired, tested, and
implemented.

Like any information technology product, a voting system starts with the
explicit definition of what the system is to do and how well it is to do
it. These requirements are then translated into design specifications that
are used to develop the system. Electronic voting systems are typically
developed by vendors and then purchased as commercial off-the-shelf (COTS)
products and implemented by state and local election administrators.
During the development, acquisition, and implementation of the systems, a
range of tests are performed and the process is managed to ensure
performance expectations are met. Together, these activities form a voting
system life cycle (see figure 5).

Figure 5: Simplified Voting System Life Cycle

Unless voting systems are properly managed throughout their life cycle,
this one facet of the election process can significantly undermine the
integrity of the whole.

Standards. Voting system standards define the functional and performance
requirements that must be met and thus provide the baseline against which
systems can be developed and tested. They also specify how the systems
should be implemented and operated. Voting system standards apply to
system hardware, software, firmware, and documentation, and they span
prevoting,  voting, and postvoting activities. They address, for example,
requirements relating to system security; system reliability (accuracy and
availability); system auditability; system storage and maintenance; and
data retention and transportation. In addition to national standards, some
states and local jurisdictions have specified their own voting system
requirements.

Development. Product development is performed by the voting system vendor.
Product development includes further establishing system requirements,
designing the system architecture, developing software, integrating
hardware and software components, and testing the system.

Acquisition. Voting system acquisition activities are performed by state
and local governments and include publishing a request for proposal,
evaluating proposals, choosing a voting system method, choosing a vendor,
writing and administering contracts, and testing the acquired system.

Operations. Operation of voting systems is typically the responsibility of
local jurisdictions. These activities include setting up systems before
voting, vote capture and counting during elections, recounts and system
audits after elections, and storage of systems between elections. Among
other things, this phase includes activities associated with the physical
environments in which the system operates. These include ensuring the
physical security of the polling place and voting equipment and
controlling the chain of custody for voting system components and
supplies. The operations phase also includes monitoring of the election
process by use of system audit logs and backups, and the collection,
analysis, reporting, and resolution of election problems.

Testing. As noted, testing is conducted by multiple entities throughout
the life cycle of a voting system. Voting system vendors conduct testing
during system development. National testing of systems is conducted by
accredited independent testing authorities. Some states conduct testing
before acquiring a system to determine how well it meets the specified
performance parameters, or states may conduct certification testing to
ensure that a system performs as specified by applicable laws and
requirements. Once a voting system is delivered by the vendor, states and
local jurisdictions may conduct acceptance testing to ensure that the
system satisfies requirements. Finally, local jurisdictions typically
conduct logic and accuracy tests prior to each election and sometimes
subject portions of the system to parallel testing during each election.

Management. Management processes ensure that each life cycle phase
produces desirable outcomes and are conducted by the organizations
responsible for each life cycle phase. Voting system vendors manage the
development phase, while states and/or local jurisdictions manage the
acquisition and operations phases. Typical management activities that span
the system life cycle include planning, configuration management, system
performance review and evaluation, problem tracking and correction, human
capital management, and user training. Management responsibilities related
to security and reliability include program planning, disaster recovery
and contingency planning, definition of security roles and
responsibilities, configuration management of voting system hardware and
software, and poll worker security training.

Electronic Voting System Performance Can Be Judged on Several Attributes

Although the debate concerning electronic voting systems is primarily
focused on security, other performance attributes are also relevant, such
as reliability and ease of use, and cost. Each of these attributes is
described here.

Security. Election officials are responsible for establishing and managing
security and privacy controls to protect against threats to the integrity
of elections.^14 Threats to election results and voter confidentiality
include potential modification or loss of electronic voting data; loss,
theft, or modification of physical ballots; and unauthorized access to
software and electronic equipment. Different types of controls can be used
to counter these threats. Physical access controls are important for
securing voting equipment, vote tabulation equipment, and ballots.
Software access controls (such as passwords and firewalls^15) are
important for limiting the number of people who can access and operate
voting devices, election management software, and vote tabulation
software. In addition, physical screens around voting stations, and poll
workers preventing voters from being watched or coerced while voting are
important to protect the privacy and confidentiality of the vote.

Reliability.  Ensuring the reliability of votes being recorded and tallied
is an essential attribute of any voting equipment and depends to a large
degree on the accuracy and availability of voting systems. Without such
assurance, both voter confidence in the election and the integrity and
legitimacy of the outcome of the election are at risk. The importance of
an accurate vote count increases with the closeness of the election. Both
optical scan and DRE systems are claimed to be highly accurate. Although
voting equipment may be designed and developed to count votes as recorded
with 100 percent accuracy, how well the equipment counts votes as intended
by voters is a function not only of equipment design, but also of how
procedures are followed by election officials, technicians, and voters. It
is also important to limit system down time so that polling places can
handle the volume of voter traffic.

^14We have described an effective security program as including, at a
minimum, (1) assigning responsibility for security, (2) assessing security
risks and vulnerabilities and implementing both manual and
technology-based security measures to prevent or counter these risks, and
(3) periodically reviewing the controls to ensure their appropriateness.
For more information, see GAO, Executive Guide: Information Security
Management, GAO/AIMD-98-68 (Washington, D.C.: May 1998).

^15A firewall is a hardware or software component that protects computers
or networks from attacks by outside network users by blocking and checking
all incoming traffic.

Ease of Use.  Ease of use (or user friendliness) depends largely on how
voters interact with the voting system, physically and intellectually.
This interaction, commonly referred to as the human/machine interface, is
a function of the system design and how it has been implemented. Ease of
use depends on how well jurisdictions design ballots and educate voters on
the use of the equipment. A voting system's ease of use affects accuracy
(i.e., whether the voter's intent is captured), and it can also affect the
efficiency of the voting process (confused voters take longer to vote).
Accessibility by diverse types of voters, including those with
disabilities, is a further aspect of ease of use.

Cost.  For a given jurisdiction, the particular cost associated with an
electronic voting system will depend on the requirements of the
jurisdiction as well as the particular equipment chosen. Voting equipment
costs vary among types of voting equipment and among different
manufacturers and models of the same type of equipment. Some of these
differences can be attributed to differences in what is included in the
unit cost. In addition to the equipment unit cost, an additional cost for
jurisdictions is the software that operates the equipment, prepares the
ballots, and tallies the votes (and in some cases, prepares the election
results reports). Other factors affecting the acquisition cost of voting
equipment are the number and types of peripherals required. Once
jurisdictions acquire the voting equipment, they also incur the cost to
operate and maintain it, which can vary considerably.

Widespread Concerns about Electronic Voting Systems Have Been Reported

Election officials, computer security experts, citizen advocacy groups,
and others have raised significant security and reliability concerns with
electronic voting systems, citing, for example, inadequacies in standards,
system design and development, operation and management activities, and
testing. In 2005, we examined the range of concerns raised by these groups
and aligned them with their relevant life cycle phases. We also examined
EAC's efforts related to these concerns. Furthermore, we identified key
practices that each level of government should implement throughout the
voting system life cycle in order to improve security and reliability.^16

The aspects of the voting system life cycle phases are
interdependent--that is, a problem experienced in one area of the life
cycle will likely affect other areas. For example, a weakness in system
standards could result in a poorly designed and developed system, which
may not perform properly in the operational phase. State and local
jurisdictions have documented instances when their electronic voting
systems exhibited operational problems during elections. Such failures led
to polling place delays, disruptions, and incorrect vote tabulations.

In reviewing the reported concerns, we have explained that many of the
security and reliability concerns involved vulnerabilities or problems
with specific voting system makes and models or unique circumstances in a
specific jurisdiction's election, and there is a lack of consensus among
elections officials, computer security experts, and others on the
pervasiveness of the concerns. We concluded in 2005 that these concerns
have caused problems with recent elections, resulting in the loss and
miscount of votes. In light of the demonstrated voting system problems;
the differing views on how widespread these problems are; and the
complexity of assuring the accuracy, integrity, confidentiality, and
availability of voting systems throughout their life cycles, we stated
that the security and reliability concerns merit the focused attention of
federal, state, and local authorities responsible for election
administration.

^16 GAO-05-956.

Inadequate National Standards

Appropriately defined and implemented standards for system functions and
testing processes are essential to ensuring the security and reliability
of voting systems across all phases of the elections process. States and
local jurisdictions face the challenge of adapting to and consistently
applying appropriate standards and guidance to address vulnerabilities and
risks in their specific election environments. The national standards are
voluntary--meaning that states are free to adopt them in whole or in part
or reject them entirely.

The Federal Election Commission (FEC) issued a set of voluntary voting
systems standards in 1990 and revised them in 2002. These standards
identify requirements for electronic voting systems. Computer security
experts and others criticized the 2002 voting system standards for not
containing requirements sufficient to ensure secure and reliable voting
systems. Common concerns with the standards involved their vague and
incomplete security provisions, inadequate provisions for some commercial
products and networks, and inadequate documentation requirements.

In December 2005, EAC issued the Voluntary Voting System Guidelines, which
include additions and revisions for system functional requirements,
performance characteristics, documentation requirements, and test
evaluation criteria for the national certification of voting systems.
These guidelines promote security measures that address gaps in prior
standards and are applicable to more modern technologies, such as controls
for software distribution and wireless operations.

As we previously reported, the 2005 Voluntary Voting System Guidelines do
not take effect until December 2007. Moreover, this version of the
standards does not comprehensively address voting technology issues. For
instance, they do not address COTS devices (such as card readers,
printers, or personal computers) or software products (such as operating
systems or database management systems) that are used in voting systems
without modification. This is significant because computer security
experts have raised concerns about a provision in the prior voting system
standards that exempted unaltered COTS software from testing and about
voting system standards that are not sufficient to address the weaknesses
inherent in telecommunications and networking services. Specifically,
vendors often use COTS software in their electronic voting systems,
including operating systems. Security experts note that COTS software
could contain defects, vulnerabilities, and other weaknesses that could be
carried over into electronic voting systems, thereby compromising their
security. Regarding telecommunications and networking services, selected
computer security experts believe that relying on any use of
telecommunications or networking services, including wireless
communications, exposes electronic voting systems to risks that make it
difficult to adequately ensure their security and reliability--even with
safeguards such as encryption and digital signatures in place.

As states and jurisdictions move to a more integrated suite of election
systems, proactive efforts to establish standards in such areas will be
essential to addressing emerging technical, security, and reliability
interactions among systems and managing risks in this dynamic election
environment. However, the 2005 guidelines do not address the emerging
trends in election systems, such as the integration of registration
systems with voting systems.

In light of this and other weaknesses in the standards, we reported in
2005 that EAC did not yet have detailed plans in place for addressing
these deficiencies. Accordingly, we recommended that EAC collaborate with
NIST and the Technical Guidelines Development Committee to define specific
tasks, measurable outcomes, milestones, and resource needs required to
improve the standards. To its credit, EAC agreed with our recommendation,
recognizing that more work was needed to further develop the technical
guidelines. Accordingly, it stated that it planned to work with NIST to
plan and prioritize its standards work within its scarce resources.

Inadequate System Design and Development

Multiple reports, including several state-commissioned technical reviews
and security assessments, raised concerns about the design and development
of secure and reliable electronic voting systems. Among other things, weak
embedded security controls and audit trail design flaws were two major
areas of concern.

           o Weak system security controls. Some electronic voting systems
           reportedly have weak software and hardware security controls.
           Regarding software controls, many security examinations reported
           flaws in how controls were implemented in some DRE systems to
           prevent unauthorized access. For example, one model failed to
           password-protect the supervisor functions controlling key system
           capabilities; another relied on an easily guessed password to
           access these functions. If exploited, these weaknesses could
           damage the integrity of ballots, votes, and voting system software
           by allowing unauthorized modifications. Regarding physical
           hardware controls, several recent reports found that certain DRE
           models contained weaknesses in controls designed to protect the
           system. For instance, reviewers were concerned that a particular
           model of DRE was set up in such a way that if one machine was
           accidentally or intentionally unplugged from the others, voting
           functions on the other machines in the network would be disrupted.
           In addition, reviewers found that the switches used to turn a DRE
           system on or off, as well as those used to close the polls on a
           particular DRE terminal, were not protected.
           o Design flaws in developing voter-verified paper audit trails. 
           Establishing a voter-verified paper audit trail involves adding a
           paper printout to a DRE system so that a voter can review and
           verify his or her ballot. Some citizen advocacy groups, security
           experts, and elections officials advocate these audit trails as a
           protection against potential DRE flaws. However, other election
           officials and researchers have raised concerns about potential
           reliability and security flaws in the design of systems using
           voter-verified paper audit trails. If voting system mechanisms for
           protecting the paper audit trail were inadequate, an insider could
           associate voters with their individual paper ballots and votes,
           particularly if the system stored voter-verified ballots
           sequentially on a continuous roll of paper. If not protected, such
           information could breach voter privacy and confidentiality.

Inadequate System Operation and Management Activities

Several reports raised concerns about the operational practices of local
jurisdictions and the actual performance of their respective electronic
voting systems during elections. These include incorrect system
configurations, inadequate security management programs, and poor
implementation of security procedures.

           o Incorrect system configuration.  Some state and local election
           reviews have documented cases in which local governments did not
           properly configure their voting systems. These improper
           configurations resulted in voters being unable to vote in certain
           races or their votes not being captured correctly by the voting
           system. 
           o Poor version control of system software.  Security experts and
           some election officials expressed concern that the voting system
           software installed at the local level may not be the same as what
           was qualified and certified at the national or state levels. These
           groups raised the possibilities that either intentionally or by
           accident, voting system software could be altered or substituted,
           or that vendors or local officials might install untested or
           uncertified versions of voting systems, knowingly or unknowingly.
           As a result, potentially unreliable or malicious software might be
           used in elections.
           o Inadequate security management programs.  Several technical
           reviews found that states did not have effective information
           security management plans in place to oversee their electronic
           voting systems. The reports noted that key managerial functions
           were not in place, including (1) providing appropriate security
           training, (2) ensuring that employees and contractors had proper
           certifications, (3) ensuring that security roles were well defined
           and staffed, and (4) ensuring that pertinent officials correctly
           set up their voting system audit logs and require them to be
           reviewed. 
           o Poor implementation of security procedures. Several reports
           indicated that state and local officials did not always follow
           security procedures. For example, reports found that a regional
           vote tabulation computer was connected to the Internet and that
           local officials had not updated it with several security patches,
           thus needlessly exposing the system to security threats. In
           addition, several reports indicated that some state and local
           jurisdictions did not always have procedures in place to detect
           problems with their electronic voting systems such as ensuring the
           number of votes cast matched the number of signatures on precinct
           sign-in sheets.

Inadequate Testing

Security experts and some election officials have expressed concerns that
the tests performed by independent testing authorities and state and local
election officials do not adequately assess electronic voting systems'
security and reliability. These concerns are intensified by what some
perceive as a lack of transparency in the testing process.

           o Inadequate security testing. Many computer security experts
           expressed concerns with weak or insufficient system testing,
           source code reviews, and penetration testing. To illustrate their
           concerns, they pointed to the fact that most of the systems that
           exhibited the weak security controls previously cited had been
           nationally certified after testing by an independent testing
           authority. Security experts and others point to this as an
           indication that both the standards and the testing program are not
           rigorous enough with respect to security.
           o Lack of transparency in the testing process.  Security experts
           and some elections officials have raised concerns about a lack of
           transparency in the testing process. They note that the test plans
           used by the independent testing authorities, along with the test
           results, are treated as protected trade secrets and thus cannot be
           released to the public. Critics say that this lack of transparency
           hinders oversight and auditing of the testing process. This in
           turn makes it harder to determine the actual capabilities,
           potential vulnerabilities, and performance problems of a given
           system. Despite assertions by election officials and vendors that
           disclosing too much information about an electronic voting system
           could pose a security risk, one security expert noted that a
           system should be secure enough to resist even a knowledgeable
           attacker.

Variability and Weaknesses in State and Local Approaches to Voting System
Standards, Testing, Operations, and Performance Measurement

In 2006, we reported on state and local government experiences in
conducting the 2004 national election.^17 Regarding voting systems, states
and jurisdictions' responses to our surveys showed that differing versions
of the national voting system standards were used (not always the most
current version); voting system life cycle management practices were not
consistently implemented; and certain types of system testing were not
widely performed. Moreover, jurisdictions reported that they did not
consistently monitor the performance of their systems, which is important
for determining whether election needs, requirements, and expectations are
met and for taking corrective actions when they are not.

States' Use of Standards Varied

States and jurisdictions reported that they applied a variety of voting
system standards, some of which were no longer current. Specifically, 44
states and the District of Columbia reported that they were requiring
local jurisdictions' voting systems to be used for the first time in the
November 2006 general election to comply with the national voting system
standards.^18 However, these states were not all using the same version of
the standards. This is troublesome because the later versions of the
standards are more stringent than the earlier versions in various areas,
including security.

More specifically, 28 of the 44 states and the District of Columbia
reported that voting systems to be used for the first time in the 2006
election comply with the 2002 voting system standards. Nine of these 28
states would also require their jurisdictions to apply the 1990 federal
standards to new voting systems and 4 of the 28 would also require
jurisdictions to use the 2005 voting system standards, which were in draft
version at the time of our survey. (One other state also expected to apply
the 2005 voting system standards.) Ten other of the 44 states reporting
said that they expected to use hybrid standards that were based on one or
more versions of the national standards, without specifying the
composition of their hybrid, and 4 states planned to use the national
standards in 2006, but did not specify a version. (Five states responded
that they did not require their voting systems to comply with any version
of the national standards or had not yet made a decision on compliance
with the standards for 2006. One state did not respond.)

^17GAO-06-450.

^18 To obtain information from state election officials on any changes
made in selected state statutory requirements since the 2000 election and
other changes made including actions taken to implement HAVA, we conducted
a Web-based survey of the 50 states and the District of Columbia. For more
details about this survey, see Appendix V in GAO-06-450.

Jurisdictions Varied Widely in Applying Security Practices

Local jurisdictions varied widely in the nature and extent of their voting
system security efforts and activities during the 2004 election. Our
research on recommended security practices shows that effective system
security management involves having, among other things, (1) defined
policies governing such system controls as authorized functions and access
and documented procedures for secure normal operations and incident
management; (2) documented plans for implementing policies and procedures;
(3) clearly assigned roles and responsibilities for system security; and
(4) verified use of technical and procedural controls designed to reduce
the risk of disruption, destruction, or unauthorized modification of
systems and their information. Jurisdictions' efforts in each of these
areas for the November 2004 general election are discussed here.

Policies and procedures. Many jurisdictions reported having written
policies and procedures for certain aspects of security related to their
voting systems, but others did not. Written security policies were more
prevalent among large jurisdictions (an estimated 65 percent) than small
jurisdictions (an estimated 41 percent). An estimated one-fifth of
jurisdictions reported that they did not have written policies and
procedures in place for transporting ballots or electronic memory, storing
ballots, or electronic transmission of voted ballots to ensure ballot
security. In addition, some jurisdictions that we visited had published
detailed voting system security policies and procedures that included such
topics as network security policies for election tabulation, procedures
for securing and protecting election equipment and software, testing
voting equipment to ensure accurate recording of votes, and disaster
recovery plans, while others omitted these topics.^19 Some jurisdictions
also took additional steps to ensure that election workers had access to,
and were trained in, the contents of the policies and procedures for
securing ballots and voting equipment.

Implementation plans. Election officials in only 8 of the 28 jurisdictions
that we visited told us that they had written plans for implementing
security aspects of their voting systems and processes. Moreover, the
contents of plans we obtained from local jurisdictions varied widely. One
of the jurisdiction's security plans covered most aspects of the voting
process, from ballot preparation through recount, while another plan was
limited to the security of its vote-tallying system in a stand-alone
environment. Of the 5 plans we reviewed, 2 covered almost all of the 8
security topics in our review that included risk assessment, physical
controls, awareness training, and incident response, while the others
covered just one or two topics.

Roles and responsibilities. In addition, security management roles and
responsibilities for the 2004 general election were not consistently
assigned.^20 According to survey responses, security responsibilities
primarily fell to local election officials (estimated at 67 percent) for
the 2004 general election, although state officials (estimated at 14
percent) and other entities (e.g., independent consultants and vendors,
estimated at 24 percent) were also

^19 To obtain a more detailed understanding of the stages of the election
process, challenges associated with it in local jurisdictions, and how
local election officials address those challenges, we visited and
interviewed officials in a nonprobability sample of 28 local election
jurisdictions in 14 states nationwide. For more details about these
visits, see Appendix V in GAO-06-450.

^20 From our local jurisdiction survey, we estimate that 90 percent of all
jurisdictions (excluding those that used only hand-counted paper ballots
on Election Day) specifically assigned responsibility for voting system
security. Jurisdictions that used only hand-counted paper ballots on
Election Day were excluded from this survey question.

Figure 7: Estimated Use of Security Controls by Local Jurisdictions in the
2004 General Election, by Jurisdiction Size

Note: More than one group may have been identified with security
responsibilities.

^aThe difference between small jurisdictions and medium jurisdictions is
statistically significant. The 95 percent confidence interval for small
jurisdictions is plus or minus 8 percentage points.

^bThe 95 percent confidence interval for small jurisdictions is plus or
minus 8 percentage points.

^cThe differences between small jurisdictions and both medium and large
jurisdictions are statistically significant. The 95 percent confidence
interval for small jurisdictions is plus or minus 8 percentage points.

Among the jurisdictions that we visited, election officials reported that
various security measures were in use during the 2004 general election to
safeguard voting equipment, ballots, and votes before, during, and after
the election. However, the measures were not uniformly reported by
officials in these jurisdictions, and officials in most jurisdictions
reported that they did not have a security plan to govern the scope,
nature, and implementation of these measures or other aspects of their
security program. The security controls most frequently cited by officials
for the jurisdictions that we visited were locked storage of voting
equipment and ballots and monitoring of voting equipment. Other security
measures mentioned during our visits included testing voting equipment
before, during, or after the election to ensure that the equipment was
accurately tallying votes; planning and conducting training on security
issues and procedures for elections personnel; and video surveillance of
stored ballots and voting equipment. Table 2 summarizes the types and
frequency of security measures reported by election officials in the
jurisdictions we visited.

Table 2: Security Controls Reportedly Used in the 2004 General Election by
Election Officials in Jurisdictions Visited by GAO

Reported security control                          Number of jurisdictions 
Locked/sealed storage of voting equipment and                25            
ballots ^a                                                                 
Monitoring of voting equipment ^a                            14            
Encrypted ballots or election results ^a                     10            
Security plans ^a                                             8            
Testing of voting equipment ^a                                7            
Control of voting machine memory cards by precinct            6            
personnel during elections ^a                                              
Video surveillance for voting equipment or ballots            5            
Security training                                             4            

Source: GAO analysis of interviews from local jurisdictions we visited.

^aOne or more jurisdictions we visited indicated this security control was
not applicable because of the voting method used.

Voting systems that can be remotely accessed introduce additional security
challenges. Based on survey responses, we estimated that a small
percentage of local jurisdictions (10 percent) provided remote access to
their voting systems for one or more categories of personnel--local
election officials, state election officials, vendors, or other
parties.^24 Some of the jurisdictions that provided this access described
a variety of protections to mitigate the risk of unauthorized remote
access, including locally controlled passwords, passwords that change for
each access, and local control of communications connections. However, the
percentage of jurisdictions with remote access may actually be higher
because 7 to 8 percent of jurisdictions did not know if remote access was
available for their systems.

Some Types of Testing Were Not Commonly Performed

To ensure that voting systems perform as intended, the systems must be
effectively tested. Voting system test and evaluation can be grouped into
various types, or stages: certification testing (national level),
certification testing (state level), acceptance testing, readiness
testing, parallel testing, and postelection voting system audits. Each of
these tests has a specific purpose and is conducted at the national,
state, or local level at a particular time in the election cycle. Table 3
summarizes these types of tests.

^24 Jurisdictions that used only hand-counted paper ballots on Election
Day were excluded from this survey question.

Table 3: Types of Testing and Evaluation for Voting Systems, with Common
Time Frames and Responsibilities

Test type        Purpose                    When conducted Responsibility  
Certification    To verify compliance of    Prior to (or   Federal         
(national)^a     voting equipment with      as a condition authorities and 
                    federal standards          of) system     independent     
                                               acceptance     laboratories^b  
Certification    To validate compliance of  Before         State election  
(state)          voting equipment with      election       authorities     
                    state-specific                                            
                    requirements                                              
Acceptance       To verify that voting      Before         State or local  
                    equipment delivered by a   election       election        
                    vendor meets state or                     authorities     
                    local requirements                                        
Readiness (logic To verify that voting      Before         Local election  
and accuracy)    equipment is functioning   election       authorities     
                    properly, usually by                                      
                    confirming that                                           
                    predictable outputs are                                   
                    produced from pre-defined                                 
                    inputs^c                                                  
Parallel         To verify accurate         During         State or local  
                    performance of voting      election       election        
                    equipment through random                  authorities     
                    selection and systematic                                  
                    evaluation of operational                                 
                    equipment                                                 
Audit            To review and reconcile    After election State or local  
                    election records to                       election        
                    confirm correct conduct of                authorities     
                    an election or uncover                                    
                    evidence of problems with                                 
                    voting equipment or                                       
                    election processes                                        

Source: GAO analysis based on GAO-02-3 and GAO-05-956.

^aWith the enactment of HAVA in 2002, responsibility for overseeing
national testing of voting systems and certifying those that met federal
standards was assigned to the EAC in HAVA S 231(a)(1) (codified at 42
U.S.C. S 15371(a)(1)). The EAC assumed this responsibility in August 2005,
when it was transferred from the National Association of State Election
Directors (NASED). Under NASED, national testing against federal standards
was called "qualification testing".

^bRefers to the U.S. Election Assistance Commission and testing
laboratories accredited by them as provided for in HAVA S 231 (codified at
42 U.S.C. S 15371).

^cReadiness testing that is conducted to confirm the proper functioning of
election equipment on Election Day just before the polls open is sometimes
called verification testing.

For the November 2004 general election, voting system testing was
conducted for almost all voting systems, but the types and content of the
testing performed varied considerably. According to survey responses, most
states and local jurisdictions employed national and state certification
testing and readiness testing to some extent, but the criteria used in
this testing were highly dependent on the state or jurisdiction. Also,
many, but not all, states and jurisdictions conducted acceptance testing
of both newly acquired systems and those undergoing changes or upgrades.
In contrast, relatively few states and jurisdictions conducted parallel
testing during elections or audits of voting systems following elections.
State and local responses to our surveys are summarized here relative to
each type of testing.

National certification. Most states continued to require that voting
systems be nationally tested and certified. For voting systems being used
for the first time in the 2004 general election, national certification
testing was almost always uniformly required. In particular, 26 of 27
states using DRE for the first time in this election, as well as the
District of Columbia, required their systems to be nationally certified,
while 9 of the 10 states using punch card equipment for the first time,
and 30 of 35 states and the District of Columbia using optical scan
equipment for the first time, reported such requirements. However, for the
2004 general election, we estimated that 68 percent of jurisdictions did
not know whether their respective systems were nationally certified. This
uncertainty surrounding the certification status of a specific version of
voting system at the local level underscores our concern that even though
voting system software may have been qualified and certified at the
national or state levels, software changes and upgrades performed at the
local level may not be.

State certification. For the November 2004 general election, 42 states and
the District of Columbia reported that they required state certification
of voting systems. Seven of these states purchased voting systems at the
state level for local jurisdictions. Officials for the remaining states
and the District of Columbia reported that responsibility for purchasing a
state-certified voting system rested with the local jurisdiction. While
state certification requirements often included national testing as well
as confirmation of functionality for particular ballot conditions, some
states also required additional features such as construction quality,
transportation safety, and documentation. Among the remaining 8 states
that did not require state certification, officials described other
mechanisms to address the compliance of voting equipment with
state-specific requirements, such as a state approval process or
acceptance of voting equipment based on federal certification.

For the 2006 general election, 44 states reported that they would have
requirements for certification of voting systems, 2 more states than for
the 2004 general election. Of the 44, all but 1 expected to conduct the
certification themselves; the remaining state reported that it would rely
on results from a national independent testing authority to make its
certification decision. In addition, 18 of the 43 states planned to
involve a national testing laboratory in their certification process.

Acceptance testing.  With regard to acceptance testing of new voting
systems, 26 states and the District of Columbia reported that
responsibility for such testing was assigned to either the state or local
level for the 2004 general election. Specifically, 8 states and the
District of Columbia reported that they had responsibility for performing
acceptance testing, 15 states required local jurisdictions to perform such
testing, and 3 states reported that requirements for acceptance testing
existed at both the state and local levels. Twenty-two states either did
not require such testing or did not believe that such testing was
applicable to them. (Two states did not know their acceptance testing
requirements for the 2004 election.)

In addition, more states required that acceptance testing be performed for
changes and upgrades to existing systems than they did for new systems--30
states in all and the District of Columbia. Similarly, election officials
at a majority of the local jurisdictions that we visited told us that they
conducted some type of acceptance testing for newly acquired voting
equipment, although they described a variety of approaches to performing
acceptance testing. For example, the data used for testing could be
vendor-supplied, developed by election officials, or both, and could
include system initialization, logic and accuracy, and tamper resistance.
Other steps, such as diagnostic tests, physical inspection of hardware,
and software configuration checks, were also mentioned as testing
activities by local election officials. Further, election officials from 3
jurisdictions that we visited said that vendors were heavily involved in
designing and executing the acceptance tests, while officials from another
jurisdiction that we visited said that vendors contributed to a portion of
their testing. In 2 jurisdictions, officials said that acceptance tests
were conducted at a university center for elections systems.

Readiness testing. Almost all states (49) and the District of Columbia
reported that they performed readiness testing of voting systems at the
state level, the local level, or both (one state did not require readiness
testing). Most states (37) required local jurisdictions to perform
readiness testing. However, 7 states reported that they performed their
own readiness testing for the 2004 general election in addition to local
testing. Five states and the District of Columbia reported that they had
no requirements for local jurisdictions to perform readiness testing but
conducted this testing themselves.

State laws or regulations in effect for the 2004 election typically had
specific requirements for when readiness testing should be conducted and
who was responsible for testing, sometimes including public demonstrations
of voting system operations. We found that most jurisdictions conducted
readiness testing, also known as logic and accuracy testing, for both the
2000 and 2004 general elections. Election officials in all of the local
jurisdictions we visited following the 2004 election reported that they
conducted readiness testing on their voting equipment using one or more
approaches, such as diagnostic tests, integration tests, mock elections,
and sets of test votes, or a combination of approaches.

Security testing.  Security testing was reportedly performed by 17 states
and the District of Columbia for the voting systems used in the 2004
general election, and 7 other states reported that they required local
jurisdictions to conduct such testing. The remaining 22 states said that
they did not conduct or require system security testing. (Three states
reported that security testing was not applicable for their voting
systems.) Moreover, we estimated that at least 19 percent of local
jurisdictions nationwide (excluding jurisdictions that reported that they
used paper ballots) did not conduct security testing for the systems they
used in the November 2004 election. Although jurisdiction size was not a
factor in whether security testing was performed, the percentage of
jurisdictions performing security testing was notably higher when the
predominant voting method was DRE (63 percent^25) and lower for
jurisdictions where the predominant method was precinct count optical scan
(45 percent).^26

25 The 95 percent confidence interval for DRE is +14/-15 percentage
points.

Parallel testing. Parallel testing was not widely performed by local
jurisdictions in the 2004 general election, although 7 states reported
that they performed parallel testing of voting systems on Election Day,
and another 6 states reported that this testing was required by local
jurisdictions.^27 We estimated that 2 percent of jurisdictions using
electronic systems for at least some of their voting conducted parallel
testing for the 2004 general election.^28 Large and medium jurisdictions
primarily performed this type of testing (7 percent and 4 percent of
jurisdictions, respectively). The percentage of small jurisdictions
performing this type of testing was negligible (0 percent). Election
officials in 2 of the 28 jurisdictions that we visited told us that they
performed parallel testing, either at the state level or at the local
jurisdiction. In both cases, the tests were conducted on voting equipment
for which security concerns had been raised in another state's voting
equipment test report. Local officials who told us that parallel testing
was not performed on their voting systems attributed this to the absence
of parallel testing requirements, a lack of sufficient voting equipment to
perform these tests, or the unnecessary nature of parallel testing because
of the stand-alone operation of their systems.

Post-election audits. Less than one-half of the states (22) and the
District of Columbia reported that they performed postelection voting
system audits for the 2004 general election. Specifically, 4 states and
the District of Columbia reported that they conducted postelection audits
of voting systems, 16 states required that audits of voting systems be
conducted by local jurisdictions, and 2 states reported that audits of
voting systems were performed at both the state and local levels.
Moreover, state laws or regulations in effect for the 2004 general
election varied in when and how these audits were to be conducted.

^26 The 95 percent confidence interval for precinct count optical scan is
+/-9 percentage points. The difference between the percentages of
jurisdictions performing security testing on DRE and central count optical
scan was not statistically significant.

^27Both EAC's Best Practices Tool Kit and the 2005 Voluntary Voting System
Guidelines recommend development of parallel testing procedures for all
types of automated voting equipment.

^28We estimated that 91 percent of jurisdictions considered parallel
testing to be not applicable.

We estimated that 43 percent of jurisdictions that used voting systems for
at least some of their voting conducted postelection voting system audits.
This practice was much more prevalent at large and medium jurisdictions
(62 percent and 55 percent, respectively) than small jurisdictions (34
percent).^29 We further estimated that these voting system audits were
conducted more frequently in jurisdictions with central count optical scan
voting methods (54 percent) than they were in jurisdictions with precinct
count optical scan voting methods (35 percent).

Jurisdictions Did Not Consistently Monitor Voting System Performance

It is important that performance be measured during system operation. As
we reported in 2001 and 2006, measuring how well voting systems perform
during a given election allows local officials to better position
themselves for ensuring that elections are conducted properly. Such
measurement also provides the basis for knowing where performance needs,
requirements, and expectations are not being met so that timely corrective
action can be taken to ensure the security and reliability of the voting
system. Jurisdictions without supporting measures for security and
reliability may lack sufficient insight into their system operations.

Overall, responses to our local jurisdiction survey show that large
jurisdictions were most likely to record voting system performance and
small jurisdictions were least likely. We estimated that 42 percent of
jurisdictions overall monitored the accuracy of voting equipment in the
2004 general election. Other measures recorded were spoiled ballots
(estimated at 50 percent of jurisdictions), undervotes (50 percent of
jurisdictions),^30 and overvotes (49 percent of jurisdictions). During our
visits to local jurisdictions, election officials in several jurisdictions
told us that measuring overvotes was not a relevant performance indicator
for jurisdictions using DREs because they do not permit overvoting, and
that undervotes were not a meaningful metric because most voters focused
on a limited range of issues or candidates and thus frequently chose not
to vote on all contests. Figure 8 shows the percentages of small, medium
and large jurisdictions that collected information on various measures of
accuracy.

^29 The 95 percent confidence interval for large jurisdictions is +/-8
percentage points, and for small jurisdictions it is +/-7 percentage
points.

^30 An estimated 25 percent of respondents selected "not applicable" to
the question on spoiled/ruined ballots in their survey response.

Figure 8: Estimated Percentages of Jurisdictions that Collected
Information on Voting Accuracy for the 2004 General Election, by
Jurisdiction Size

^aThe differences between small jurisdictions and both medium and large
jurisdictions are statistically significant. 

^bThe 95 percent confidence interval for small jurisdictions in these
categories is plus or minus 6 percentage points.

^cThe difference between small jurisdictions and large jurisdictions is
statistically significant. The 95 percent confidence interval for small
jurisdictions is plus or minus 6 percentage points.

We estimated that 15 percent of jurisdictions measured voting system
failure rates and 11 percent measured system downtime.^31 A higher
percentage of large and medium jurisdictions collected these performance
data than small jurisdictions. Collection of these data was also related
to the predominant voting method used by a jurisdiction, with
jurisdictions that predominantly used DREs more likely to collect system
data than those that used precinct count or central count optical scan
voting methods (an estimated 45 percent of jurisdictions versus 23 percent
or 10 percent, respectively). Figure 9 shows the percentages of small,
medium, and large jurisdictions that collected information on voting
equipment failures and downtime. Figure 10 shows the percentages by
predominant voting method of all jurisdictions that collected data on
equipment failures.

^31 An estimated 66 percent of respondents selected the response "not
applicable" for the survey questions on measurement of pieces of equipment
that failed and equipment downtime.

Figure 9: Estimated Percentages of Jurisdictions that Collected
Information on Voting Equipment Reliability for the 2004 General Election,
by Jurisdiction Size

^aThe differences between all categories of jurisdiction size are
statistically significant.

Figure 10: Estimated Percentages of Jurisdictions that Collected
Information on Voting Equipment Failures for the 2004 General Election, by
Predominant Voting Method

Note: The differences between DRE and both central count and precinct
count optical scan voting methods are statistically significant.

^aThe 95 percent confidence interval for DRE is plus or minus 13
percentage points.

^bThe 95 percent confidence interval for central count optical scan
percentages is plus 7 or minus 5 percentage points.

^cThe 95 percent confidence interval for precinct count optical scan
percentages is plus 8 or minus 7 percentage points.

Further, an estimated 55 percent of all jurisdictions kept a written
record of issues and problems that occurred on Election Day, which could
be a potential source of performance data. Large jurisdictions were more
likely to keep a written record of issues or problems that occurred on
Election Day. Specifically, we estimated that 79 percent of large
jurisdictions kept such records, compared with 59 percent of medium
jurisdictions and 52 percent of small jurisdictions.^32 The
responsibilities for monitoring or reporting voting system performance
most often rested with local jurisdictions. We estimated that 83 percent
of local jurisdictions had local officials responsible for performance
monitoring or reporting, while states or other organizations (such as
independent consultants or vendors) held such responsibilities in 11
percent and 13 percent of jurisdictions, respectively.

Addressing Voting System Challenges Requires the Combined Efforts of All Levels
of Government

The challenges in ensuring that voting systems perform securely and
reliably are not unlike those faced by any technology user--application of
well-defined standards for system capabilities and performance; effective
integration of the people, processes, and technology throughout the voting
system life cycle; rigorous and disciplined performance of security and
testing activities; objective measurement to determine whether the systems
are performing as intended; and analytical and economically justified
basis for making informed decisions about voting system investment
options. These challenges are complicated by other conditions common to
both the national elections community and other information technology
environments: the distribution of responsibilities among various
organizations, technology changes, funding opportunities and constraints,
changing requirements and standards, and public attention. Although
responsibility for voting system performance falls largely on local
governmental units, state and federal governments have important roles to
play as well. Therefore, all levels of government need to work together to
address these challenges, under the leadership of the EAC.

To assist the EAC in executing its leadership role, we previously made
recommendations to the commission aimed at better planning its ongoing and
future activities relative to, for example, system standards and
information sharing. While the EAC agreed with the recommendations, it
told us that its ability to effectively execute its role is resource
constrained.

^32The differences between large jurisdictions and both medium and small
jurisdictions are statistically significant.

Establishing and Applying Current and Comprehensive Standards

The extent to which states and local jurisdictions adopt and consistently
apply up-to-date voting systems standards directly affects the security
and reliability of voting systems during elections. For the 2006 general
election, a substantial proportion of states and jurisdictions had yet to
adopt the most current federal voting system standards or related
performance measures, meaning that the systems they employ may not perform
as securely and reliably as desired. Beyond this, decisions by states and
local jurisdictions to apply these latest standards for the 2008 election
present additional challenges such as (1) whether the systems can be
tested and certified in time for the election and (2) adopting standards
that are now undergoing revision rather than continued use of earlier
standards or later adoption of even newer standards.

EAC plays an important role in ensuring the timely testing and
certification of voting systems against the latest standards and in
informing state and local decisions on whether to adopt these standards
for the 2008 election. Accordingly, we have recommended that EAC define
tasks and time frames for achieving the full operational capability of the
national voting system certification program. These management elements
would need to take into account estimating testing capacity and expected
volume for the testing laboratory accreditation program, establishing
protocols and time frames for reviewing certification packages, and
setting norms for timely consideration and decision making regarding
system certifications. Sharing this information with state and local
election officials would help them to plan for system upgrades, testing,
and state certification to meet their upcoming election cycles.

States and local jurisdictions must also consider the timely adoption of
standards in light of the additional work that is currently under way and
planned to address known weaknesses in the national standards. For
example, in addition to establishing minimum functional and performance
requirements for voting systems, standards can also be used to govern
integration of election systems, such as the accuracy, reliability,
privacy, and security of components and interfaces. Accordingly, we have
recommended that the EAC collaborate with NIST and the Technical
Guidelines Development Committee to define the specific tasks, measurable
outcomes, milestones, and resource needs required to improve the voting
system standards. Identifying the incremental improvements to standards
for several future election cycles and coordinating these with states and
local jurisdictions would help those officials plan for these cycles and
prepare the public for expected changes in voting technologies, security
and reliability features, and compensating controls.

Ensuring that Necessary Security, Testing, and Operational Activities Are
Effectively Performed

Maximizing the performance of the voting systems that jurisdictions
currently have and those they plan to use in the next general election
means taking proactive steps between now and November 2008 to ensure that
these systems perform as intended. These steps include activities aimed at
securing, testing, and preparing these systems for operation. Although the
vast majority of jurisdictions performed security, testing, and
operational activities in one form or another for the 2004 general
election, the extent and nature of these activities varied among
jurisdictions and depended on the availability of resources (financial and
human capital) committed to them. The challenge facing all voting
jurisdictions will be to ensure that these activities are fully and
properly performed, particularly in light of the security and reliability
concerns that have been reported with electronic voting systems.

Security, testing, and operational activities are to a large degree
responsive to--and limited by--formal state and local directives. For
2004, election officials for some states identified various state and
local directives for managing the security and reliability of their voting
systems, including security plans, security testing, system acceptance
testing, and voting equipment auditing. When appropriately defined and
implemented, such directives can promote the effective execution of
security and testing practices across all phases of the election process.
As voting technologies and requirements evolve, states and local
jurisdictions face the challenge of adapting and implementing the
directives to meet the needs of their specific election environments.

Managing the People, Processes, and Technology as Components of the Overall
Process

As previously stated, jurisdictions need to manage the triad of people,
processes, and technology as interrelated and interdependent parts of the
total voting process. Given the amount of time that remains between now
and the November 2008 elections, jurisdictions' voting system performance
is more likely to be influenced by improvements in poll worker system
operation training, voter education about system use, and vote casting and
counting procedures than by changes to the physical systems. The challenge
for voting jurisdictions is thus to ensure that these people and process
issues are dealt with effectively.

In this regard, the election management decisions and practices of states
and local jurisdictions can benefit from the experiences and results of
those with comparable election environments. In 2004 and again in 2006,
EAC compiled such information into guidance documents for widespread use
by election officials. However, as the election environment and voting
systems continue to evolve, additional lessons and topics will undoubtedly
surface. Accordingly, we have recommended that the EAC establish a process
and schedule for periodically compiling and disseminating recommended
practices for security and reliability across the system life cycle and
that the practices be informed by information it collects on the problems
and vulnerabilities of these systems. Incorporating the feedback obtained
through actual voting system development, acquisition, preparation, and
operations into practical guidance will allow the election community to be
more robust and efficient.

Gathering and Using Reliable System Performance Measures and Data and Making
Informed Investment Decisions

Reliable measures and objective data are needed for jurisdictions to know
whether the technology they use is meeting the needs of the user
communities (both the voters and the officials who administer the
elections). While the vast majority of jurisdictions reported that they
were satisfied with the performance of their respective technologies in
the November 2004 elections, this satisfaction was based mostly on the
subjective impressions of election officials rather than on objective data
that measured voting system performance. Although these impressions should
not be discounted, informed decision making on voting system operations
and technology investment requires more objective data. The immediate
challenge for jurisdictions is to define measures and begin collecting
data so that they can definitely know how their systems are performing.

States and local jurisdictions can benefit from sharing performance data
on voting systems, including information on problems and vulnerabilities.
However, the diffused and decentralized nature of our election system
impedes timely and accurate collection and dissemination of this type of
information for particular voting system models. Accordingly, we have
recommended that the EAC develop a process and associated time frames for
sharing information on voting system problems and vulnerabilities across
the election community. The national voting system certification process
established in January 2007 provides a mechanism for election officials to
report problems and vulnerabilities with their systems to the EAC. Not yet
defined are the mechanisms to collect and disseminate information on
problems and vulnerabilities that are identified by voting system vendors
and independent groups outside of the national certification process.

In addition, foreseeable changes in technology require jurisdictions to
determine whether a particular technology will provide benefits over its
useful life that are commensurate with life-cycle costs (acquisition as
well as operation and maintenance) and to assess whether these collective
costs are affordable and sustainable. Thus, the long-term challenge for
jurisdictions is to view and treat voting systems as capital investments
and to manage them as such, including basing decisions on technology
investments on clearly defined requirements and reliable analyses of
quantitative and qualitative return on investment.

In closing, I would like to say again that electronic voting systems are
an undeniably critical link in the overall election chain. While this link
alone cannot make an election, it can break one. The problems that some
jurisdictions have experienced and the serious concerns that have surfaced
highlight the potential for continuing difficulties in upcoming national
elections if these challenges are not effectively addressed. The EAC plays
a vital role related to ensuring that election officials and voters are
educated and well informed about the proper implementation and use of
electronic voting systems and ensuring that jurisdictions take the
appropriate steps--related to people, process, and technology--that are
needed regarding security, testing, and operations. More strategically,
the EAC needs to move swiftly to strengthen the voting system standards
and the testing associated with enforcing them. However, the EAC alone
cannot ensure that electronic voting system challenges are effectively
addressed. State and local governments must also do their parts. Moreover,
critical to the commission's ability to do its part will be the adequacy
of resources at its disposal and the degree of cooperation it receives
from entities at all levels of government.

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

Contact and Acknowledgements

For further information, please contact Randolph C. Hite at (202) 512-3439
or by e-mail at [email protected] . Other key contributors to this
testimony were Neil Doherty, Nancy Glover, Paula Moore, and Kim Zelonis.

United States Government Accountability Office

GAO

Testimony

Before the Subcommittee on Financial Services and General Government,
Committee on Appropriations, House of Representatives

For Release on Delivery

Expected at 2:00 p.m. EST Wednesday, March 7, 2007

ELECTIONS

All Levels of Government Are Needed to Address Electronic Voting System
Challenges

Statement of Randolph C. Hite, Director

Information Technology Architecture and Systems

GAO-07-576T

Abbreviations

COTS commercial off-the-shelf
DRE direct recording electronic
EAC Election Assistance Commission
HAVA Help America Vote Act of 2002
NIST National Institute of Standards and Technology

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. However, 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.

www.gao.gov/cgi-bin/getrpt?GAO-07-576T.

To view the full product, including the scope
and methodology, click on the link above.

For more information, contact Randolph C. Hite at (202) 512-3439 or
[email protected].

Highlights of GAO-07-576T, a report to the Subcommittee on Financial
Services and General Government, Committee on Appropriations, House of
Representatives

March 7, 2007

ELECTIONS

All Levels of Government Are Needed to Address Electronic Voting System
Challenges

Since the 2000 national elections, concerns have been raised by various
groups regarding the election process, including voting technologies.
Beginning in 2001, GAO published a series of reports examining virtually
every aspect of the elections process. GAO's complement of reports were
used by Congress in framing the Help America Vote Act of 2002, which,
among other things, provided for replacement of older voting equipment
with more modern electronic voting systems and established the Election
Assistance Commission (EAC) to lead the nation's election reform efforts.
GAO's later reports have raised concerns about the security and
reliability of these electronic voting systems, examined the EAC's efforts
to address these concerns, and surveyed state and local officials about
practices used during the 2004 election, as well as plans for their
systems for the 2006 election.

Using its published work on electronic voting systems, GAO was asked to
testify on (1) the contextual role and characteristics of electronic
voting systems, (2) the range of security and reliability concerns that
have been reported about these systems, (3) the experiences and management
practices of states and local jurisdictions regarding these systems, and
(4) the longstanding and emerging challenges facing all levels of
government in using these systems.

Voting systems are one facet of a multifaceted, year-round elections
process that involves the interplay of people, processes, and technology,
and includes all levels of government. How well these systems play their
role in an election depends in large part on how well they are managed
throughout their life cycle, which begins with defining system standards;
includes system design, development, and testing; and concludes with
system operations. Important attributes of the systems' performance are
security, reliability, ease of use, and cost effectiveness.

A range of parties knowledgable about elections or voting systems have
expressed concerns about the security and reliability of electronic voting
systems; these concerns can be associated with stages in the system life
cycle. Examples of concerns include vague or incomplete voting system
standards, system design flaws, poorly developed security controls,
incorrect system configurations, inadequate testing, and poor overall
security management.

For the 2004 national elections, states' and local governments' responses
to our surveys showed that they did not always ensure that important life
cycle and security management practices were employed for their respective
electronic voting systems. In particular, responses indicated that the
most current standards were not always adopted and applied, security
management practices and controls were employed to varying degrees, and
certain types of system testing were not commonly performed. Moreover,
jurisdictions' responses showed that they did not consistently monitor the
performance of their systems.

In GAO's view, the challenges faced in acquiring and operating electronic
voting systems are not unlike those faced by any technology user--adoption
and application of well-defined system standards; effective integration of
the technology with the people who operate it and the processes that
govern this operation; rigorous and disciplined performance of system
security and testing activities; reliable measurement of system
performance; and the analytical basis for making informed, economically
justified decisions about voting system investment options. These
challenges are complicated by other conditions such as the distribution of
responsibilities among various organizations and funding opportunities and
constraints. Given the diffused and decentralized allocation of voting
system roles and responsibilities across all levels of government,
addressing these challenges will require the combined efforts of all
levels of government, under the leadership of the EAC. To assist the EAC
in executing its leadership role, GAO has previously made recommendations
to the commission aimed at better planning its ongoing and future
activities relative to, for example, system standards and information
sharing. While the EAC agreed with the recommendations, it stated that its
ability to effectively execute its role is resource constrained.

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. However, 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.

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 ( www.gao.gov ). 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
www.gao.gov and select "Subscribe to Updates."

Order by Mail or Phone

The first copy of each printed report is free. Additional copies are $2
each. A check or money order should be made out to the Superintendent of
Documents. GAO also accepts VISA and Mastercard. Orders for 100 or more
copies mailed to a single address are discounted 25 percent. Orders should
be sent to:

U.S. Government Accountability Office 441 G Street NW, Room LM Washington,
D.C. 20548

To order by Phone: Voice: (202) 512-6000 TDD: (202) 512-2537 Fax: (202)
512-6061

To Report Fraud, Waste, and Abuse in Federal Programs

Contact:

Web site: www.gao.gov/fraudnet/fraudnet.htm E-mail:
[email protected] Automated answering system: (800) 424-5454 or (202)
512-7470

Congressional Relations

Gloria Jarmon, Managing Director, [email protected] (202) 512-4400 U.S.
Government Accountability Office, 441 G Street NW, Room 7125 Washington,
D.C. 20548

Public Affairs

Paul Anderson, Managing Director, [email protected] (202) 512-4800
U.S. Government Accountability Office, 441 G Street NW, Room 7149
Washington, D.C. 20548

References

Visible links
  28. http://www.gao.gov/cgi-bin/getrpt?GAO-02-90
  29. http://www.gao.gov/cgi-bin/getrpt?GAO-02-3
*** End of document. ***