#8 Absentee Voting
A study of accessibility issues and potential
design solutions for designers, procurement
officials, and consumers.
W. Bradley Fain, Ph.D.
Carrie Bell
July 2012
1
This monograph documents the results of research conducted in the Accessibility Evaluation Facility
(AEF) of the Georgia Tech Research Institute. The AEF has performed a series of accessibility
evaluations for both industry and government customers. Although information about the outcome of
specific evaluations has been withheld, the data presented in this monograph is based upon general
outcomes and lessons learned stemming from research conducted at the Georgia Tech Research
Institute.
For additional information about this monograph, please contact:
Dr. Brad Fain
Georgia Tech Research Institute
ELSYS/HSID/HSEB Mail Code 0837
Atlanta, GA 30332-0837
Voice: (404) 407-7261
Fax: (404) 407-9261
brad.fain@gtri.gatech.edu
2
Table of Contents
Table of Contents ...................................................................................................................................... 3
List of Tables ............................................................................................................................................ 6
List of Figures ........................................................................................................................................... 7
Introduction............................................................................................................................................... 8
Assistive Technologies (AT) Commonly Used for Voting .................................................................. 8
Applicable Guidelines and Policies for the Voting Process................................................................ 17
Help America Vote Act of 2002 (HAVA) ...................................................................................... 17
Voting System Standards (VSS) and Voluntary Voting System Guidelines (VVSG) ................... 18
Americans with Disabilities Act Guidelines (ADA) ....................................................................... 22
Section 508 Guidelines.................................................................................................................... 23
Section 255 Guidelines.................................................................................................................... 25
Information and Communication Technology (ICT) Standards and Guidelines ............................ 26
ANSI/HFES 200- Human Factors Engineering of Software User Interfaces ................................. 32
Human Factors Design Standard ..................................................................................................... 33
Applicable Laws for Absentee Voting ................................................................................................ 33
Voting Accessibility for the Elderly and Handicapped Act of 1984 ............................................... 33
Uniformed and Overseas Citizens Absentee Voting Act of 1986 (UOCAVA) .............................. 33
Military and Overseas Voter Empowerment Act of 2009 (MOVE) ............................................... 34
Evaluation Methodologies ...................................................................................................................... 35
How do I define the goals of the evaluation? ...................................................................................... 36
How do I perform a task analysis? ...................................................................................................... 37
How do I develop a Method of Test? .................................................................................................. 38
Which evaluation techniques should I use? ........................................................................................ 40
How should I document the results of the evaluation? ....................................................................... 45
Accessibility Issues and Potential Solutions........................................................................................... 47
Context for Potential Solutions ........................................................................................................... 47
Issues with the Absentee Voting Process ............................................................................................ 50
Obtaining the Ballot ........................................................................................................................ 50
Voters cannot sign the application form. ..................................................................................... 50
Reading the Ballot ........................................................................................................................... 51
Text-only paper ballots cannot be read by voters with visual impairments. ............................... 51
Marking the Ballot .......................................................................................................................... 52
Ballots are not provided in a format that can be marked electronically. ..................................... 52
Voters who need assistance marking the ballot cannot vote privately. ....................................... 53
Error feedback is not provided. ................................................................................................... 54
Paper-based vote verification must be visually inspected. .......................................................... 55
Returning the Ballot ........................................................................................................................ 56
Physical submission of the ballot is difficult for some users. ..................................................... 56
Correct postage is not indicated. .................................................................................................. 57
Issues with Voting Technology........................................................................................................... 58
Physical Ability Factors .................................................................................................................. 58
The voting device does not provide or support necessary accessibility options. ........................ 58
The voting device is not portable. ............................................................................................... 59
3
Voters may not be able to physically hold a portable voting device. .......................................... 60
Some components of voting devices may be difficult for users with mobility impairments to
reach. ............................................................................................................................................ 61
Some components of devices may be difficult for users with mobility impairments to see. ...... 63
Glare makes it difficult for some users to see the display. .......................................................... 64
System response time to user input is slow. ................................................................................ 66
Audio and Speech Output ............................................................................................................... 67
The voting device does not provide a headphone jack. ............................................................... 67
The placement of the headphone jack allows the headphone cord to interfere with use of the
device. .......................................................................................................................................... 68
The voting device does not support t-coil coupling..................................................................... 70
The volume level is insufficient. ................................................................................................. 71
The rate at which speech output is provided is not adjustable. ................................................... 72
Speech output via speakers may be inappropriate due to privacy concerns. ............................... 73
Speech output is difficult to understand due to poor sound quality or interference. ................... 75
Speech output is not repeated. ..................................................................................................... 76
Speech output requires users to listen to large blocks of text without the ability to pause, start
over, or skip ahead. ...................................................................................................................... 77
Speech output is inaccurate. ........................................................................................................ 78
Non-verbal audio output is not meaningful. ................................................................................ 79
Issues with Ballot Design and Presentation ........................................................................................ 80
Instructions and Help ...................................................................................................................... 80
The system does not provide sufficient instructions. .................................................................. 80
Timing of instructions is inappropriate........................................................................................ 81
Instructions may be difficult for users with cognitive impairments to understand. .................... 82
User Interface Design ...................................................................................................................... 83
The system does not provide clear and consistent page titles. .................................................... 83
Display elements do not receive focus in the correct reading order. ........................................... 84
The system does not provide clear and consistent navigation controls. ...................................... 85
The mapping of hardware controls to user interface functions is inconsistent............................ 86
When scrolling is required (and it should be used sparingly and consistently), the user is not
notified that scrolling is needed. .................................................................................................. 87
Onscreen controls that are small and close together may be difficult to activate without
activating adjacent controls. ........................................................................................................ 88
Complex or inconsistent user interface screens may be difficult for users to understand. .......... 89
System time-outs may cause problems for some users. .............................................................. 90
The voting device changes the ballot automatically based on assumptions about user intent. ... 91
The voting device does not provide sufficient feedback to the user............................................ 92
The system does not provide clear and consistent status information. ........................................ 93
Information Presentation ................................................................................................................. 94
Icons used in place of text labels may be ambiguous. ................................................................. 94
Small text and icons are difficult for users with low vision to perceive...................................... 95
Non-text elements are not accessible to some users with visual impairments. ........................... 96
Control names are not descriptive. .............................................................................................. 97
Candidates are listed on multiple columns or pages. ................................................................... 98
Multiple races are displayed on one page of an electronic ballot. ............................................... 99
4
Write-In Voting ............................................................................................................................. 100
Use of the write-in option is confusing...................................................................................... 100
Typing the name of a write-in candidate is difficult.................................................................. 101
Ballot Review ................................................................................................................................ 102
Users are not given the opportunity to review their options before submitting their vote. ....... 102
Users do not notice when the review screen differs from their intended selections.................. 103
The system does not provide adequate notification of or protection against overvoting and
undervoting. ............................................................................................................................... 104
Revising a ballot is excessively complex or difficult. ............................................................... 105
Policy-Related Issues ........................................................................................................................ 106
Obtaining accurate information on voter registration is difficult as service members transition
out of active duty status. ............................................................................................................ 106
Registering to vote is different from state to state. .................................................................... 107
References ......................................................................................................................................... 108
5
List of Tables
Table 1: Requirements for voting accessibility mandated by the Help America Vote Act of 2002, as
amended by H.R. 2239. .......................................................................................................................... 17
Table 2: Voting System Standards of 2002 and Voluntary Voting System Guidelines of 2005 and 2007.
................................................................................................................................................................ 18
Table 3: ADA-ABA Accessibility Guideline Sections Relevant to Voting Device Accessibility. ........ 23
Table 4: Potentially Applicable Section 508 Standards. ........................................................................ 23
Table 5: Potentially Applicable Section 255 Standards. ........................................................................ 25
Table 6: Information and Communication Technology (ICT) Standards and Guidelines. .................... 27
Table 7: Applicable ANSI/HFES 200 Guidelines. ................................................................................. 33
6
List of Figures
Figure 1: Magnifying glass. ...................................................................................................................... 9
Figure 2: Neckloop. ................................................................................................................................ 10
Figure 3: Earphones connected to electronic device. ............................................................................. 10
Figure 4: Screen reader. .......................................................................................................................... 11
Figure 5: Braille display. ........................................................................................................................ 11
Figure 6: Examples of alternative input devices (track ball, left, and foot-operated mouse, right). ...... 12
Figure 7: Mouth stick.............................................................................................................................. 14
Figure 8: Reaching aid. ........................................................................................................................... 15
Figure 9: Examples of input switches. .................................................................................................... 16
Figure 10: Sip and puff device................................................................................................................ 17
Figure 11: Recommended process for absentee voting. ......................................................................... 48
Figure 12: iPad absentee voting concept. ............................................................................................... 48
Figure 13: Return envelopes with and without additional postage amount displayed. .......................... 57
Figure 14: Portable voting device. .......................................................................................................... 59
Figure 15: The touchscreen of a voting device is too high, causing reach problems for a seated user. . 61
Figure 16: The touchscreen of a voting device is low enough for a seated user to reach it. .................. 62
Figure 17: The angle of the display results in glare for a seated user. ................................................... 64
Figure 18: An adjustable display allows a seated user to adjust the display angle to reduce glare. ....... 65
Figure 19: Headphone jack location allows the cord to interfere with use of the machine. ................... 68
Figure 20: Headphone jack location prevents the cord from interfering with use of the machine. ........ 69
Figure 21: A volume knob is provided to allow the user to adjust the output volume. .......................... 71
Figure 22: Potentially sensitive audio is output through speakers. ........................................................ 73
Figure 23: A headphone jack is provided for private listening. ............................................................. 74
7
Introduction
This document contains a set of issues related to the process of absentee voting, primarily
focusing on issues associated with paper ballots, ballot design, and the presentation of information on a
small electronic device used for marking ballots. Additionally, this monograph focuses on issues
associated with the physical device and associated hardware, the physical environment in which the
device may be used, and the steps in the process before and after the actual marking of the ballot
(obtaining a ballot, printing and signing it, submitting it). The issues herein are divided into three
general categories.

Issues with the Absentee Voting Process

Issues with Voting Technology

Issues with Ballot Design and Presentation
Research conducted in relation to the absentee voting monograph was largely collected from
interviews and surveys conducted with service members who had recently become injured or disabled
as a result of military combat. According to the U.S. Department of Defense (2012), as of July 10,
2012, 49,008 U.S. troops have been wounded during the current conflicts in Iraq and Afghanistan,
many returning home with a range of disabilities including loss of limbs, impaired vision, and
traumatic brain injury (TBI). Thus, several aspects of the voting process present accessibility
challenges for wounded service members, including the accessibility of the physical space in which
voting occurs, the design of the ballot (legibility, consistency, and the organization and presentation of
information), and the technologies by which ballots are displayed to and marked by voters.
Assistive Technologies (AT) Commonly Used for Voting
Assistive technologies (AT) are pieces of equipment or software that are used to increase the
functional capabilities of people with disabilities. The following is a list of AT that people might use
to aid in interacting with casting an absentee ballot:
Magnifying glass - A magnifying glass or magnifier is a handheld lens that is used to increase
the size of an image. Typical magnification powers range from 2x to 10x. Some magnifying glasses
have an integrated light, which may increase visibility for some users. Some magnifying glasses are in
the form of a rectangular strip that can be overlaid on a line of text.
o
o
o
Used by: Magnifiers are used by people with low vision, who may require larger text and
images to be able to read and understand content.
Used for: On a ballot marking device, magnifiers are used to read button labels and ballot
information.
AT impact on voting: Magnifiers allow people with low vision to view information
presented in a size that is otherwise too small for them to read. The use of a magnifier
requires the user to get very close (within arm’s reach) to the object of interest. This may
require the user to assume an uncomfortable position (e.g., bending forward at the waist,
rising from a reclined position, if in a hospital bed). The design and placement of the
voting technology to be viewed may prevent a user from getting close enough with a
magnifier to read the information of interest.
8
Figure 1: Magnifying glass.
Screen magnifiers – A screen magnifier is a software application that presents enlarged screen
output on a computer’s display. Most often, the software magnifies only a portion of the screen that
the user selects. Most screen magnifiers are capable of 16 times magnification, and some may be
capable of 32 times magnification or more. However, increasing the screen magnification also reduces
the amount of information that can be displayed, so users typically use the lowest magnification setting
that is still usable for them. Other features of a screen magnifier may include color inversion or other
high contrast options as well as cursor customization to increase the visibility of the mouse cursor.
o Used by: Screen magnifiers are typically used by individuals with low vision.
o Used for: Screen magnifiers are used to enlarge portions of a visual display.
o AT impact on voting: Screen magnifiers allow people with low vision to view information
presented in a size that is otherwise too small for them to read. A device equipped with a
screen magnifying application must include some means of selecting portions of the screen
for magnification (e.g., a mouse, keypad, or touchscreen).
Neckloop – Neckloops are assistive listening devices that transmit magnetic signals to t-coils
inside hearing aids. The t-coils transform magnetic signals into sound. Most hearing aids allow users
to switch between microphone mode and t-coil mode. When in t-coil mode, the hearing aid transmits
sound only from the device connected to the neckloop (although it may also transmit noise from
interfering magnetic fields). Neckloops typically have 3.5 mm plugs that are compatible with common
audio output jacks. Silhouettes and ear links provide the same functionality as neckloops.
o
o
o
Used by: Neckloops are often used by individuals with hearing impairments –especially
those who also have vision impairments.
Used for: Neckloops can transmit audio from a device to a user.
AT impact on voting: Devices are often equipped with forms of audio output that make
the device accessible to users with visual impairments. However, this audio output is
inaccessible to users who also have hearing impairments. Neckloops enable users with both
vision and hearing impairments to receive information from the device, provided that the
device is equipped with a 3.5mm audio output jack, because the neckloops interface with
the user’s hearing aid. The user can adjust the neckloop to amplify only the sound coming
from the device interface, so that ambient background noise is filtered. Any type of verbal
information, alerts, or feedback can be provided through a neckloop.
9
Figure 2: Neckloop.
Earphones – Earphones are common listening devices that can be plugged into electronic
devices, typically via a 3.5mm plug. Earphones can be used in the same manner as neckloops, but they
do not require hearing aids.
o
o
o
Used by: Earphones are often used by individuals with visual impairments who require an
audio interface.
Used for: Earphones transmit verbal information and other audio information from a
machine to a user.
AT impact on voting: Earphones enable users with low or no vision to receive verbal
information and other auditory information, such as feedback from the machine. Earphones
alone might not enable independent voting for some users unless the machine also supports
speech recognition.
Figure 3: Earphones connected to electronic device.
Screen readers – A screen reader is a software application that interprets output typically
displayed on a computer screen into a format that is usable for an individual user. The output may be
presented in the form of text-to-speech, auditory icons, or a Braille display. Some of the more
10
commonly used screen readers are JAWS (Job Access With Speech) by Freedom Scientific and
Window-Eyes by GW Micro. However, due cost of these types of screen readers, individuals may
instead choose to use an online screen reader (such as NonVisual Desktop Access) or a screen reader
that is included in a computer’s operating system (such as Microsoft Narrator or Macintosh
VoiceOver). Users control the screen reader using keyboard commands or, in the case of VoiceOver, a
combination of both keyboard commands and trackpad gestures.
A Braille display may be used in combination with or instead of a text-to-speech screen reader. Some
users with visual impairments prefer to use only a Braille display due to privacy concerns. However, a
Braille display is not necessarily a suitable option because only 10% of individuals who are blind know
how to read Braille. Also, learning Braille as an adult is similar to learning a new language; it can take
several months to over a year to learn.
o Used by: Screen readers are typically used by individuals with visual impairments such as
blindness and low vision and by individuals with cognitive impairments.
o Used for: Screen readers are used to convert visual text into a format that is interpretable
by users who cannot see the text – most often into an audio format.
o AT impact on voting: Screen readers can make electronic devices accessible to the
visually impaired. Given privacy concerns, auditory outputs produced by screen readers
should be accessible via a 3.5 mm headphone jack rather than speakers.
Figure 4: Screen reader.
Figure 5: Braille display.
11
Alternative input devices – A variety of input devices are available for individuals unable to
use a standard keyboard and/or mouse. These alternative input devices may be used by individuals
with upper mobility impairments, visual impairments, or cognitive impairments.
Alternative mice inputs include a joysticks, trackballs, vertical mice, roller mice, or foot operated mice.
A joystick and vertical mouse allow users to adopt a more comfortable hand position that does not
require rotation of the wrist. Alternative devices for keyboards include on screen keyboards,
ergonomic keyboards, expanded keyboards, miniature keyboards, and one-handed keyboards. A
pointing device, such as a mouse, is used to select characters on an on screen keyboard. Ergonomic,
expanded, miniature, and one-handed keyboards are designed for individuals with limited dexterity and
fine motor control. There are a wide variety of ergonomic keyboards, each designed with a particular
limitation in mind.
o Used by: Alternative input devices are typically used by individuals with limitations in
dexterity, fine motor control, and range of motion. Users who cannot grip a traditional
mouse may be able to use a trackball or roller mouse more effectively. A foot operated
mouse provides point-and-click functionality for users with severe upper mobility
impairments. It is also appropriate for users who can use a keyboard but cannot easily
switch from the keyboard to the mouse.
o Used for: Alternative mice and keyboards are typically used to interact with personal
computers. However, other machines (e.g., kiosks) can be equipped to enable input from
these devices. Many of these devices connect via a USB cable.
o AT impact on voting: Alternative input devices can enable users to interact with a machine
with greater comfort and fewer errors. Also, individuals with limited mobility may be
unable to reach traditional controls, and may therefore require alternative input devices.
Figure 6: Examples of alternative input devices (track ball, left, and foot-operated mouse, right).
Speech recognition software – A computer equipped with a microphone and speech
recognition software can receive voice commands and dictation from the user. Some speech
recognition software allows the user to operate all aspects of a computer using only verbal commands.
The speech recognition software may be speaker-dependent, meaning the individual user has to “train”
the software to recognize and understand the user’s voice and commands. Alternatively, the software
may be speaker independent. The software can be designed to recognize a limited number of
commands that are appropriate for a given context (e.g., user says “next” to view the next page), which
can improve the software’s recognition performance.
12
o
o
o
Used by: A fully functional speech interface does not require the use of upper limbs. When
coupled with an auditory display, a speech interface does not require vision. Therefore,
speech recognition software is used by individuals with upper mobility impairments,
dexterity impairments, and/or visual impairments.
Used for: Speech recognition software interprets verbal input from the user. It can be used
to move cursors, type words, and execute commands.
AT impact on voting: Well-developed speech recognition software could support all forms
of user input to a device that would otherwise be performed manually. The software can
enable users to move a cursor, navigate among pages, select options from a multiple choice
list, etc. Note that the user’s interactions should remain private, which could be
problematic when they are voting for write-in candidates (for a solution, see Dawkins &
Gilbert, 2010). Although some speech recognition technology can achieve 99% accuracy,
such as Dragon NaturallySpeaking, this high level of accuracy can only be achieved under
optimal conditions – in a quiet room where the speaker has good diction and speaks at a
consistent rate.
Manipulation stick – A manipulation stick is a rod, typically a wooden dowel, used as an aid
to increase a user’s reach or strength. Common end attachments are rubber tips and hooks. Rubber
tips are used to press against an object (for example, to close a cupboard door or activate a switch).
Hooks are used to pick things up or to pull objects (for example, to pull a lever). It is not uncommon
for a manipulation stick to have a rubber tip on one end and a hook on the other. Many people who
use manipulation sticks have both a short stick and a long stick.
o
o
o
Used by: Manipulation sticks are typically used by individuals who have limited strength in
their hands or arms, or who have limited reach capability (i.e., people with upper mobility
impairments).
Used for: A manipulation stick can facilitate tasks associated with use of a voting device.
Common tasks include activating controls and inserting smart cards into direct-recording
electronic (DRE) voting machines.
AT impact on voting: Manipulation sticks increase a user’s reach. The manipulation stick
can also increase a user’s leverage. Manipulation sticks are sometimes used in combination
with an electric wheelchair. The user positions the stick, and then uses the force generated
by the wheelchair’s motor to apply the necessary pressure to activate a control or open a
drawer. Manipulation sticks are only useful when they are available to the user. Users who
have both a short stick and a long stick may not carry both, and may find that they need the
one that they don’t have. Someone with a long manipulation stick may find that the stick is
too long to use easily. For example, to use the longer stick, the user must often be at a
further distance from the object being manipulated, which means that a lot of extra
maneuvering may be required.
Some voting booths will not provide enough room for a user to obtain the necessary
distance, which may mean the user will be unable to use the stick. A greater degree of fine
motor control may also be required to use a longer stick. Additionally, the ends of the stick
may not be suitable for use with some devices. For example, the rubber end of the
manipulation stick may be too large to activate some controls, or the gap on the hook may
be insufficient to grab onto some objects with adequate leverage. Also, touchscreens that
13
respond to skin moisture or that discharge electrical current are not compatible with
manipulation sticks.
Mouth stick – A mouth stick is similar to a manipulation stick, but it is held in the mouth. The
mouth stick is held by the teeth and lips, and is controlled by neck and lip movement. Many mouth
sticks are made of aluminum, and can be equipped with end attachments such as clips or rubber points.
o
o
o
Used by: Mouth sticks are most commonly used by quadriplegics who have little or no
ability to move their limbs.
Used for: Mouth sticks are used to press buttons, manipulate lightweight objects, or hold
items such as pens.
AT impact on voting: Mouth sticks can be used to interact with a voting device if the
machine is at a proper height for a seated user. Voters might need assistance getting
appropriately positioned at the booth. He or she might have limited reach with the mouth
stick, depending on the extent of upper torso mobility, so upper or distant parts of the
display might be inaccessible. Touchscreens that respond to skin moisture are not
compatible with manipulation sticks. Also, touchscreens that discharge electrical current in
response to touch are incompatible with some types of manipulation sticks (i.e., those that
do not conduct electricity).
Figure 7: Mouth stick.
Reaching aid – A reaching aid is a 1 to 2 foot long device with a trigger handle used to open
and close the end for the purpose of grasping objects that are difficult to reach. Error! Reference
source not found. shows a typical reaching aid.
o
o
o
Used by: Reaching aids are often used by people who use a wheelchair or who have an
upper mobility impairment that limits their reach distance.
Used for: Reaching aids are used to insert and retrieve smart cards into DRE voting
machine, to manipulate paper ballots, and to retrieve printouts from the voting machine
(i.e., a voter-verifiable paper trail).
AT impact on voting: Reaching aids enhance the ability of people with limited reach to
perform voting tasks that require the user to reach or grasp items. Such tasks may include
inserting and retrieving cards and ballots and retrieving outputs.
14
Figure 8: Reaching aid.
Touchscreens – There are several types of touchscreen technology in use including resistive,
capacitive, infrared, and surface acoustic wave (SAW) touchscreens. A resistive touchscreen panel
consists of multiple layers, two of which are electrically conductive layers separated by a small gap.
When pressure is applied to the outer surface, the two layers make contact causing a change in the
electrical current. Because this type of touchscreen relies on pressure to detect touch, the material with
which the pressure is applied is irrelevant. However, resistive touchscreens typically offer only 75%
clarity and may not be appropriate for users with low vision. A capacitive touchscreen detects touch
based on the electrical properties of the object that is touching the screen. Certain types of objects
(e.g., the tip of a pencil, gloved fingers, or a prosthetic arm) do not produce detectable touches because
they are not grounded. Infrared touchscreens respond to both human touch and other objects such as
styluses. However, these types of touchscreens can be problematic for users that hover their finger
over a control before selecting it. SAW touchscreens also respond to both human touch and other
objects such as styluses, but the technology may be too expensive to be a feasible solution.
o Used by: Touchscreens may be easier to use for individuals that experience difficulty using
both a keyboard and a mouse. They may also be more user friendly for individuals with
cognitive disabilities because of the one-to-one mapping between the control and the
display. Since the controls and display are the same and can change from one screen to the
next, touchscreen technology is also useful when space is limited.
o Used for: All interactions with a machine that might otherwise be supported by a keyboard,
keypad, or buttons.
o AT impact on voting: Touchscreens may be easier to use for many users, including those
without disabilities. However, certain disabilities make touchscreens inaccessible,
including visual impairments, upper limb tremors, and dexterity impairments. Individuals
with these impairments will need another type of assistive technology to supplement or
replace the touchscreen.
Input switches – A variety of input switches accommodate various disabilities. Switches can
be activated by a hand, foot, head, or other body part. A sip-and-puff switch is controlled by the user’s
breath. Many switches have two input channels, allowing two actions (e.g., cursor advance and
select). Switches must interface directly with the hardware to be activated, typically by wire, although
wireless switches exist.
o
o
Used by: Input switches are used by individuals with dexterity impairments and upper limb
mobility impairments.
Used for: Input switches act as accessible buttons when connected to another device whose
controls are inaccessible. The functionality of the switch depends on the software of the
15
o
other device. The device must be programmed to accept switch input, and might respond
input by advancing a cursor among display items or selecting a display item. For example,
a sip-and-puff interface can advance the cursor each time the user sips air through a tube,
and select an item when the user puffs through the tube.
AT impact on voting: Impaired coordination of upper limbs can cause users to make
frequent mistakes as they interact with controls – particularly highly sensitive touchscreens
– as they often over- or under-shoot their intended targets. The use of an input switch can
greatly reduce the number of errors that are committed. Also, input switches enable users
with limited reach or other upper limb impairments to interact with electronic devices.
Figure 9: Examples of input switches.
Sip and puff –A sip and puff (SNP) controller is a system used to control a device by turning
air pressure into electrical signals. This technology allows users to control devices such as computer
mice or motorized wheelchairs by inhaling/sipping and exhaling/puffing on a “wand” or tube. Devices
that use SNP technology require specific amounts of air pressure to be "sipped" or "puffed" by the
user. A calibration process, with the aid of an assistant or technician, is usually required before the
user begins using the SNP device.
o Used by: Sip and puff technologies are used by individuals with upper mobility
impairments and other dexterity limitations.
o Used for: Sips and puffs can be used to effectively navigate a computer screen and
accomplish mouse button clicks. SNP technology allows users with very limited control of
their bodies to be able to interact with electronics.
o AT impact on voting: For people with motor impairments or dexterity limitations,
operating a traditional computer mouse or switch can be very difficult. SNP technology
enables users to interact with electronic devices. A sip-and-puff input device combined
with scanning software allows users the capability to utilize many keyboard-accessible
programs as well.
16
Figure 10: Sip and puff device.
Applicable Guidelines and Policies for the Voting Process
Some of the laws surrounding various disability requirements have created a need for more
closely examining voting accessibility for voters who with physical and cognitive limitations. The
following guidelines and regulatory policies have been applied to the voting process and are useful in
ascertaining voters’ rights within the US as well as abroad, where US citizens and service members
reside.
Help America Vote Act of 2002 (HAVA)
HAVA was passed to modernize the administration of federal elections. HAVA provides
federal funding to for programs such as provisional voting, voting information, statewide voter
registration lists and identification requirements for first-time registrants, administrative complaint
procedures, and updated/upgraded voting equipment. HAVA was critical in the development of voting
system guidelines for people with disabilities. This act led to required improvements to the quality,
reliability, accuracy, accessibility, affordability, and security of voting equipment, election systems,
and technology.
Table 1: Requirements for voting accessibility mandated by the Help America Vote Act of 2002, as amended by H.R.
2239.
Paragraph
Text of Requirement
Title III(a)(3)(A)
The voting system shall be accessible for individuals with disabilities, including
nonvisual accessibility for the blind and visually impaired, in a manner that
provides the same opportunity for access and participation (including privacy and
independence) as for other voters
Title III(a)(3)(B)
The voting system shall satisfy the requirement of subparagraph (A) through the
use of at least one direct recording electronic voting system or other voting system
equipped for individuals with disabilities at each polling place, and such voting
system shall provide a mechanism for voter-verification of results which separates
the function of vote generation from the function of vote casting in a manner
analogous to that described in section 4 with respect to the separation of paper
ballot generation and paper ballot verification and preservation, but does not
require the use of paper.
17
Paragraph
Text of Requirement
Title III(a)(3)(C)
The equipment deployed in accordance with subparagraph (B) shall meet the
voting system standards for disability access and voter-verification of results as
outlined in this paragraph in accordance with the deadline set forth in section 5(a),
provided that if it does not and an interim paper system is deployed in accordance
with section 5(b), disabled voters shall have the option of using the interim paper
system with the assistance of an aide of the voter's personal selection or using the
voting system otherwise put in place for use by disabled voters at the time in
question in accordance with the Help America Vote Act of 2002, as in effect prior
to the enactment of this Act, except that the deadline set forth in section
301(a)(3)(C) of such Act (42 U.S.C. 15481(a)(3)(C)) is moved forward from
January 1, 2007, to January 1, 2006
Title III(a)(3)(D)
Election officials shall be instructed in the rights of the disabled to vote with the
assistance of an aide of their selection under the Voting Rights Act of 1965
Voting System Standards (VSS) and Voluntary Voting System Guidelines
(VVSG)
The Voting System Standards (Federal Election Commission, 2002) contains a list of
accessibility standards for voting machines and ballots that were later updated and supplemented with
the Voluntary Voting System Guidelines (Election Assistance Commission, 2007). These updated
guidelines reflect changes in technology and election practices. The VSS and VVSG are necessary for
testing the voting systems to determine if their ability to provide the basic, albeit necessary
functionality, accessibility and security capabilities that are required. The voting guidelines
significantly increase security requirements for voting systems that allow more people with disabilities
to vote privately and independently. Both sets of guidelines are voluntary, and are therefore listed
separately from Section 508 standards, which are mandatory. The VVSG was adopted in 2005 to
update the Voting Systems Standards but has not been officially adopted. Thus, both sets of guidelines
are used to address sound design principles in this monograph.
Table 2: Voting System Standards of 2002 and Voluntary Voting System Guidelines of 2005 and 2007.
Source & Section
Text of Standard / Guideline
General
VSS 3.2.2.2
There is no restriction on space allowed for the installation of voting systems,
except that the arrangement of these systems shall not impede performance of
their duties by polling place officials, the orderly flow of voters through the polling
place, or the ability for the voter to vote in private;
VSS 3.2.4.1
All systems shall provide voting booths or enclosures for poll site use. Such
booths or enclosures may be integral to the voting system or supplies as
components of the voting system, and shall: a. Be integral to, or makes provision
for, the installation of, the voting device; b. Ensure by its structure stability against
movement or overturning during entry, occupancy, and exit by the voter; c.
Provide privacy for the voter, and be designed in such a way as to prevent
observation of the ballot by any person other than the voter; and d. Be capable of
meeting the accessibility requirements of Section 2.2.7.1;
VSS 2.2.7.2(i)
Provide a secondary means of voter identification or authentication when the
primary means of doing so uses biometric measures that require a voter to
possess particular biological characteristics;
18
Source & Section
Text of Standard / Guideline
VVSG 3.1.1(1)
All eligible voters are to have access to the voting process without discrimination.
The voting process must be accessible to individuals with disabilities. The voting
process includes access to the polling place, instructions on how to vote, initiating
the voting session, selecting among contest choices, review of the ballot, final
submission of the ballot, and getting help when needed;
VVSG 3.2.4-C
Instructional material for the voter SHALL conform to norms and best practices for
plain language;
VVSG 3.2.4-C.3
The system SHOULD use familiar, common words and avoid technical or
specialized words that voters are not likely to understand;
VVSG 3.2.4-C.4
The system SHOULD issue instructions on the correct way to perform actions,
rather than telling voters what not to do.
Machine Specifications
VSS 2.2.7.2(a)
Not require the voter to bring their own assistive technology to a polling place;
VSS 2.2.7.2(g)
For a system that requires a response by a voter in a specific period of time, alert
the voter before this time period has expired and allow the voter additional time to
indicate that more time is needed;
VSS 2.2.7.2(h)
For a system that provides sound cues as a method to alert the voter about a
certain condition, such as the occurrence of an error, or a confirmation, the tone
shall be accompanied by a visual cue for users who cannot hear the audio
prompt;
VSS 2.4.3.3(l)
Provide sufficient computational performance to provide responses back to each
voter entry in no more than three seconds;
VVSG 3.2.3.1-A.3
The voting system SHALL issue all warnings in a way that preserves the privacy
of the voter and the confidentiality of the ballot;
VVSG 3.2.3.2-B
No information SHALL be kept within an electronic CVR that identifies any
accessibility feature(s) used by a voter;
VVSG 3.2.4-A
The voting station SHALL provide instructions for all its valid operations;
VVSG 3.2.4-B
The voting system SHALL provide a means for the voter to get help directly from
the system at any time during the voting session;
VVSG 3.2.6-C.1
On touch screens, the sensitive touch areas SHALL have a minimum height of 0.5
inches and minimum width of 0.7 inches. The vertical distance between the
centers of adjacent areas SHALL be at least 0.6 inches, and the horizontal
distance at least 0.8 inches;
VVSG 3.2.6.1-D
If the system has not completed its visual response within one second, it SHALL
present to the voter, within 0.5 seconds of the voter's action, some indication that
it is preparing its response;
VVSG 3.3.1-A
The Acc-VS [Accessible Voting Station]SHALL be integrated into the
manufacturer’s complete voting system so as to support accessibility for disabled
voters throughout the voting session.
General Ballot Design
VSS 2.4.3.3(h)
Allow the voter, before the ballot is cast, to review his or her choices and, if the
voter desires, to delete or change his or her choices before the ballot is cast;
VSS 2.4.3.3(i)
For electronic image displays, prompt the voter to confirm the voter's choices
before casting his or her ballot, signifying to the voter that casting the ballot is
19
Source & Section
Text of Standard / Guideline
irrevocable and directing the voter to confirm the voter's intention to cast the
ballot;
VVSG 3.1.1(2)
Each cast ballot must accurately capture the selections made by the voter. The
ballot must be presented to the voter in a manner that is clear and usable. Voters
should encounter no difficulty or confusion regarding the process for recording
their votes;
VVSG
3.2(a)(1)(A)(i)
Except as provided in subparagraph (B), the voting system (including any lever
voting system, optical scanning voting system, or direct recording electronic
system) shall permit the voter to verify (in a private and independent manner) the
votes selected by the voter on the ballot before the ballot is cast and counted;
VVSG
3.2(a)(1)(A)(ii)
Except as provided in subparagraph (B), the voting system (including any lever
voting system, optical scanning voting system, or direct recording electronic
system) shall provide the voter with the opportunity (in a private and independent
manner) to change the ballot or correct any error before the ballot is cast and
counted (including the opportunity to correct the error through the issuance of a
replacement ballot if the voter was otherwise unable to change the ballot or
correct any error);
VVSG 3.2.2-A
If the voter selects more than the allowable number of choices within a contest,
the voting system SHALL notify the voter of the effect of this action before the
ballot is cast and counted;
VVSG 3.2.2.1-E
The VEBD (voter-editable ballot device) SHALL provide navigation controls that
allow the voter to advance to the next contest or go back to the previous contest
before completing a vote on the contest(s) currently being presented (whether
visually or aurally);
VVSG 3.2.6-A
Voting systems SHALL NOT require page scrolling by the voter.
Controls
VSS 2.2.7.2(f)(1)
For a device with touchscreen or contact-sensitive controls, provide an input
method using mechanically operated controls or keys that shall be tactilely
discernible without activating the controls or keys;
VSS 2.2.7.2(f)(2)
For a device with touchscreen or contact-sensitive controls, provide an input
method using mechanically operated controls or keys that shall be operable with
one hand and not require tight grasping, pinching, or twisting of the wrist;
VSS 2.2.7.2(f)(3)
For a device with touchscreen or contact-sensitive controls, provide an input
method using mechanically operated controls or keys that shall require a force
less than 5 lbs (22.2 N) to operate;
VSS 2.2.7.2(f)(4)
For a device with touchscreen or contact-sensitive controls, provide an input
method using mechanically operated controls or keys that shall provide no key
repeat function;
VSS 2.4.3.3(b)
Enable the voter to easily identify the selection button or switch, or the active area
of the ballot display that is associated with each candidate or ballot measure
response.
Visual Display
VSS 2.2.5.2.2(b)
All error messages requiring intervention by an operator or precinct official shall
be displayed or printed unambiguously in easily understood language text, or by
means of other suitable visual indicators;
20
Source & Section
Text of Standard / Guideline
VSS 2.2.7.2(e)(1)
For electronic image displays, permit the voter to adjust the contrast setting;
VSS 2.2.7.2(e)(2)
For electronic image displays, permit the voter to adjust color settings, when color
is used;
VSS 2.2.7.2(e)(3)
For electronic image displays, permit the voter to adjust the size of the text so that
the height of capital letters varies over a range of 3 to 6.3 millimeters;
VSS 2.3.1.1(f)
Ensure that vote response fields, selection buttons, or switches properly align with
the specific candidate names and/or issues printed on the ballot display, ballot
card or sheet, or separate ballot pages;
VSS 2.4.3.1(a)
To facilitate casting a ballot, all systems shall provide text that is at least 3
millimeters high and provide the capability to adjust or magnify the text to an
apparent size of 6.3 millimeters;
VSS 3.4.9(b)
Information or data displays shall be large enough to be readable by voters and
operators with no disabilities and by voters with disabilities consistent with the
requirements defined in Section 2.2.7 of the Standards;
VSS 3.4.9(d)
Color coding shall be selected so as to assure correct perception by voters and
operators with color blindness; and shall not be used as the only means of
conveying information, indicating an action, prompting a response, or
distinguishing a visual element;
VSS 3.4.9(e)
The system's display shall not use flashing or blinking text objects, or other
elements having a flash or blink frequency, greater than 2 Hz and lower than 55
Hz;
VVSG 3.2.4-F
The use of color by the voting system SHOULD agree with common conventions:
(a) green, blue or white is used for general information or as a normal status
indicator; (b) amber or yellow is used to indicate warnings or a marginal status; (c)
red is used to indicate error conditions or a problem requiring immediate attention;
VVSG 3.2.4-G
When an icon is used to convey information, indicate an action, or prompt a
response, it SHALL be accompanied by a corresponding linguistic label;
VVSG 3.2.5-A
No voting system display screen SHALL flicker with a frequency between 2 Hz
and 55 Hz;
VVSG 3.2.5-D
Voting systems SHALL provide a minimum font size of 3.0mm (measured as the
height of a capital letter) for all text intended for voters or poll workers;
VVSG 3.2.5-E
A voting station that uses an electronic image display SHALL be capable of
showing all information in at least two font sizes, (a) 3.0-4.0 mm and (b) 6.3-9.0
mm, under control of the voter. The system SHALL allow the voter to adjust font
size throughout the voting session while preserving the current votes;
VVSG 3.2.5-G.2
The system MAY achieve legibility of paper records by supporting magnification of
those records. This magnification MAY be done by optical or electronic devices.
The manufacturer MAY either: 1) provide the magnifier itself as part of the system,
or 2) provide the make and model number of readily available magnifiers that are
compatible with the system;
VVSG 3.2.5-I
The voting station SHALL be capable of showing all information in high contrast
either by default or under the control of the voter. The system SHALL allow the
voter to adjust contrast throughout the voting session while preserving the current
votes. High contrast is a figure-to-ground ambient contrast ratio for text and
informational graphics of at least 6:1;
21
Source & Section
Text of Standard / Guideline
VVSG 3.2.5-J
The default color coding SHALL support correct perception by voters with color
blindness;
VVSG 3.2.5-K
Color coding SHALL NOT be used as the sole means of conveying information,
indicating an action, prompting a response, or distinguishing a visual element.
Auditory Display
VSS 2.2.7.2(b)(1)
Provide audio information and stimulus that communicates to the voter the
complete content of the ballot;
VSS 2.2.7.2(b)(2)
Provide audio information and stimulus that provides instruction to the voter in
operation of the voting device;
VSS 2.2.7.2(b)(3)
Provide audio information and stimulus that provides instruction so that the voter
has the same vote capabilities and options as those provided by the system to
individuals who are not using audio technology;
VSS 2.2.7.2(b)(4)
Provide audio information and stimulus that for a system that supports write-in
voting, enables the voter to review the voter's write-in input, edit that input, and
confirm that the edits meet the voter's intent;
VSS 2.2.7.2(b)(5)
Provide audio information and stimulus that enables the voter to request repetition
of any information provided by the system;
VSS 2.2.7.2(b)(6)
Provide audio information and stimulus that supports the use of headphones
provided by the system that may be discarded after each use;
VSS 2.2.7.2(b)(7)
Provide audio information and stimulus that provides the audio signal through an
industry standard connector for private listening using a 1/8 inch stereo
headphone jack to allow individual voters to supply personal headsets;
VSS 2.2.7.2(b)(8)
Provide audio information and stimulus that provides a volume control with an
adjustable amplification up to a maximum of 105 dB that automatically resets to
the default for each voter;
VSS 2.2.7.2(c)
Provide, in conformance with FCC Part 68, a wireless coupling for assistive
devices used by people who are hard of hearing when a system utilizes a
telephone style handset to provide audio information;
VSS 2.2.7.2(d)
Meet the requirements of ANSI C63.19-2001 Category 4 to avoid electromagnetic
interference with assistive hearing devices;
VVSG 3.2.3.1-A.2
During the voting session, the audio interface of the voting system SHALL be
audible only to the voter.
Americans with Disabilities Act Guidelines (ADA)
The ADA is a civil rights law that prohibits discrimination based on disability. The mission of
the ADA is to make it possible for everyone with a disability to live a life of freedom and equality,
which includes the right to vote both privately and independently. It is similar to the American with
Disabilities Act in that both laws make it illegal to discriminate against people with disabilities. This
law opened up opportunities to people with disabilities by implementing changes to public policy,
healthcare law and policy changes, and public law changes for those with physical, mental and
cognitive disabilities.
ADA guidelines for building accessibility were codified in 28 CFR Part 36, Appendix A. The code
provides more detailed guidelines than those provided by the ADA-ABA guidelines for some aspects
22
of building accessibility. Several sections listed may not pertain explicitly to absentee voting, but parts
of those sections may be relevant.
Table 3: ADA-ABA Accessibility Guideline Sections Relevant to Voting Device Accessibility.
Section
306
308
309
703.3
707
904.3.3
Title
Knee and Toe Clearance
Reach Ranges
Operable Parts
Braille
Automatic Teller Machines and Fare Machines
Check Writing Surfaces
Section 508 Guidelines
Originally added as an amendment to the Rehabilitation Act of 1986, Section 508 requires that
all Federal information that is accessible electronically must be accessible for those with disabilities.
Information must be accessible in a variety of ways, which are specific to each disability. Section 508
applies specifically to federal departments and agencies, although various state and local governments
have also adopted legislation based on Section 508. Section 508 does not apply to procurement in the
private sector. However, Section 508 does apply to private organizations, including religious
organizations that provide services for local, state, or federal governments. Furthermore, the Section
508 technical standards represent best practices for accessibility, and are used in this document as
recommendations for designing accessible voting devices. Note that many of these guidelines were the
listsVSS
Section
guidelines
potentially
applicable
to voting
basis for4 the
(see 508
Table
2). Tablethat
4: are
Potentially
Applicable
Section
508 devices.
Standards.
Table 4: Potentially Applicable Section 508 Standards.
508 Standard that may apply
Situations in which it could apply
1194.25(a) Self-contained products shall be usable by people
with disabilities without requiring an end-user to attach assistive
technology to the product. Personal headsets for private
listening are not assistive technology.
This standard applies if the device requires
users to attach items of assistive technology
(excepting audio headsets) to the device in
order to access it.
1194.25(b) When a timed response is required, the user shall
be alerted and given sufficient time to indicate more time is
required.
This standard applies if there are any
instances where the device limits the time
allowed for users to perform an action or
provide a response.
1194.25(c) Where a product utilizes touchscreens or contactsensitive controls, an input method shall be provided that
complies with 1194.23(k)(1) through (4).
This standard applies if the device has a
touchscreen or other non-mechanical,
contact sensitive controls.
1194.23(k)(1) Controls and keys shall be tactilely discernible
without activating the controls or keys.
Standard 1194.23(k)(3) only applies if key
repeat is supported (e.g., keys repeat entry
when pressed for a specified period of time).
1194.23(k)(2) Controls and keys shall be operable with one
hand and shall not require tight grasping, pinching, or twisting of
the wrist. The force required to activate controls and keys shall
be 5 lbs. (22.2 N) maximum.
1194.23(k)(3) If key repeat is supported, the delay before repeat
23
Note: A best practice would be to apply
1194.23(k) to all mechanically operated
controls such as buttons, knobs, and
switches, whether or not 1194.25(c) applies.
Some believe this approach reflects the true
508 Standard that may apply
Situations in which it could apply
shall be adjustable to at least 2 seconds. Key repeat rate shall
be adjustable to 2 seconds per character.
intent of the guidelines.
1194.23(k)(4) The status of all locking or toggle controls or keys
shall be visually discernible, and discernible either through touch
or sound.
1194.25(d) When biometric forms of user identification or control
are used, an alternative form of identification or activation, which
does not require the user to possess particular biological
characteristics, shall also be provided.
This standard applies if the device uses
biometric forms of user identification, such
as a fingerprint scanner or a retinal scanner.
1194.25(e) When products provide auditory output, the audio
signal shall be provided at a standard signal level through an
industry standard connector that will allow for private listening.
The product must provide the ability to interrupt, pause, and
restart the audio at any time.
This standard applies if the device provides
voice output.
1194.25(f) When products deliver voice output in a public area,
incremental volume control shall be provided with output
amplification up to a level of at least 65 dB. Where the ambient
noise level of the environment is above 45 dB, a volume gain of
at least 20 dB above the ambient level shall be user selectable.
A function shall be provided to automatically reset the volume to
the default level after every use.
This standard applies if the device provides
voice output.
1194.25(g) Color coding shall not be used as the only means of
conveying information, indicating an action, prompting a
response, or distinguishing a visual element.
This standard applies if color coding is used
to convey information on the device.
1194.25(h) When a product permits a user to adjust color and
contrast settings, a range of color selections capable of
producing a variety of contrast levels shall be provided.
This standard applies if the device’s display
uses color and the device permits the user to
adjust color and contrast settings.
1194.25(i) Products shall be designed to avoid causing the
screen to flicker with a frequency greater than 2 Hz and lower
than 55 Hz.
This standard applies if the device’s display
or elements on the display flash with a
frequency between 2 and 55 Hz.
1194.31(a) At least one mode of operation and information
retrieval that does not require user vision shall be provided, or
support for assistive technology used by people who are blind or
visually impaired shall be provided.
This standard always applies.
1194.31(b) At least one mode of operation and information
retrieval that does not require visual acuity greater than 20/70
shall be provided in audio and enlarged print output working
together or independently, or support for assistive technology
used by people who are visually impaired shall be provided.
This standard always applies.
1194.31(c) At least one mode of operation and information
retrieval that does not require user hearing shall be provided, or
support for assistive technology used by people who are deaf or
hard of hearing shall be provided.
This standard always applies.
1194.31(d) Where audio information is important for the use of a
product, at least one mode of operation and information retrieval
shall be provided in an enhanced auditory fashion, or support for
assistive hearing devices shall be provided.
This standard applies if audio information is
important for use of the device. For
example, audio information (speech output,
etc.) for which there is not a redundant visual
24
508 Standard that may apply
Situations in which it could apply
alternative would be covered by this
standard.
1194.31(e) At least one mode of operation and information
retrieval that does not require user speech shall be provided, or
support for assistive technology used by people with disabilities
shall be provided.
This standard always applies.
1194.31(f) At least one mode of operation and information
retrieval that does not require fine motor control or simultaneous
actions and that is operable with limited reach and strength shall
be provided.
This standard always applies.
Section 255 Guidelines
Section 255 of the Telecommunications Act of 1996 requires that telecommunication products
be accessible to people with disabilities. Telecommunications products must be accessible to people
with disabilities with little difficulty or expense. If this is not possible, manufacturers must create their
products to be compatible with adaptive equipment used by people with disabilities.
Telecommunications products covered include telecommunication devices, such as telephones and fax
machines, and products that have a telecommunication service capability, such as computers with
modems. Section 255 does not apply to products made before the law took effect on February 8, 1996.
Table 5 lists Section 255 guidelines that are potentially applicable to voting devices.
Table 5: Potentially Applicable Section 255 Standards.
1193.41 Input, control, and mechanical functions. Input, control, and mechanical functions shall be
locatable, identifiable, and operable in accordance with each of the following, assessed independently:
(a) Operable without vision. Provide at least one mode that does not require user vision.
(b) Operable with low vision and limited or no hearing. Provide at least one mode that permits operation
by users with visual acuity between 20/70 and 20/200, without relying on audio output.
(c) Operable with little or no color perception. Provide at least one mode that does not require user
color perception.
(d) Operable without hearing. Provide at least one mode that does not require user auditory perception.
(e) Operable with limited manual dexterity. Provide at least one mode that does not require user fine
motor control or simultaneous actions.
(f) Operable with limited reach and strength. Provide at least one mode that is operable with user
limited reach and strength.
(g) Operable without time-dependent controls. Provide at least one mode that does not require a
response time. Alternatively, a response time may be required if it can be by-passed or adjusted by the
user over a wide range.
(h) Operable without speech. Provide at least one mode that does not require user speech.
(i) Operable with limited cognitive skills. Provide at least one mode that minimizes the cognitive,
memory, language, and learning skills required of the user.
1193.43 Output, display, and control functions. All information necessary to operate and use the
product, including but not limited to, text, static or dynamic images, icons, labels, sounds, or incidental
operating cues, shall comply with each of the following, assessed independently:
25
(a) Availability of visual information. Provide visual information through at least one mode in auditory
form.
(b) Availability of visual information for low vision users. Provide visual information through at least one
mode to users with visual acuity between 20/70 and 20/200 without relying on audio.
(c) Access to moving text. Provide moving text in at least one static presentation mode at the option of
the user.
(d) Availability of auditory information. Provide auditory information through at least one mode in visual
form and, where appropriate, in tactile form.
(e) Availability of auditory information for people who are hard of hearing. Provide audio or acoustic
information, including any auditory feedback tones that are important for the use of the product, through
at least one mode in enhanced auditory fashion (i.e., increased amplification, increased signal-to-noise
ratio, or combination). For transmitted voice signals, provide a gain adjustable up to a minimum of 20
dB. For incremental volume control, provide at least one intermediate step of 12 dB of gain.
(f) Prevention of visually-induced seizures. Visual displays and indicators shall minimize visual flicker
that might induce seizures in people with photosensitive epilepsy.
(g) Availability of audio cutoff. Where a product delivers audio output through an external speaker,
provide an industry standard connector for headphones or personal listening devices (e.g., phone-like
handset or earcup) which cuts off the speaker(s) when used.
(h) Non-interference with hearing technologies. Reduce interference to hearing technologies (including
hearing aids, cochlear implants, and assistive listening devices) to the lowest possible level that allows a
user to utilize the product.
(i) Hearing aid coupling. Where a product delivers output by an audio transducer which is normally held
up to the ear, provide a means for effective wireless coupling to hearing aids.
Information and Communication Technology (ICT) Standards and Guidelines
Proposed updates to Section 508 and Section 255 (Disabled Persons' Telecommunications
Access) were published in 2010 under the title “Information and Communication Technology
Standards and Guidelines” (see http://www.access-board.gov/sec508/refresh/draftrule.htm#admin508). The guidelines address accessibility issues that have emerged from recent
technological developments. They include guidance for the design of electronic documents which may
be relevant to ballot design.
The ICT standards and guidelines provide guidance for systems with closed and open functionality.
Systems with closed functionality are “locked down,” preventing users from altering settings (e.g., font
size); they also do not provide peripheral hardware connections. Voting devices should have open
functionality, to the extent that users should be able to modify settings. However, an optimal voting
machine should not require voters to attach assistive technology (see VSS 2.2.7.2(a), which states,
“[The voting machine shall] not require the voter to bring their own assistive technology to a polling
place”). Therefore, a voting device should be open with respect to display and control settings, but
closed with respect to peripheral hardware connections. The latter recommendation should not be
applied to sub-optimal voting devices whose internal functionality does not support all varieties of
disabilities.
26
Table 6: Information and Communication Technology (ICT) Standards and Guidelines.
Functional Performance Criteria
302 Functional Performance Criteria
302.1 Without Vision. Where a visual mode of operation is provided, ICT shall provide at least one
mode of operation that does not require user vision.
302.2 With Limited Vision. Where a visual mode of operation is provided, ICT shall provide at least
one mode of operation that magnifies, one mode that reduces the field of vision required, and one
mode that allows user control of contrast.
302.3 Without Perception of Color. Where a visual mode of operation is provided, ICT shall provide
at least one mode of operation that does not require user perception of color.
302.4 Without Hearing. Where an auditory mode of operation is provided, ICT shall provide at least
one mode of operation that does not require user hearing.
302.5 With Limited Hearing. Where an auditory mode of operation is provided, ICT shall provide at
least one mode of operation which improves clarity, one mode that reduces background noise, and one
mode that allows user control of volume.
302.6 Without Speech. Where a spoken mode of operation is provided, ICT shall provide at least one
mode of operation that does not require user speech.
302.7 With Limited Manipulation. Where a manual mode of operation is provided, ICT shall provide
at least one mode of operation that does not require fine motor control or operation of more than one
control at the same time.
302.8 With Limited Reach and Strength. Where a manual mode of operation is provided, ICT shall
provide at least one mode of operation that is operable with limited reach and limited strength.
302.9 Minimize Photosensitive Seizure Triggers. ICT shall provide at least one mode of operation
that minimizes the potential for triggering photosensitive seizures.
Hardware
402 Closed Functionality
402.1 General. Except for personal headsets and audio loops, ICT with closed functionality shall be
operable without requiring the user to attach assistive technology and shall conform to 402
Advisory 402.1 General. Self-service machines, information kiosks, set-top boxes, and devices like
printers, copiers, fax machines, and calculators have closed functionality because their functionality is
self-contained. Their design precludes the user from adding peripherals or software to them.
ICT has closed functionality for many reasons. These reasons include design or policy. ICT may have
closed functionality in practice even though the manufacturer did not design or develop ICT to be
closed. Computers which are “locked down” to the extent that end users cannot adjust settings are
functionally closed.
Exceptions: 1. Audible tones shall be permitted instead of speech for visible output that is not
displayed for security purposes, including, but not limited to, asterisks representing personal
identification numbers.
402.2 Speech Enabled. ICT shall be speech enabled. Operating instructions and orientation, visible
transaction prompts, user input verification, error messages, and all displayed information for full use
shall be accessible to and independently usable by individuals with vision impairments. Speech shall
be delivered through a mechanism that is readily available to all users, including, but not limited to, an
industry standard connector or a telephone handset. Speech shall be recorded or digitized human, or
synthesized. Speech shall be coordinated with information displayed on the screen.
27
402.2.1 User Control. Speech for any single function shall be automatically interrupted when a
transaction is selected. Speech shall be capable of being repeated and paused.
402.3 Volume Control. ICT shall provide volume control complying with 402.3.
402.3.1 Private Listening. Where speech required by 402.2 is delivered through a mechanism for
private listening, ICT shall provide a mode of operation for controlling the volume.
402.3.2 Speaker Volume. Where sound is delivered through speakers on ICT, incremental volume
control shall be provided with output amplification up to a level of at least 65 dB. Where the ambient
noise level of the environment is above 45 dB, a volume gain of at least 20 dB above the ambient level
shall be user selectable. A function shall be provided to automatically reset the volume to the default
level after every use.
402.4 Characters. Characters displayed on the screen shall be in a sans serif font. Characters shall
be 3/16 inch (4.8 mm) high minimum based on the uppercase letter “I”. Characters shall contrast with
their background with either light characters on a dark background or dark characters on a light
background.
403 Biometrics
403.1 General. Where provided, biometrics shall not be the only means for user identification or
control.
Exception: Where at least two biometric options that use different biological characteristics are
provided, ICT shall be permitted to use biometrics as the only means for user identification or control.
404 Preservation of Information Provided for Accessibility
404.1 General. ICT that transmits or converts information or communication, shall not remove nonproprietary information provided for accessibility or shall restore it upon delivery.
Advisory 404.1 General. This provision applies to conversion techniques, such as encoding, signal
compression, and format transformation. Examples of ICT that encode, compress, or transform
include firewalls, routers, and gateways.
One example of ICT preserving information provided for accessibility is a media player that displays
embedded captions from a captioned video and does not strip out the captioning.
Another example of ICT preserving information provided for accessibility is converting a document into
a new format while retaining information about the identity, operation, and state of the interface
elements.
This provision does not require the addition or translation of information. For example, this is not a
requirement to change voice mail to text or to vocalize captions.
406 Standard Connections
406.1 General. Where connection points are provided, at least one of each type of connection shall
conform to industry standard non-proprietary formats.
Advisory 406.1 General. The intent of this provision is to ensure compatibility with assistive
technologies by requiring the use of standard connections on ICT.
Examples of connection points include expansion slots, ports, and connectors for cables.
Industry standard non-proprietary formats include wireless connections to ICT, such as infrared (IR)
and Bluetooth.
407 Operable Parts
407.1 General. Where provided, operable parts of ICT shall conform to 407.
28
407.2 Contrast. Where provided, keys and controls shall contrast visually from background surfaces.
Characters and symbols shall contrast visually from background surfaces. Visual contrast shall be
either light-on-dark or dark-on-light.
407.3 Tactilely Discernible. At least one tactilely discernible input control shall be provided for each
function. Where provided, key surfaces not on active areas of display screens shall be raised above
surrounding surfaces. Where touch or membrane keys are the only method of input, each key shall be
tactilely discernible from surrounding surfaces and adjacent keys.
407.3.1 Identification. Operable parts shall be tactilely discernible without activation.
407.4 Key Repeat. Where an alphabetic keyboard with key repeat is provided, the delay before the
key repeat shall be adjustable to at least 2 seconds and the key repeat rate shall be adjustable to 2
seconds per character.
407.5 Numeric Keys. Where provided, numeric keys shall be arranged in a 12-key ascending or
descending telephone keypad layout. The number five key shall be tactilely distinct from the other
keys.
Advisory 407.5 Numeric Keys. A telephone keypad and a keypad on a computer keyboard differ in
one significant feature, ascending versus descending numerical order of the layout. Both types of
keypad layout conform to this provision.
407.9 Operation. Operable parts shall be operable with one hand and shall not require tight grasping,
pinching, or twisting of the wrist. The force required to activate operable parts shall be 5 pounds (22.2
N) maximum.
407.10 Privacy. ICT shall provide the opportunity for the same degree of privacy of input and output to
all individuals.
407.11 Receipts, Tickets, and Transactional Outputs. Where receipts, tickets, or other outputs are
provided as a result of a transaction, speech output shall include all information necessary to complete
or verify the transaction.
Exception: Operable parts of ICT that is portable shall not be required to conform to 407.13 through
407.16
407.13 Clear Floor Space. A clear floor or ground space conforming to 36 CFR Part 1191 Appendix
D, Section 305 shall be provided.
407.14 Height. Operable parts of ICT shall be placed within one or more of the reach ranges
conforming to 36 CFR Part 1191 Appendix D, Section 308.
407.15 Visibility. Where a display screen is provided, it shall be visible from a point located 40 inches
(1015 mm) above the center of the clear floor space in front of the ICT.
407.16 Braille Instructions. Where speech is required by 402.2, braille instructions for initiating the
speech mode of operation shall be provided. Braille shall conform to 36 CFR Part 1191 Appendix D,
Section 703.3.
408 ICT with Two Way Voice Communication
408.1 General. ICT that provides two way voice communication shall conform to 408.
Exception: Where ICT is a closed system, conformance to standards other than ITU-T
Recommendation G.722 shall be permitted where equivalent or better acoustic performance is
provided and where conversion to ITU-T Recommendation G.722 at the borders of the closed system
is supported.
408.2 Volume Gain. For transmitted voice signals, ICT shall provide a gain adjustable to a minimum
of 18 dB. For incremental volume control, ICT shall provide at least one intermediate step of 12 dB
gain.
29
408.3 Magnetic Coupling. Where ICT delivers output by an audio transducer which is typically held
up to the ear, ICT shall provide a means for effective magnetic wireless coupling to hearing
technologies.
408.4 Minimize Interference. Interference with hearing technologies (including hearing aids, cochlear
implants, and assistive listening devices) shall be reduced to the lowest possible level and shall
conform to 408.4.
408.4.2 Digital Wireline. ICT in the form of digital wireline devices shall conform to TIA 1083,
Telephone Terminal Equipment Handset Magnetic Measurement Procedures and Performance
Requirements (incorporated by reference in Chapter 1).
408.5 ITU-T Recommendation G.722. ICT shall transmit and receive speech that is digitally encoded
in the manner specified by ITU-T Recommendation G.722 for encoding and storing audio information
(incorporated by reference in Chapter 1).
408.6 Real Time Text Functionality. Where ICT provides real time voice communication, ICT shall
provide real time text functionality and shall conform to 408.6.
408.6.1 Display of Real Time Text. Where provided, multi-line displays shall be compatible with real
time text systems used on the network.
408.6.2 Text Generation. Where provided, features capable of text generation shall be compatible
with real time text systems used on the network.
408.6.3 Interoperability. Where ICT interoperates outside of its closed system, or where ICT
connects to other systems, ICT shall conform to 408.6.3.1 or 408.6.3.2.
Platforms and Applications
501 General
501.1 Scope. The provisions of Chapter 5 shall apply where required by Chapter 1 or where
referenced by a requirement in this document.
Advisory 501.1 Scope. Examples of platforms are desktop, embedded operating systems (including
mobile), web browsers, plug-ins to web browsers which render a particular media or format, and sets of
components which allow other applications to execute.
Applications may be web-based or client-side software. Examples of applications are email clients,
word processors, help desk systems, content management systems, e-learning courseware, and
terminal emulation.
Exception: Platforms and applications that have closed functionality and that conform to 402 shall not
be required to conform to 502.
502 Interoperability with Assistive Technology
502.1 General. Platforms, platform software toolkits, and applications shall conform to 502.
502.2 Accessibility Services. Platforms and their software toolkits shall provide a documented set of
accessibility services that support a mode of operation for applications running on the platform to
interoperate with assistive technology and shall conform to 502.2. Applications that are also platforms
shall expose the underlying platform accessibility services or implement other documented accessibility
services.
502.2.1 Object Information. The object role, state(s), boundary, name, and description shall be
programmatically determinable.
502.2.2 Row, Column, and Headers. The row and column an object is in, and the headers for the row
and column for that component, if it is in a data table that has row or column headers, shall be
programmatically determinable.
30
502.2.3 Values. The current value and any minimum or maximum values, if the component represents
one of a range of values, shall be programmatically determinable.
502.2.4 Label Relationships. The relationship that a component has as a label for another
component, or of being labeled by another component, shall be programmatically determinable.
502.2.6 Text. The text contents, text attributes, and the boundary of text rendered to the screen shall
be programmatically determinable.
502.2.7 Actions. A list of actions that can be executed on an object shall be programmatically
determinable. Applications shall allow assistive technology to programmatically execute available
actions on objects.
502.2.8 Focus Cursor. Applications shall expose information and mechanisms necessary to track and
modify focus, text insertion point, and selection attributes of user interface components.
502.2.9 Event Notification. Notification of events relevant to user interactions, including but not
limited to changes in the component’s state(s), value, name, description, or boundary, shall be
available to assistive technologies.
502.3 Documented Accessibility Usage. Where platform documentation is available to application
developers, platforms and applications shall conform to 502.3.
502.3.1 User Control of Accessibility Features. Platforms shall provide a mode of operation for user
control over platform features that are defined in the platform documentation as accessibility features.
502.3.2 No Disruption of Accessibility Features. Applications shall not disrupt platform features that
are defined in the platform documentation as accessibility features.
503 Applications
503.1 General. Applications shall conform to 503.
503.2 User Preferences. Applications shall provide a mode of operation that allows user preferences
for platform settings for color, contrast, font type, font size, and focus cursor.
Exception: Applications that are designed to be isolated from their underlying platforms, including web
applications, shall not be required to conform to 503.2.
503.3 Alternative User Interfaces. Where an application provides an alternative user interface that
functions as assistive technology, the application shall use platform and other industry standard
accessibility services to provide the alternate user interface.
503.4 User Controls for Captions and Video Description. Where ICT displays video with
synchronized audio content, ICT shall provide user controls for closed captions and video description
conforming to 503.4.
503.4.1 Caption Controls. Where user controls are provided for the selection of volume, ICT shall
provide user controls for the selection of captions in at least one location that is comparable in
prominence to the location of the user controls for volume.
503.4.2 Video Description Controls. Where user controls are provided for the selection of channels,
ICT shall provide user controls for the selection of video description in at least one location that is
comparable in prominence to the location of the user controls for channels.
503.4.3 On-screen Menus. Where an on-screen menu is provided for the selection of volume or
channels, ICT shall provide for the selection of captions and video description at the same menu level
as that of volume and channel selection.
Documentation and Support Services
602 Documentation
602.1 General. Documentation that supports the use of ICT shall conform to 602.
31
602.2 Accessibility and Compatibility Features. Documentation shall list and explain how to use the
accessibility and compatibility features of the ICT that support the technical requirements of this
document. Documentation shall include accessibility features that are built-in and accessibility features
that provide compatibility with assistive technology.
Advisory 602.2 Accessibility and Compatibility Features. One example of an accessibility feature
is the ability to access commands and navigate using the keyboard. Voice recognition, screen readers,
and alternative keyboards rely upon keyboard control of features for accessible and efficient operation.
Keyboard navigation includes support for the following: cursor keys (up, down, left and right arrows),
tab and shift-tab (to cycle through fields), enter or spacebar (to select or activate), hot keys, macros,
and other keyboard acceleration mechanisms.
Where ICT components are designed to be part of an integrated system, this provision requires that
the documentation explains how to configure the system to support accessibility.
602.3 Materials. When ICT support services provide documentation, documentation materials shall
conform to 602.3.
602.3.1 WCAG Conformant. Documentation in electronic format shall conform to all Level A and
Level AA Success Criteria and all Conformance Requirements in WCAG 2.0 (incorporated by
reference in Chapter 1).
602.3.2 Alternate Formats. Alternate formats shall be provided upon request.
603 Support Services
603.1 General. ICT support services including, but not limited to, help desks, call centers, technical
support, and training services shall conform to 603.
603.2 Information on Accessibility and Compatibility Features. ICT support services and training
shall include information on the accessibility and compatibility features required by 602.2 to be listed
and explained in ICT documentation.
Advisory 603.2 Information on Accessibility and Compatibility Features. A best practice is for ICT
support services to provide training programs about the following topics: accessibility requirements for
individuals with disabilities; methods of communication used by individuals with disabilities; assistive
technology commonly used with ICT products; designing for accessibility; solutions for accessibility and
compatibility of ICT with assistive technology; the use of people-first language; and sensitivity training
concerning disability issues.
603.3 Effective Communication. ICT support services shall accommodate the communication needs
of individuals with disabilities.
Advisory 603.3 Effective Communication. To be effective, communication with individuals with
disabilities should include alternate methods of communication for both in-person and remote
communication. Examples of alternative methods are sign language interpreters, assistive listening
systems, TTYs, real time captioning, and telecommunications relay services. Examples of
telecommunication relay services are TTY speech-to-speech and video relay.
603.3.1 Materials Provided. When support services provide documentation, the documentation
materials shall conform to 602.
ANSI/HFES 200- Human Factors Engineering of Software User Interfaces
The HFES 200 guidelines, in accordance with American National Standards Institute (ANSI)
procedures, focus on user-system interfaces and interfaces between operating systems, middleware,
application software, and other software layers. The objective of the HFES 200 guidelines is to
provide requirements and recommendations that will increase the accessibility and ease of use of
32
software by people with a wide range of capabilities. Table 7 lists HFES 200 guidelines that are
applicable to voting devices. (These guidelines are also incorporated in the exhaustive VSS
document.)
Table 7: Applicable ANSI/HFES 200 Guidelines.
Applicable Guideline
Notes
8.4 General Control and Operation Guidelines
8.4.12 Facilitate navigation to the location of errors
When software detects that users have entered
invalid data, it should notify them in a way that
allows the users to identify and navigate easily to
the location of the error.
8.5 Compatibility with assistive technology
8.5.10 Enable appropriate presentation of tables
When displaying tabular data or data in columns,
the application provides assistive technology with
information about the data, including any row or
column names.
Human Factors Design Standard
Based on the 1996 Human Factors Design Guide, the Human Factors Design Standard (HFDS)
is an exhaustive list of best practices and principles involved in the development, design, selection and
analysis of Federal Aviation Administration systems and equipment. Although the HFDS was written
to guide the design of FAA systems, these human factors design principles provide strong guidance for
most systems with a human in the loop, including voting systems. The applicable HFDS guidelines are
not listed in this document as there are many guidelines that carry multiple exceptions. Although an
exhaustive list of guidelines are available, only those that are applicable have been cited herein.
Applicable Laws for Absentee Voting
This monograph presents guidelines and standards that are applicable to the administration of
federal elections. These are regarded as a list of best practices that should be followed in non-federal
elections as well. Each set of guidelines/standards can be categorized as compulsory or voluntary.
Voting Accessibility for the Elderly and Handicapped Act of 1984
Under this act, polling places across the US are required to be physically accessible to people
with disabilities for federal elections. States are also required to make registration and voting aids
available for people with disabilities as well as elderly voters. Information regarding elections is
required to provide information by telecommunications devices for the deaf (TDDs or TTYs).
Uniformed and Overseas Citizens Absentee Voting Act of 1986 (UOCAVA)
In 1986, Congress passed the Uniformed and Overseas Citizens Absentee Voting Act
(UOCAVA), which provides the legal basis for absentee voting by U.S. citizens who are active
members of a uniformed service as well as their spouses and dependents. Active duty service
members may request an absentee ballot by completing the Federal Post Card Application (FPCA), or
by other means allowed by the states. VAOs are available to assist personnel in requesting an absentee
ballot. The absentee ballot is then delivered through the mail, in person, or electronically. Although
procedures vary from state to state, ballots received electronically typically must be printed out and
33
completed by hand. Completed ballots may be returned by mail, fax, e-mail, or (in rare instances)
electronic submission. This process can be somewhat complicated. For example, in order to return a
ballot by mail, the voter may have to place the completed ballot in a privacy envelope, complete a
voter affidavit, and place all materials into a return envelope that must be signed and sealed. Once
service members have transitioned to civilian status, UOCAVA no longer applies. Veterans must
follow the same absentee voting procedures that are expected of the general civilian population.
Military and Overseas Voter Empowerment Act of 2009 (MOVE)
This Act amends the Uniformed and Overseas Citizens Absentee Voting Act (UOCAVA). It
requires states to electronically deliver voter registration materials and blank ballots to uniformed and
overseas voters. It also eliminates some state laws that caused disenfranchisement (e.g., requirements
for official notarization, which cannot be obtained abroad).
34
Evaluation Methodologies
There are two basic methods of acquiring information about human performance: direct
measurement and derived assessments. If your goal is to evaluate the ability of a specific individual to
perform in a technological system, then you place that individual in the system, give him or her a
representative series of tasks to complete, and observe his or her performance. This direct
measurement approach allows you to collect very good data about the ability of a single individual to
interact with a given device, although the findings may not generalize to a larger population. More
robust direct measurement studies adopt effective sampling strategies. By testing a diverse and
representative group of users, the evaluator can generalize findings to a larger population segment with
much better accuracy. Effective sampling for accessibility evaluations can be very challenging. Not
only are there a variety of disability types to consider, but a variety of functional abilities within each
disability type to consider as well.
There are eight major categories of functional limitations that can be addressed in an
accessibility evaluation: upper mobility impairments, lower mobility impairments, hardness of hearing,
deafness, low vision, blindness, speech impairments, and cognitive impairments. Within each category
of functional limitation there are a variety of subtypes and levels of disability. For example, in the
upper mobility category, someone may have a hand tremor that impacts fine motor control, a muscular
abnormality that impacts strength, or an injury that limits reach. A user’s level of experience may also
have a significant impact on the outcome of the evaluation. The accessibility problems experienced by
someone who has been blind since birth may be significantly different than the accessibility problems
experienced by someone who has recently lost his or her vision.
Derived assessment methods, such as functional assessments, checklist evaluations, or expert
inspections, do not seek to measure human performance directly. Instead, such techniques are used to
determine the extent to which a device meets formal or informal guidelines that have been associated
with desired human performance objectives. A checklist evaluation, also known as a heuristic
evaluation, is employed where a considerable body of knowledge about a specific domain of human
performance relative to the product of interest exists (Nielsen, 1994). For example, a significant
amount of research related to the readability of computer displays under normal lighting conditions has
been performed. Researchers have analyzed the existing data and developed design guidance for the
selection of a font size for important information on computer displays that applies to users with
normal or correctable vision. Instead of performing detailed user-in-the-loop test procedures,
evaluators can simply measure the font size and viewing distance, and compare the results to
established design guidelines.
Occasionally, our knowledge about a specific domain is either incomplete, or the results of
previous research are so ambiguous or conflicting, that the development of useful guidelines is
impossible. Also, it may be impossible to formulate design guidelines without knowledge of the
operational environment of the device. For example, many studies have expressed the importance of
providing sufficient contrast on labels or other display items that must be discernible from the
background. However, it is very difficult to specify a minimum contrast ratio without knowing details
about the device and the context in which it will be used. The contrast ratio of a label must be greater
if the label is to be read outdoors as opposed to a label that will only be read indoors. Also, the size,
coloration, and position of the control can impact contrast requirements. Therefore, it may be
impossible to develop a design guideline that is specific enough to objectively test. In such cases,
35
other methods of evaluating accessibility must be employed or the expertise of the evaluator must be
relied upon.
Given the challenges of sampling from a diverse population and our somewhat limited
knowledge of designing for people with disabilities, it is necessary to combine data from both direct
measurement and derived assessment evaluation techniques to obtain the desired degree of confidence
in the evaluation results. Gaps in our knowledge in designing for people with disabilities can be
addressed by performing user testing. Similarly, the sampling issues associated with human
performance testing for accessibility can be addressed by using the derived assessment results to
eliminate the need to test some populations and focus the assessment on others.
A variety of evaluation techniques are available to the evaluator interested in measuring
accessibility. However, there are five main high-level activities common to all accessibility
evaluations:





Define evaluation goals
Perform a task analysis
Develop a Method of Test
Perform the evaluations
Document the results
How do I define the goals of the evaluation?
It is critical to the success of the evaluation effort to determine the specific goals of the
evaluation and the intended audience of the test data prior to designing the data collection method.
Failure to consider the goal of the evaluation may result in incomplete data or wasted efforts. If the
goal of the evaluation is to measure conformance with a specific set of requirements, such as the
VVSG, then the accessibility evaluation should focus on checklist evaluation techniques supported by
user testing when needed. If the evaluation goal is to determine if a particular device is accessible to a
particular individual, then the evaluation can focus on user testing supported by a functional
assessment.
Evaluators may need to know which device among several similar devices is most accessible.
If asked to choose among several different options, evaluators should use empirical evidence of
accessibility whenever possible. For example, a variety of different desktop printers are available from
a variety of different manufactures. Each device can differ in terms of the accessibility features they
provide and the accessibility barriers inherent in the designs of the equipment. An accessibility
evaluation can assist someone in choosing a device that is most accessible to a given individual or a
class of individuals. In order to measure general accessibility and compare the results among multiple
products, both checklist evaluations and user testing are important.
Evaluators may also need to know how well the device conforms to a specific set of guidelines
or technical requirements. Commonly, evaluations are performed to measure the conformance with
the Section 508 standards for the purpose of preparing a Voluntary Product Accessibility Template
(VPAT). A VPAT is used to allow companies to document and communicate the accessibility of a
particular product to a potential Federal Government customer. An accessibility evaluation for this
purpose relies heavily on the checklist evaluation approach; however, user testing may be required to
document conformance with some technical requirements.
36
Accessibility evaluations can also be used to advise design. If conducted as part of an iterative
design process, the results of an accessibility evaluation can be used to correct design deficiencies and
assist developers in selecting from a list of potential design alternatives. Both the checklist evaluation
and user testing play an important role in this type of evaluation.
Once the purpose of the evaluation has been determined, the evaluation goals should be
expressed as a desired outcome of the evaluation. Candidate evaluation goals might include the
following:




Measuring conformance with applicable standards of Section 508
Comparing the accessibility of product A to product B
Selecting a design alternative from among two candidate alternatives by selecting the
alternative that maximizes accessibility to users without vision
Determining the functional abilities required to operate the product
Proper evaluation goals formulated in terms of a desired outcome will assist in the selection of
accessibility evaluation techniques.
How do I perform a task analysis?
Task identification is the first, and most important, step in conducting a task analysis (Sanders
& McCormick, 1993; Drury, Paramore, Van Cott, Grey, & Corlett, 1988). The focus of the assessment
is the system comprised of the user and the device that the user is intended to operate. Task
identification is a structured analysis of all the user actions that are associated with the usage of a
particular device or class of devices (Drury, 1983). There are three potential sources of information
that the analyst may use to identify tasks. The first source is to observe users as they operate the
device (or a comparable device). Users perform tasks with a device because of the requirements of
their jobs - and thus it is beneficial to understand how users interact with devices as part of their
everyday routine. By observing how users interact with a device in realistic work environments,
researchers can decompose user actions into specific tasks.
The second source of information is technical documentation associated with the device, such
as user manuals and training materials. These documents may provide information on tasks that are
not readily observed, such as tasks performed only during initial setup of the device or in response to
some malfunction. These documents may also provide information on alternative methods of
performing tasks, beyond what current users may be observed doing. However, it should be noted that
most documentation is centered on the technology, not on the user. Researchers should use caution
when relying on technical documentation as the sole source of task analysis information. It is very
tempting to derive a list of product features from the technical documents rather than a list of user
tasks.
The third source of information about user tasks comes from the development of design
reference scenarios that describe how users are envisioned to use the device in various circumstances.
This source is particularly important when analyzing a new product or addition of new interfaces to an
existing product. In such cases, it is not possible to observe users interacting with the new device, and
technical documentation may still be immature. A design reference scenario is simply a narrative of
the key elements in the work environment, and the usage of the device in that environment. A typical
set of design reference scenarios would include perhaps 8-12 scenarios, each differing in the
circumstances or aspects of functionality that are used.
37
After completing the task identification, it is often useful to denote tasks that are associated
with maintenance, troubleshooting, or initial setup. Depending on the type of equipment and the
complexity of the device, maintenance, troubleshooting, and initial set-up tasks may be performed by
someone other than the end user, such as a technician or someone with specialized training.
Next, each task should be assigned a priority level. A prioritization of the task list should be
developed based on an estimate of the essential or core features of the device, versus advanced features
and features related to device set-up and maintenance (Fain & Folds, 2002). Priority may be divided
into three levels. Priority 1 tasks must be able to be successfully completed irrespective of impairment
in order for the product to be usable for all users. An inability to perform a priority 1 task because of
an impairment would likely severely limit the accessibility of the product under evaluation for users
with that impairment. Priority 2 tasks are secondary tasks that may be performed on an occasional
basis to access advanced functionality. The inability to perform a priority 2 task because of an
impairment, while not critical to the basic use of the product, may negate the value of advanced
features of the product. Priority 3 tasks are tertiary tasks that are not necessarily performed by all users
of the device, but must be performable by some operator on occasion. These tasks include initial setup
tasks that are not ordinarily repeated, major troubleshooting tasks, and major maintenance tasks that
users are expected to perform, albeit infrequently. The inability to perform a priority 3 task because of
an impairment would not affect the basic accessibility of the product unless the product is to be used
by a single user (or group of users all sharing the same impairment). Maintenance tasks may be
associated with any priority level. Some routine maintenance tasks that any user might perform, such
as loading paper, are judged to be priority level 1. Maintenance tasks that are performed rarely or only
by specially-trained users, however, are judged to be priority level 3.
The final step in the process is the development of task scenarios. The basic unit or component
of an accessibility test is a task scenario (or task sequence). Task scenarios are operationally realistic
arrangements of tasks. A series of scenarios (approximately 8-12) that exercises the tasks identified in
the task analysis should be developed. Repetition of key tasks is encouraged. The set of scenarios
should fully exercise the user interface components of the device. For simple devices, the set of
scenarios may include every single task identified during the task identification stage of the analysis.
For more complex devices, it is acceptable to formulate the scenarios so that a representative set of
tasks are selected from the full set of tasks.
When developing the scenarios, the evaluator should be mindful that every component of
interest on the device is covered. For example, the task scenarios should cover tasks that result in
common errors in addition to the more routine tasks. Some of the task conditions may have to be
simulated. For example, when evaluating a voting device, it is important to evaluate the error
notifications given on the review screen that appears after all selections have been made. An overvoting error can be produced (with behind-the-scenes software, for example) and the participant can be
instructed to go through all of the steps involved in correcting the error without actually having to
induce the error.
How do I develop a Method of Test?
A Method of Test (MOT) document should be created to establish the plan for addressing the
test objectives. The MOT document serves as a coordination tool amongst evaluators and forces those
performing the evaluation to consider the details of the evaluation prior to data collection.
The first section of the MOT should identify the reason for performing the test, and the system
that is being tested. Evaluators should pay careful attention to documenting the configuration of the
38
system. Any additional software or other modifications above the baseline configuration of the system
should be clearly noted. In addition, each test objective should be described, and a summary of
planned test sessions should be outlined. The second section should describe, in considerable detail,
the methods that will be used in performing the test. Evaluation goals should be stated clearly. In
addition, the MOT should outline how the evaluation will produce an outcome that meets the stated
goals.
The number and type of participants that are required by the test should be described. For
example, a test objective that involves user-in-the-loop testing should describe the specific functional
limitations that are required to participate in the test. Care should be taken to choose participants that
represent 1) the user population that is expected to interact with the products and 2) the variety of
functional limitations that the test is designed to consider. For example, when studying users with low
vision, it is desirable to consider several different levels of general impairment as well as specific
impairments, such as central field degeneration, in order to evaluate a more complete range of
impairments. In addition to participant requirements, any specialized equipment required for the
evaluation, such as force meters or assistive technologies, should be specified. Additional resources,
such as personal attendants for participants or sign language interpreters for those without hearing,
should be documented.
A step-by-step documentation of the procedure should be included in the MOT. The procedure
section should document participant training, the particular tasks that the participant will perform, and
any debriefing activities performed as a part of the test. A detailed description of the data that will be
recorded should be provided. Common usability metrics for voting machines, which also may be
informative for accessibility evaluations, include effectiveness, efficiency, and satisfaction (Campbell
& Byrne, 2009; Laskowski, Autry, Cugini, Killam, & Yen, 2004; NIST, 2007). Effectiveness and
efficiency are objective measures, whereas satisfaction is a subjective measure. Measures of
effectiveness can be derived from the number of errors committed by a user, the number of help
requests, and the number of tasks successfully completed. A measure of efficiency is derived from the
time required to complete a given set of tasks. A measure of satisfaction is derived from the user’s
subjective responses to a series of questions or statements.
The effectiveness of a voting machine is arguably the most important performance measure
(NIST, 2007), and it is often measured simply by computing error rate. Errors include unintentional
votes for candidates and residual votes (i.e., voting for zero or two candidates in a race). The
correlation between unintentional vote rates and residual vote rates have been found to be low and
statistically insignificant (Campbell & Byrne, 2009), which suggests that evaluators should consider
using both measures.
Methods of data reduction and analysis should describe how the data will be interpreted. If any
statistical analyses are planned, a description of each analysis should be included. The criteria
specifying how the evaluator will determine if a test objective has been met should be included as well.
For example, the test criteria might be that all of the checklist evaluation items must receive a pass
rating, at least sixty percent of the users must be able to operate the device without experiencing
difficulty, and no more than ten percent of participants should experience extreme difficulty when
operating the device.
39
Which evaluation techniques should I use?
The three main accessibility evaluation techniques are functional assessments, checklist
evaluations, and user testing. The accessibility evaluator will likely combine two or more of the
techniques described below to produce results that meet the goals of the evaluation.
Functional Assessment. The goal of a functional assessment is to determine the human
abilities needed to interact with the device. The functional assessment should be performed at the task
level. Required abilities should be identified for each task.
For example, selecting a menu item from the control panel of a machine requires vision in
order to identify the menu item label, reach in order to select the item, fine motor control in order to
select the button associated with the menu item, and strength to press the button. For the vision
requirement, the evaluator should identify the font size of the label and determine the position of the
operator during normal operation of the device. Based on these observations, the required level of
visual acuity can be calculated. The evaluator might determine that in order to complete the menu item
selection task, operators should have 20/70 or better visual acuity based on the size of the label and the
position of the operator. Similarly, the fine motor control ability can be assessed by measuring the
target and the positions of the surrounding controls. The activation of a small button with many
closely-arranged surrounding controls will require more fine motor control ability than a large button
with fewer surrounding controls. The evaluator might express fine motor control requirements in
terms of a user’s ability to select a 0.2 cm button from a grid of buttons arranged 0.5 cm apart.
Strength requirements can be measured directly. Using a calibrated force meter, the evaluator can take
direct measurements of the force required to activate the button associated with the menu item
selection. Reach requirements can also be assessed by direct measurement. Therefore, the evaluator
might say that the functional abilities required to perform a menu selection from the control panel are
20/70 or better vision, the ability to select 0.2 cm buttons from a grid of buttons spaced 0.5 cm apart, a
reach of 10 cm, and the ability to apply 0.2 lbs of force on a 0.2 cm button. Notice that the task did not
require hearing or the ability to speak.
The results of the assessment are used in determining the focus and scope of user testing. If the
device does not require hearing to operate, then there is no reason to test users that are deaf or users
with limited hearing. Also, if the evaluator identifies a number of instances where a particular ability
is exercised, it might be useful to focus user testing on the collection of data relative to that ability. For
example, if the functional ability testing identifies instances where reach requirements may exceed the
abilities of seated users, it might be useful to test a number of users in a variety of personal mobility
aids. However, if the functional ability for a given task clearly exceeds the capabilities of a given user
population, there is little reason to proceed with testing of associated tasks. For example, given that
the task of selecting a menu item from a control panel requires at least partial vision, there is little
reason to test a user without vision on that particular task.
An important exception to this general rule occurs when the goal of the evaluation is to provide
design feedback. The evaluator may want to know if the device would be accessible if obvious
accessibility problems are removed. In this instance, the evaluator would simply skip or assist the user
in the performance of task components that would otherwise be extremely difficult or impossible to
perform for a given user group. In the case of an evaluation involving a user who is blind, the
evaluator would assist the user by reading the menu label or perhaps directing the user to the position
of the associated control. By skipping only the task components associated with obvious accessibility
deficiencies, the evaluator can collect data about the remainder of the task components. A designer
40
could use the data to determine if additional accessibility interventions are required beyond the
correction of the obvious problems.
In addition, the results of the functional assessment can be used during the checklist evaluation
when the evaluation item is based on a human performance measurement. For example, when
designing for someone with limited strength, an important accessibility guideline is to not require the
user to apply more than five pounds of linear force to operate the equipment.
Finally, the functional assessment might be an important point of data in and of itself. By
collapsing the requirements of functional ability across all of the tasks, the evaluator can identify the
minimum set of abilities required to operate the device. This data might help consumers select
products that match their own abilities.
Checklist Evaluation. In order to perform a checklist evaluation, the evaluator must be skilled
in using measurement equipment such as torque meters and force meters, when necessary, as well as
the general operation of the device to be evaluated. Also, the evaluator must be prepared to use sound
judgment and apply accepted principles when checklist items are ambiguous or ill-defined. In order to
complete the evaluation, the evaluator first selects the requirements, standards, and design guidelines
that both meet the evaluation goals and are applicable to the device. For example, if the primary
evaluation goal is to test conformance with Section 508, then the checklist evaluation should be
comprised of applicable Section 508 standards and related design guidance. If the goal of the
evaluation is to measure accessibility in general, then the evaluator should select a broad range of
recognized design guidelines, standards and principles. Potential checklist items can be derived from
the accessibility literature found in peer reviewed journal articles, standards found in government
regulations such as Section 508, and sound human factors design principles.
The United States Department of Justice published an accessibility checklist for polling places
(DoJ, 2004). A similar checklist was published by the Wisconsin Government Accountability Board
(Disability Rights Wisconsin, 2011). The checklists address issues associated with physical access to
the polling place and voting machine, but they do not address the accessibility of the voting machine
interface.
A checklist for the usability of ballots is available from the Brennan Center for Justice (Norden,
et al., 2008). Although the usability checklist does not specifically address accessibility issues,
adherence to its usability principles will improve accessibility to users with disabilities. Ballot
designers should consider using an accessibility checklist that is specific to the voting machine on
which the ballot will be displayed.
Once the list of checklist items is defined, the evaluator should define a method of evaluating
each checklist item. Methods of evaluation may include direct observation, measurement, or expert
inspection. Direct observation may be used when the checklist item evaluation method requires the
evaluator to inspect the device for a particular design element or characteristic. For example, a
checklist item derived from a guideline that recommends that the use of controls that required pinching
be avoided might instruct the evaluator to inspect the device for such controls. The evaluator may
determine that the device passes or fails the checklist criteria by simply inspecting the device to see if a
control that requires pinching is present. Other checklist items may require the use of a tool in order to
obtain some useful measurement. For example, a checklist item related to a guideline that
recommends that users not be required to apply more than 5.0 pounds of linear force might instruct the
evaluator to measure all linear control actions using a force meter. If any of the control action force
requirements exceed 5.0 pounds for proper operation, the evaluator should fail the checklist item.
41
Some checklist items rely on the expertise of the evaluator for proper determination of a pass/fail
rating. If the checklist item is ambiguous or ill-defined, the evaluator must make a reasonable
interpretation of the guideline based on prior experience and apply domain knowledge in order to make
a pass/fail judgment. If a checklist item derived from a guideline requiring sufficient contrast is
created, it is up to the evaluator to determine what sufficient contrast is. The adequacy of the contrast
will, in part, depend on the operating environment of the equipment and the size and coloration of the
user interface elements.
After the evaluation criteria have been defined for each checklist item, the evaluator performs
the evaluation by assigning a pass/fail rating to each checklist item. The evaluator should record all
measurements and keep a log of observations made during the evaluation. If the evaluation criteria
require the judgment of the evaluator, the evaluator should defend his or her judgment by documenting
the thought process used to arrive at the conclusion. After the evaluation is complete, a second
qualified evaluator should review the first evaluator’s findings.
User Testing. In some ways, user testing for accessibility is very similar to user testing for
usability. The evaluators should be properly trained in overall evaluation methodology, the specific
protocol for a given test, and the special concerns that arise when dealing with users with special
needs. The evaluator should be very well acquainted with the device under evaluation and should be
able to recognize and if possible rectify unusual problems that may arise, irrespective of whether the
problem is a consequence of a participant’s behavior. For example, if the device is a direct recording
electronic voting machine, the evaluator should know how to handle smart card reading errors, how to
restore the system to its proper configuration for the test, and whether the machine is malfunctioning in
a way that will render the test invalid.
The evaluation facility should be configured according to the requirements of the device and
the intended user population that will be participating in the evaluation. For example, if users are
expected to utilize assistive technologies during the course of the test, the device should be configured
to facilitate the use of specialized equipment. When testing software or hardware devices in an
integrated network environment, assistive technology software, such as screen readers or voice
recognition software, should be installed and configured prior to performing the test. In addition,
proper accommodations must be provided for participants with disabilities. These accommodations
include the physical space and any functional accommodations that are appropriate for the impairments
of the user. For example, access to the building and maneuvering space around the device should be
considered if the intended user population includes persons who use wheelchairs. If required, space
should also be provided for personal assistants or sign language interpreters.
In general, observers should not be present in the evaluation room while accessibility
evaluations are taking place, unless such observers are playing a direct role in the evaluation protocol
(e.g., watching for certain errors). Customer personnel and other interested parties may be
accommodated by allowing them to observe video feeds from the evaluation room. Video cameras
should be positioned so as to give a good view of the subject and a good view of the device during the
test. An audio recording should be made for the purpose of crosschecking key event reports. The
video recording should have a time of day counter, including seconds that can be used to cross-check
task performance times.
Accessibility evaluations should be conducted using participants who are properly trained and
representative of the user population of interest. Although ease of access to users is always a
consideration, care should be taken to ensure that the “user population of interest” is not defined based
42
on the types of impairments that happen to be conveniently present. A typical accessibility evaluation
might include representatives from the following user populations:







Users with upper mobility limitations, including users with limited strength, limited
reach, and limited fine motor control
Users with lower mobility limitations, including users who use a manual wheelchair, a
powered wheelchair, and users who use personal mobility aids such as scooters
Users who are hard of hearing
Users who are deaf
Users who have visual limitations including users who are color blind, users with poor
visual acuity, and users with central field obstructions
Users who are blind
Users who have speech limitations
Users with cognitive limitations should also be tested, if possible. Cognitive limitations are
increasingly common among veterans of Operation Iraqi Freedom and Operation Enduring Freedom,
due to high incidence of traumatic brain injury (TBI). Cognitive limitations that are commonly
associated with TBI include attention deficit, limited working memory, and language difficulty.
Because TBI symptoms depend on which brain areas are injured, it may be difficult to obtain a
representative sample of users. If the evaluator desires a representative sample, he or she should not
treat potential participants with TBI as a homogenous group. In general, it is best to test each
participant separately. Evaluation sessions should be scheduled so that the pace of the session is not
hurried, allowing participants to take breaks as desired.
Proper informed consent must be obtained before proceeding with any aspect of the test. The
participants’ rights to discontinue participation at any time should be respected. Care should be given
to ensure that each participant is fully capable of providing informed consent. Obtaining proper
informed consent is especially critical when working with users with certain cognitive impairments.
Evaluators should consult with a local research university’s Institute Review Board’s guidelines for
obtaining informed consent from users with disabilities.
Pilot testing of the test methods and evaluation instruments should be conducted before
proceeding with the evaluation. Internal pilot testing is often less formal than external pilot testing,
and is generally performed with participants recruited from within the organization performing the
testing. Documents, methods, and other materials should be updated as appropriate in accordance with
the pilot testing results. Internal pilot testing need not be performed with actual or simulated
impairments, although it is beneficial to do so when possible. External pilot testing should be
conducted with users who have some impairment of interest and are not part of the evaluation team.
Again, documents, methods, and other materials should be refined as appropriate in accordance with
the pilot testing results.
A test session should begin with participant orientation and training. The nature of the training
is largely determined by the extent to which the device’s learnability or intuitiveness is of interest. If
the device’s learnability or intuitiveness is not of interest, participants should be provided detailed
training on the use of the device. Most accessibility evaluations will fall into the second category. The
evaluator should point out each of the components that will be evaluated during the test session.
Training should also include instruction and practice on key event reporting. The evaluator should also
inform the participant about the purpose of the test and how the data collected during the test will be
43
used. If an assistant (e.g., a sign-language interpreter) accompanies the participant, the individual
giving the assistance should also be properly briefed about the procedures to be followed.
The evaluation should begin with the evaluator telling the participant about the scenario that is
going to be performed. Tests of voting device usability are conducted with either the directed or
undirected approach. In the directed approach, participants are given a list of candidates for whom
they are instructed to vote. In the undirected approach, participants are allowed to make their own
selections. The directed approach enables experimenters to easily identified mistakes. The VVSG
(2007) specifies that voting devices should be tested with the directed approach.
Once the background for the scenario has been described to the participant, the evaluator
should guide the participant through the scenario as required. There may be a tendency for an
evaluator to want to assist the participant in the completion of a task. Care should be exercised to
ensure that any assistance offered by the evaluator does not unduly affect the outcome of the
evaluation.
At the beginning of a task, there is a clear “start” event - which can be generated by the
evaluator, by the participant, or by the device. The evaluator or an observer should use a stopwatch to
measure task performance time. The end of the task is similarly identified by a clear event, which can
be a device outcome, a participant-generated event, or an evaluator-generated event.
The evaluator should be at the particularly sensitive to the fact that a given device may not be
very usable for all users with disabilities, and that this could lead to some degree of frustration on the
user’s part. Evaluators should be prepared to intervene and move on to the next task to avoid
unnecessary frustration. In general, the test session should be conducted with the evaluator directing
the participant in the performance of the tasks.
The thinking-aloud approach to usability testing (Rubin, 1994; Dumas & Redish, 1993), while
popular, is not particularly well suited to accessibility evaluations because it may create unnatural task
performance demands that affect the validity of task performance measures (not to mention that
thinking-aloud data is of questionable validity). The thinking-aloud approach also makes it difficult to
collect data when testing individuals with speech impairments. As an alternative to continuous
thinking aloud, which emphasizes the stream of consciousness of the participant, the key-event
reporting method should be used. In this method, certain key events that are of interest to the
evaluation are identified in advance. The evaluator should brief the participant about the key events
and ask the subject to report them when they occur. Examples of key events that are often of interest
are as follows:
1.
2.
3.
4.
5.
6.
“I can’t find X”
“I can’t figure out how to do Y”
“I didn’t expect that to happen”
“I see that I have made an error” or “I didn’t mean to do that”
“I don’t know why that happened”
“I don’t know what to do next”
When the tasks associated with a given component of the device have been completed (or at
whatever time specified by the test procedure), a user rating of accessibility should be collected. The
rating should ordinarily be made using a Likert-type scale with an even number of rating points, unless
there is a clear need to have a neutral point in the scale. A four-point scale is preferred unless there is a
strong reason to use more scale points. Anchors are along the lines of:
44
1.
2.
3.
4.
= Completely unacceptable
= Marginally unacceptable
= Marginally acceptable
= Completely acceptable
The System Usability Scale (SUS) is commonly implemented in usability evaluations of a
variety of devices (Brook, 1996; Greene, Byrne, & Everett, 2006), and could also be applied to
accessibility evaluations of voting devices. The SUS is a standardized scale comprising 10 statements
to which participants indicate agreement on five point scales. A composite score is calculated from the
10 items. Note that the traditional SUS uses a five point scale, which enables participants to give a
neutral response on each item. Because such responses limit the usefulness of the data, evaluators
should consider modifying the scale to include only four response options.
The evaluator should record any pertinent observations that are made during the test session.
For example, if the participant makes a mistake that is obvious to the evaluator but is not part of a
participant comment, then the evaluator should make a note of the occurrence. The evaluator should
also make a note of any system malfunction or other event that might influence the interpretation of the
test results. For example, if the participant was inadvertently interrupted during the performance of the
test, then the evaluator should note the interruption. Important events that are not necessarily
verbalized should also be recorded. These events might include the beginning and end of task
sequences, the number and type of errors, or a record of tasks that are successfully completed.
The final activity in the test session should be a debriefing of the participant. The debriefing
should include opportunity for the subject to share any comments on the device and/or the test
methods. If possible and appropriate, the debriefing should include a quick review of the session,
including a summary of the subject’s performance (e.g., “You were able to complete 4 of the 5 tasks.
The task you did not complete was retrieving the printout from the output slot.”). This review may
help prompt the participant to make insightful comments. All serious comments offered by the
participant during this debrief should be fully documented. Ad hoc comments made by the participant
in earlier stages of the evaluation session, but not repeated by the participant during the debriefing,
should be documented if the evaluator judges them to be important and germane to the accessibility of
the product.
Data analysis involves the computation of task durations and the tabulations of user comments
and errors. Each error should be documented and categorized. A summary tabulation of errors by
error type should be generated. Analysis of data that involves interpretation of user comments should
be cross-checked between analysts.
How should I document the results of the evaluation?
The final step in the process is to integrate the results of the evaluation and document the
findings. Checklist items should be documented alongside with user test results when possible. For
example, when documenting a problem with retrieving the smart card from a slot on a voting device,
the evaluator might organize the results as follows:
The smart card is only ejected 0.2” from the slot, which does not provide sufficient grasping
area. Five out of eight users with upper mobility limitations could not retrieve the smart card
from the slot without assistance.
45
If the purpose of the evaluation is to provide design feedback, the results of the evaluation
should be organized by device component. The evaluator should precisely describe accessibility issues
for each disability type that are associated with each device component. For example, the evaluator
might offer the following feedback to designers:
Touchscreen buttons are not easily identifiable by touch, making it difficult for visually
impaired users to identify key control elements. Users with fine motor control limitations will
find the touchscreen difficult to use because of the size and spacing of the buttons on the
screen.
If the purpose of the evaluation is to document the general accessibility of the device, then the
results should be reported by disability type and task priority. For example, the evaluator might offer
the following observations:
Blind users, in general, found the device very difficult to use because no alternative to the
touchscreen was provided. Users without vision were able to complete 20% of the priority 1
tasks, 16% of the priority 2 tasks, and 10% of the priority 3 tasks.
Regardless of how the results are organized, key accessibility issues should be documented in
the conclusion of the evaluation report. It is important to clearly describe the accessibility issue and
the population that will likely be impacted by the issue. When possible, illustrations should be used to
communicate key points.
46
Accessibility Issues and Potential Solutions
Absentee voting has potential accessibility issues in the following areas:




Absentee voting process
o Obtaining the ballot
o Reading the ballot
o Marking the ballot
o Returning the ballot
Voting technology
o Physical ability factors
o Audio and speech output
Ballot design and presentation
o Instructions and help
o User interface design
o Information presentation
o Write-in voting
o Ballot review
Policy barriers
The following sections contain detailed information about common accessibility issues in each
of these areas. In addition, the populations impacted by the accessibility issues are discussed, and
potential solutions are proposed.
Context for Potential Solutions
Many of the proposed solutions proceed from the assumption that some basic capabilities are in
place to support accessible voting for a particular population that was studied to identify the issues
associated with absentee voting. The population surveyed included recently wounded military
veterans residing in a VA hospital or similar facility, although many of the issues identified also apply
to other individuals who have recently become disabled or have similar injuries and limitations. Many
of these individuals reside in long-term care/rehabilitation facilities and must vote absentee because
they do not currently reside in the state with which they are registered to vote.
Underlying many of the solutions presented below is the premise that appropriate blank ballots
could be provided to absentee voters electronically, in a flexible format that could be displayed in a
variety of different ways to accommodate the needs of voters with various types of disabilities. If
absentee ballots could be distributed electronically, a number of the issues described in the next section
could be alleviated, including obtaining, marking, and submitting the ballot. This streamlined process
may be similar to the image shown in Figure 11.
47
Figure 11: Recommended process for absentee voting.
In addition to availability of electronic ballots, another premise underlying some of the
proposed solutions is that the ballot could be marked using an electronic “ballot marking device” that
provides an accessible way to interact with the ballot (see Figure 12).
Figure 12: iPad absentee voting concept.
48
Additional details about the functionality and use of the ballot marking device are presented
below.

Voters will use an iPad-based (or similar hardware) portable absentee voting system to vote.
This includes obtaining a ballot, marking it electronically in an accessible way, and submitting
the ballot (printing it out and mailing it, or whatever other process policy requires them to
follow).

A “voting facilitator” familiar with the use of the absentee voting system will bring the device
to the voter’s hospital room, either as a handheld device or installed on an adjustable stand
compatible with beds and/or wheelchairs.

The voter should be able to mark the ballot mostly independently, but the voting facilitator may
set the system up for the voter, assist with problems during voting, and help with submitting the
ballot (printing, physical handling of the ballot, etc.).

There may also be a “voting room” available at the facility, with one or more of the portable
absentee voting systems installed as voting kiosks, so that voters who are able to and/or prefer
to vote in that environment could go to the voting room and use the device there. Voting
facilitators will still be available to assist, but one voting facilitator may be responsible for
several systems instead of the one-on-one assistance provided when bringing the system to a
room.
49
Issues with the Absentee Voting Process
Obtaining the Ballot
In many states, voters must apply for an absentee ballot. Individuals with impaired mobility
may need to apply remotely rather than in-person. In other states, voters may download the absentee
ballot from a government website.
Voters cannot sign the application form.
Detailed Description: Many applications for absentee ballots require a voter signature. This is
sometimes used as a fraud prevention measure – the signature on the ballot is matched to the
signature on the application. However, some voters with upper mobility impairments are
unable to provide a signature.
Population Impacted: Voters with upper mobility impairments; voters who are blind.
Potential Solutions: Allow voters to electronically sign applications, using other identifiable
information (such as a social security number). If registering online, voters could use their
social security number or a secure password to verify their identity. This would allow them to
“sign” their own application and not rely on outside resources for a signature.
Allow voters to obtain permanent absentee status. Voters could obtain permanent absentee
status by applying in-person with a legal form of identification. To prevent fraud, the voter
could be given a personal identification number to place on their ballots.
Applicable Guidelines:
Section 508 – 1194.31
ICT – 302.7
VSS – 2.2.7.2(i)
50
Reading the Ballot
Many voters who have low vision or are blind are unable to use mail-in ballots without
assistance.
Text-only paper ballots cannot be read by voters with visual impairments.
Detailed Description: Absentee ballots are typically text-only paper documents. If voters are
unable to see the text, they will be unable to vote independently. Although voters may ask for
assistance in reading the ballot, their privacy would be compromised by such assistance.
Braille absentee ballots are offered upon request in some states. However, voters who have
recently sustained injuries that affect their visual abilities are unlikely to be able to read Braille
ballots.
Population Impacted: Voters who are blind; voters with low vision.
Potential Solutions: Provide ballots online that can be marked electronically. Electronic
ballots can be used with assistive technologies such as screen readers that do not require the
ability to read Braille. Security concerns might prohibit ballot-casting via the internet. As an
alternative to the internet, voters could be required to print and mail the ballot after marking it
with assistive technology.
Provide a Braille absentee ballot. Braille ballots should be made available upon request in all
states. (However, as mentioned above, the percentage of users with visual impairments who
are able to read Braille is fairly low, so other methods of providing accessible ballots are also
necessary.)
Applicable Guidelines:
ICT – 302.1, 503.3
ADA-ABA – 703.3
51
Marking the Ballot
This section covers accessibility issues associated with marking and casting the ballot in
privacy.
Ballots are not provided in a format that can be marked electronically.
Detailed Description: In order to complete a paper ballot, voters with visual or upper mobility
impairments may have to seek assistance, preventing them from marking the ballot privately
and independently. Very little in the way of assistive technology is available to help users
mark a paper ballot. Stray marks on a paper ballot may make it difficult to determine the
voter’s intent, and paper ballots do not allow any error detection and notification for users.
Populations Impacted: Voters with upper mobility impairments; voters who are blind; voters
with low vision.
Potential Solution: Provide ballots in a format that can be marked electronically. Ballots that
can be marked electronically would allow voters to use screen readers and other assistive
technologies to mark the ballot without assistance. Electronic marking of ballots would also
eliminate issues resulting from stray marks on paper ballots, and could reduce errors by
allowing notification of undervote and overvote conditions.
Due to security issues, actually casting a ballot electronically may not be feasible at this time,
but the ability to mark the ballot electronically and then print it out for submission via
allowable methods (mail, fax, etc.) would increase the accessibility of the ballot.
Applicable Guidelines:
Section 508 – 1194.31(f)
ICT – 302.1, 302.7, 503.3
VVSG – 3.2.2-A
HFDS – 8.15.8.18, 8.15.11.2.1
52
Voters who need assistance marking the ballot cannot vote privately.
Detailed Description: If voters are physically unable to mark the ballot, they may require
assistance from another person. Markings are made directly adjacent to candidate names on
typical ballots. Therefore, voter privacy is compromised when assistance is required.
Populations Impacted: Voters with upper mobility impairments; voters with low vision.
Potential Solution: Separate the list of candidates from the response fields. Candidates could
be listed on a sheet to be read by the voter. A separate response sheet – to be marked by the
voter’s assistant – could contain a list of arbitrary letters. The letters would be paired with the
candidates on the document viewed by the voter. The voter would be required to dictate only
the letter associated with their chosen candidate, thus ensuring privacy.
Applicable Guidelines:
VVSG – 3.2.3.1-A.1, 3.1.1(1)
Section 508 – 1194.31(a)
ICT – 407.10
53
Error feedback is not provided.
Detailed Description: Overvotes and undervotes are common errors that can be detected by
electronic voting systems used at polling places. Such error detection is not supported by paper
based mail-in ballots. Voters are less likely to correct mistakes if they do not receive error
feedback.
Population Impacted: All voters, especially those with cognitive disabilities.
Potential Solution: Provide an electronic absentee ballot. A web application that supports
overvote and undervote detection could be provided. The ballot need not be cast electronically
(it could be printed and mailed or faxed, if allowable).
Applicable Guidelines:
VSS – 2.2.7.2(h), 2.2.5.2.2(b)
VVSG – 3.2(a)(1)(A)(ii)
HFDS – 8.15.8.18, 8.15.11.2.1
ICT – 402.2
54
Paper-based vote verification must be visually inspected.
Detailed Description: After completing a ballot, voters should be able to review their votes to
ensure they will be cast as intended. This is often enabled by a paper-based record that is
printed from the voting device. Such paper-based records are only accessible to those who can
read text.
Populations Impacted: Voters who are blind; voters with low vision; voters with cognitive
impairments, particularly those who have difficulty reading.
Potential Solution: Provide an electronic review screen. Electronic reviews can be used with
assistive technologies such as screen readers. Electronic ballots could be printed by the voting
device and read into a verification machine, which could be accessible to those with screen
readers of other assistive technologies.
Note: Many counties require the vote verification to occur on a separate device (other than the
voting device). Electronic ballots are printed by the voting device and read into a verification
machine, which could be accessible. Fraud is prevented unless both machines are tampered
with.
Applicable Guidelines:
VVSG – 3.2(a)(1)(A)(i)
ICT – 302.1-2, 302.7, 407.10, 407.11
HAVA 2002 – para III(a)(3)(B), para III(a)(3)(C)
55
Returning the Ballot
This section covers accessibility issues associated with submitting the ballot.
Physical submission of the ballot is difficult for some users.
Detailed Description: Most current absentee voting requires at least some physical handling of
the completed ballot (for example, to print out, sign, and mail or scan the completed ballot).
However, physical manipulation of ballots can be difficult or impossible for some users with
upper mobility impairments. Furthermore, it can be difficult for users with visual impairments
to sign printed ballots, and it can be difficult for users with cognitive impairments to follow
multiple steps required to prepare a ballot for submission. Reducing or eliminating
requirements for physical manipulation of paper ballots would make voting easier for voters
with disabilities.
Populations Impacted: Voters with upper mobility impairments; voters who are blind; voters
with low vision; voters with cognitive impairments.
Potential Solutions: Reduce or eliminate requirements for physical manipulation of paper
ballots. The ballot design should reduce or eliminate requirements for physical manipulations
as much as possible, such as paper folding, and limit the number of steps required to submit the
ballot.
Allow electronic submission of ballots. Currently, no states currently permit ballots to be
submitted electronically over the Internet. However, current technologies are moving in the
direction where this will soon be a possibility. For example, West Virginia piloted the use of
Internet voting in 5 counties in 2010 for UOCAVA voters (EAC, 2011). This would eliminate
the physical manipulation of paper ballots, reducing difficulty particularly for those with upper
mobility impairments.
Applicable Guidelines:
ICT – 302.7
Section 508 – 1194.31(e)
56
Correct postage is not indicated.
Detailed Description: Most absentee ballots contain many envelopes and inserts, which may
require additional postage stamps.
Populations Impacted: All absentee voters, especially those with cognitive impairments.
Potential Solution: Indicate the required amount of postage on the outer envelope. Weigh the
materials that are to be mailed by the voter, and display the corresponding postage amount on
the envelope where postage will be affixed.
Figure 13: Return envelopes with and without additional postage amount displayed.
57
Issues with Voting Technology
Physical Ability Factors
Physical ability factors that affect physical interaction with a voting device may include reach,
vision, muscle strength, and range of motion. These factors may affect voters’ ability to interact
effectively with the voting device.
The voting device does not provide or support necessary accessibility options.
Detailed Description: Voters may not have the ability to physically interact with a traditional
voting device. Voters without the use of their arms may not be able to physically touch the
voting device. These individuals may utilize other devices, such as the “sip-and-puff”, eye
tracking, head tracking, and speech recognition user interfaces. Voting devices that do not
include options for changing the user interface settings such as high contrast, font size, audio
output, may exclude some voters from having the opportunity to vote, particularly if they vote
absentee and do not have a personal aide or voter representative to assist them. The voting
system may not support write-in voting, or enable the voter to review their write-in input to
confirm that it meets voter’s intent.
Populations Impacted: Voters who are blind; voters with low vision; voters with upper
mobility impairments.
Potential solution: Provide multiple input methods for voting, including support for write-in
voting. Given the various types and severity of injuries, no single solution will be accessible
for all individuals with disabilities. Therefore, it is necessary to provide multiple methods of
input. Voting systems utilizing touchscreen displays must provide at least one other input
method that can be used by individuals with visual impairments and upper mobility
impairments. These methods may include speech recognition software or other software
applications, or external peripherals, such as a mouth stick, sip and puff, or a reaching aid.
Applicable Guidelines:
HAVA 2002 – para III(a)(3)(C)
VSS – 2.2.7.2(f)(2), 2.2.7.2(f)(3), 2.2.7.2(f)(4), 2.2.7.2(b)(4)
Section 508 – 1194.25(c), 1194.25(d), 1194.31(d)
ICT – 302.1,302.2-302.9, 402.2, 403.1, 407.10, 503.3, 603.3
58
The voting device is not portable.
Detailed Description: The voting device may be large stationary kiosk type units and not
portable. Voters who are bed ridden or unable to leave their homes are unable to physically go
to the polling place and vote. For these individuals, the primary voting option is to complete a
paper ballot which then gets mailed to a local election office. Paper ballots are not generally an
accessible technology, and if individuals are unable to fill them out on their own, they have to
compromise their privacy to receive the necessary assistance. This is also burdensome for
those with vision impairments. If the voting device is portable, then they may remain in their
residential facility while voting.
Populations Impacted: Voters with lower mobility impairments; voters who are blind; voters
with low vision.
Potential solution: Provide a portable voting device. Use of a tablet computer as the voting
device can be used by voters who are bed ridden, homebound, or living in a residential or
assisted living facility. The design needs to be flexible enough to accommodate the constraints
of a person with physical disabilities or vision impairments without being overly burdensome
to the user. The goal should be to design an optimal voting solution that considers such things
as: button configuration; size, shape, and form of the tactile controls; and the manner in which
the controls interact with the tablet computer or other technology.
Figure 14: Portable voting device.
Applicable Guidelines:
VVSG – 3.1.1(1), 2.2.7.2(f)(1), 2.2.7.2(f)(2), 2.2.7.2(f)(3), 2.2.7.2(f)(4), 2.4.3.3(b)
Section 508 – 1194.31(a),(b), (c), (d), (e), (f)
ICT – 302.1, 302.2, 302.7, 302.8, 407.2, 407.3, 407.5, 407.9, 407.10, 503.3
59
Voters may not be able to physically hold a portable voting device.
Detailed Description: Users with upper mobility impairments may not be able to physically
hold a portable voting device. The portable voting device may be mounted on a stand
compatible with the bed or wheelchair, which will allow the user to view and interact with the
voting device.
Population Impacted: Voters with upper mobility impairments.
Potential Solutions: Provide a mount or stand that holds the portable voting device. This
would allow voters to use the device without holding it. The mount could be similar to a
television wall mount or computer desk mount with an adjustable arm to allow the portable
device to be moved around once attached to bedside or wheelchair arm. The mounting brackets
should be easily attachable/detachable for ease of mounting onto bed or wheelchair.
Provide a mobile kiosk that allows voters to approach the device from a wheelchair or other
mobility aid. Designers must consider things such as the parallax in viewing displays (whether
sitting or standing).The kiosk would need to be adjustable such that those who need to stand
can see the screen as clearly as those in a wheelchair. Human Factors Design Standards
(HFDS) for workstation design would need to be adhered to for this type of device (see
applicable guidelines below).
Applicable Guidelines:
VVSG – 3.1.1(1), 2.2.7.2(f)(1), 2.2.7.2(f)(2), 2.2.7.2(f)(3), 2.2.7.2(f)(4), 2.4.3.3(b)
Section 508 – 1194.31(a),(b), (c), (d), (e), (f)
ICT – 302.1, 302.2, 302.7, 302.8, 407.2, 407.3, 407.5, 407.9, 407.10, 503.3
HFDS – 10.1.1, 10.1.2, 10.2.1
60
Some components of voting devices may be difficult for users with mobility impairments to
reach.
Detailed Description: Users of wheelchairs or other personal mobility devices may have
limited reach capabilities, when voting in a designated “voting room” at a residential facility.
Some users have power wheelchairs that allow them to raise and lower their seats to increase
their reach capabilities; however, many users have manual wheelchairs that do not allow for
these sorts of adjustments. Some wheelchair users are unable to shift their upper bodies,
limiting their access to only controls within arm’s length. Thus, it is important that the
components of a voting device that users must interact with are operable from a seated position,
without requiring excessive leaning or reaching.
Figure 15: The touchscreen of a voting device is too high, causing reach problems for a seated user.
Population Impacted: Voters with lower mobility impairments.
Potential Solutions: Position the operator controls in a position that ensures easy access by
all users.

The position of any operable control should be determined with respect to a vertical
plane that is 48 inches in length, centered on the operable control, and at the maximum
protrusion of the product within the 48 inch length.

Where any operable control is 10 inches or less behind the reference plane, the height
shall be 54 inches maximum (according to Section 508) or 48 inches maximum
(according to ADA-ABA) and 15 inches minimum above the floor. (See note below.)

Where any operable control is more than 10 inches and not more than 24 inches behind
the reference plane, the height shall be 46 inches maximum and 15 inches minimum
above the floor.

Operable controls shall not be more than 24 inches behind the reference plane.
61
Note that Section 508, which was based on older ADA guidelines, specifies a maximum height
of 54 inches. The newest ADA-ABA guidelines specify a maximum height of 48 inches, and a
forthcoming update to Section 508 is expected to follow suit.
Figure 16: The touchscreen of a voting device is low enough for a seated user to reach it.
Allow adjustment of component positions. Allow users to adjust the position of components to
meet their specific needs. Components could be adjustable between two or more discrete
positions, or freely within a range of positions.
Provide area for knee and toe clearance beneath the machine to allow a forward approach by
wheelchair users. Sufficient knee clearance will allow a forward-approaching voter in a
wheelchair to move closer to the machine. The knee clearance area beneath the machine
should extend from 9 inches to 27 inches above the floor, with a minimum width the 30 inches.
The toe clearance area should extend above the floor, with a depth of at least 17 inches and a
width of 30 inches.
Applicable Guidelines:
VSS – 2.2.7.1 (b)
VVSG – 3.3.5.1-B.3, 3.3.5.1-B.4
ICT – 407.13, 407.14, 407.15
ADA-ABA – 308.2.1, 308.2.2, 308.3.1, 308.3.2
Section 508 – 1194.25(j), 1194.31(f)
ISO/IEC 71 – 8.3.1
62
Some components of devices may be difficult for users with mobility impairments to see.
Detailed Description: If the display of a voting device is optimized for standing users, users
who are viewing the display from a wheelchair or other personal mobility device may have
difficulty seeing the contents of the display. Many LCD screens are optimized for viewing
from a particular angle, and demonstrate a sharp drop-off in contrast and readability when
viewed at angles that are off-axis. Also, voters who are seated in a wheelchair may be able to
reach a component, but may be unable to adequately see it. For example, a voter might be able
to press the keys on a keypad, but may be unable to see the key labels due to their orientation.
Population Impacted: Voters with lower mobility impairments.
Potential Solutions: Allow adjustment of component positions, including angle of the display.
A vertical tilt adjustment for the display will help to accommodate both seated and standing
users. Where possible, allow users to adjust the position of components to meet their specific
needs. Components could be adjustable between two or more discrete positions, or freely
within a range of positions. For example, a display that can be tilted vertically can better
accommodate both standing and seated users. Ensure that viewing envelope of the selected
display accommodates both standing and seated users. The viewing envelope of the display,
which describes the range of eye positions from which the contents of the display are visible,
can be determined via simple geometry based on the viewing angles of the display and the
height and angle at which the display is mounted. Ensure that the display’s viewing envelope
includes the eye positions of both seated and standing users.
Place components so that all users can see them. When designing a voting device, consider the
needs of both standing and seated users and attempt to place components in locations and
orientations that are visible for all users. For example, a display placed with the screen
perpendicular to the ground may be visible for both standing and seated users, although the
height of the display and the viewing envelope of the display must also be considered.
Provide a mobile kiosk that allows voters to approach the device from a wheelchair or other
mobility aid. Designers must consider things such as the parallax in viewing displays (whether
sitting or standing) and adjustability, such that those who need to stand can see the screen as
clearly as those in a wheelchair. Human Factors Design Standards (HFDS) for workstation
design would need to be adhered to for design of this type of device (see applicable guidelines
below).
Applicable Guidelines:
ADA-ABA – 308.2.1, 308.2.2, 308.3.1, 308.3.2, 707.7.1
Section 508 – 1194.25(j)
HFDS – 5.1.26, 5.11.1, 6.4.1.27
ISO/IEC 71 – 8.3.1
VVSG – 3.3.5-C
ICT – 407.15
63
Glare makes it difficult for some users to see the display.
Detailed Description: Glare on display screens is a general usability issue for all users.
However, standing users, or users without upper mobility problems, can more easily shield the
display with their bodies or adjust their viewing angle to avoid glare, whereas users in
wheelchairs may have more difficulty shielding the display or adjusting their viewing angle.
Also, because seated users may be looking up at a display, glare from overhead lights may be
more problematic.
Figure 17: The angle of the display results in glare for a seated user.
Populations Impacted: Voters with mobility impairments; voters who are blind; voters with
low vision.
Potential Solutions: Use an anti-glare display. Matte finish displays reduce glare by making
the surface of the display less reflective. Chemical coatings can also be used to reduce glare.
Provide an adjustable angle display. If the vertical tilt angle of the display is adjustable, users
can reposition the screen to reduce glare, instead of having to change their physical position
relative to the screen.
64
Figure 18: An adjustable display allows a seated user to adjust the display angle to reduce glare.
Use a hood to shield the display. A hood over the display can block direct light that causes
glare. However, a poorly designed hood might also block visibility for tall users.
Position the device in an area where glare is not an issue. Sunlight and overhead lighting are
common sources of glare. If possible, position voting booths in areas where glare is less likely,
such as an area that does not receive direct sunlight. Alternatively, reduce overhead lighting in
the vicinity of the machine or direct it away from the display.
Applicable Guidelines:
Section 255 – 1193.43(b)(2)(a)
HFDS – 5.2.3
ISO/IEC 71 – 8.4.2, 8.4.4
65
System response time to user input is slow.
Detailed Description: Users rely on timely feedback from the system in response to their
inputs. If feedback is not provided in a timely fashion, users may conclude that their input was
not accepted and try again, leading to multiple activation errors. This is a usability problem for
all users, but it may exacerbate difficulties for users who are more prone to making input errors
(e.g., users who are blind or who lack fine motor control).
Population Impacted: All voters.
Potential Solutions: Minimize system response lag time. The system should provide timely
feedback to the user. The system should provide some response to user input within 500 ms. If
the system response to a user input takes longer than 500 ms, an interim “in progress”
indication should be displayed to acknowledge that the input was received and is being
processed.
Allow easy recovery from errors. Eliminating lags that induce errors should be the first
priority, but designing for easy recovery from errors should they occur is also beneficial. The
interface design should prevent errors (such as overvotes) where possible, and should notify
users of potential errors (such as undervotes) and allow them to easily make changes if they
desire to do so. One way to achieve this is to provide a pop-up error notification as soon as the
error is detected, rather than at the end of the ballot marking process. The user could then
return directly to the location of the error to make changes when the error notification is
dismissed.
Applicable Guidelines:
Section 508 – 1194.25(b), 1194.31(f)
HFDS – 2.6.1, 6.3.1.5, 6.3.1.6, 8.8.2.24, 8.15.8.13, 8.15.11.1.13, 8.15.11.1.14
66
Audio and Speech Output
This section covers accessibility issues associated with audio and speech output as it relates to
voters hearing recorded audio for instructions, help, and casting of the ballot.
The voting device does not provide a headphone jack.
Detailed Description: Users with visual impairments often rely on audio output to interact
with voting devices. Interference from ambient noise may make it difficult for users to
perceive information provided via audio, especially for users who are hard of hearing. Use of
headphones allows users to hear audio output more clearly, and also enhances privacy.
However, some voting devices do not provide a headphone jack or headphone jack is not
standard size.
Populations Impacted: Voters who are blind; voters with low vision; voters who are hard of
hearing.
Potential solution: Provide a headphone jack so that users can connect personal headsets to
the device. The headphone jack should use a standard headphone connector (a 3.5 mm pin is
most common). External audio playback through speakers should be disabled when
headphones are connected. Ensure headphone jack is 1/8 inch stereo headphone jack to allow
individual voters to supply personal headsets. The use of a standard sized headphone jack
would allow voters who want to use their own headphones the ability to do so easily.
Applicable Guidelines:
VSS – 2.2.7(b)(6) or 2.2.7(b)(7)
ADA-ABA – 707.5
Section 508 – 1194.25(e), 1194.31(a), 1194.31(b), 1194.31(c), 1194.31(d)
Section 255 – 1193.43(e)(2)(f)
ICT – 302.1, 302.2 406.1, 407.10, 408.4
HFDS – 8.18.4.2
67
The placement of the headphone jack allows the headphone cord to interfere with use of the
device.
Detailed Description: Headphones are typically attached to voting devices by a cord. The
placement of the headphone jack could result in the headphone cord resting in an area where it
interferes with the use of the machine (for example, the cord may hang over the controls).
Moreover, if users must change positions to interact with different parts of the machine,
headphone cords may restrict range of motion or become unplugged during movement.
Populations Impacted: Voters who are hard of hearing; voters who are blind; voters who have
low vision; voters with upper mobility impairments; voters with lower mobility impairments.
Potential Solutions: Position the headphone jack in a location that allows sufficient range of
motion to perform all tasks without interfering with the use of the voting device. Consider the
positions that users, both seated and standing, will occupy when using the device, and place the
headphone jack in a location central to those locations. Also consider wheelchair users who
take either a forward or parallel approach. This will allow users the necessary range of motion
so they don’t accidentally unplug their headphones. For example, the headphone jack could be
placed on the front of the device directly below the smart card slot.
Provide support for wireless headphone connectivity. Wired connectors are still the most
common methods for connecting headphones, but wireless technologies such as Bluetooth are
becoming more prominent. Providing support for wireless headphone connectivity would
eliminate interference from headphone cords.
Figure 19: Headphone jack location allows the cord to interfere with use of the machine.
68
Figure 20: Headphone jack location prevents the cord from interfering with use of the machine.
Applicable Guidelines:
ADA-ABA – 308.2.1, 308.2.2, 308.3.1, 308.3.2
Section 508 – 1194.25(j)
HFDS – 6.4.1.12
69
The voting device does not support t-coil coupling.
Detailed Description: Users who rely on audio output from the voting device and who also
have limited hearing may need to use t-coils to hear audio output. Assistive devices such as
neckloops can transform electrical currents from headphone jacks to magnetic fluctuations.
These magnetic fluctuations are transformed into sound by t-coils located inside hearing aids.
The t-coils transform magnetic signals into sound. Most hearing aids allow users to switch
between microphone mode and t-coil mode. When in t-coil mode, the hearing aid transmits
sound only from the device connected to the neckloop (although it may also transmit noise
from interfering magnetic fields). Neckloops typically have 3.5 mm plugs that are compatible
with common audio output jacks. Silhouettes and ear links provide the same functionality as
neckloops. Neckloops are often used by individuals with hearing impairments – especially
those who also have vision impairments.
Voting devices are often equipped with forms of audio output that make the device accessible
to users with visual impairments. However, this audio output is inaccessible to users who also
have hearing impairments. Neckloops enable users with both vision and hearing impairments
to receive information from the machine, provided that the device is equipped with a 3.5mm
audio output jack, because the neckloops interface with the user’s hearing aid. The user can
adjust the neckloop to amplify only the sound coming from the device interface, so that
ambient background noise is filtered. Any type of verbal information, alerts, or feedback can
be provided through a neckloop.
Populations Impacted: Voters who are hard of hearing.
Potential solution: Equip the voting device with a t-coil-compatible headphone jack. The
headphone jack should be the standard size of 3.5mm.
Applicable Guidelines:
VVSG – 3.1.1(1), 3.3.3-C.2
VSS – 2.2.7.2(b)(1), 2.2.7.2(b)(2), 2.2.7.2(b)(3), 2.2.7.2(b)(4), 2.2.7.2(d)
Section 508 – 1194.25(e), 1194.25(f); 1194.31(a), 1194.31(b), 1194.31(c), 1194.31(d)
ICT – 302.1, 302.2, 302.4, 302.5
70
The volume level is insufficient.
Detailed Description: Some users may have difficulty hearing audio output at default volume
levels, particularly if the device is located in a noisy environment. Users with vision
impairments may rely exclusively on auditory information to use the device, so it is important
to ensure that the output volume can be adjusted to sufficient level.
Populations Impacted: Voters who are hard of hearing; voters who are blind; voters with low
vision.
Potential solution: Provide sufficient output volume and range of adjustment through the builtin speakers and the headphone output. The range of volume available should be implemented
as described in Section 508 guideline 1194.25(f), minimum output amplification of 65dB,
where ambient noise level is above 45dB (a volume gain of at least 20dB above the ambient
level shall be user selectable. Note that an automatic default should reset the volume after
every use.
Figure 21: A volume knob is provided to allow the user to adjust the output volume.
Applicable Guidelines:
Section 508 – 1194.25(f), 1194.31(a), 1194.31(b), 1194.31(c), 1194.31(d)
HFDS – 8.18.4.2
ICT – 302.5, 402.3, 402.3.1, 402.3.2, 408.2
VSS – 2.2.7.2(b)(8)
71
The rate at which speech output is provided is not adjustable.
Detailed Description: Users who are accustomed to using speech output often become
proficient at understanding speech output delivered at a very high rate. However, users with
recent visual disabilities, especially when coupled with cognitive impairments, may not be as
proficient with speech output systems, and may need speech played back at a slower rate.
Populations Impacted: Voters with cognitive impairments; voters who are hard of hearing;
voters who are blind; voters with low vision.
Potential Solutions: Provide adjustable rate of speech output within user settings of the voting
device. Augmentative and alternative communication is typically much slower than speech,
with users generally producing 8–10 words per minute (Higginbotham, Shane, Russell, &
Caves, 2007). Rate enhancement strategies can increase the user's rate of output to around 12–
15 words per minute, and as a result enhance the efficiency of communication.
Provide a control that allows the user to request that voice messages be repeated. This will
allow users to listen to messages again if they did not hear or understand the message initially.
Ideally, the repeat control should be a physical control (as opposed to an on-screen control) and
should be tactilely discernible to increase the accessibility of the control for users with visual
impairments.
Applicable Guidelines:
ADA-ABA – 707.5.1
Section 508 – 1194.25(e), 1194.31(a), 1194.31(b), 1194.31(c)
Section 255 – 1193.43(e)(2)(d)
HFDS – 8.18.4.2
VSS – 2.2.7.2(b)(5)
Additional Comments: It is important to note that, although the primary issue is that the rate
of speech output is not adjustable, there are additional issues that may arise, even if the rate is
adjustable. If not implemented properly, speech may become distorted at high or low speeds.
Also, one system GTRI observed allowed users to adjust the “rate,” which merely shifted the
pitch of the voice, without actually reducing or extending the amount of time required to listen
to the speech output. Research would need to be conducted to determine the best fit among
speech synthesis, digitized speech, and other variances amongst options as they relate to speech
output for voting devices.
72
Speech output via speakers may be inappropriate due to privacy concerns.
Detailed Description: Audio output may be required for some users to interact with voting
devices. The audio output may need to be at a relatively high volume to overcome ambient
noise or to accommodate users who are hard of hearing. However, much of the information
provided by voting devices should remain private, rather than being broadcast over speakers.
Populations Impacted: Voters who are blind; voters with low vision; voters who are hard of
hearing.
Potential Solutions: Provide a headphone jack so that users can connect personal headsets to
the device. The headphone jack should use a standard headphone connector (a 3.5 mm pin is
most common). External audio playback through speakers should be disabled when
headphones are connected. Ensure headphone jack is 1/8 inch stereo headphone jack to allow
individual voters to supply personal headsets. The use of a standard sized headphone jack
would allow voters who want to use their own headphones the ability to do so easily.
Avoid outputting potentially sensitive information through speakers. Due to privacy concerns,
information regarding a user’s votes should not be broadcast through speakers. Rather, it
should be delivered only through the headphone jack.
Figure 22: Potentially sensitive audio is output through speakers.
73
Figure 23: A headphone jack is provided for private listening.
Applicable Guidelines:
VVSG – 3.2.3.1-A.2, 3.2.3.1-A.3
VSS –2.2.7(b)(7)
ICT –302.1, 302.2, 302.5, 402.3.1, 406.1, 407.10, 408.4
ADA-ABA – 707.4, 707.5
Section 508 – 1194.25(e), 1194.31(d)
Section 255 – 1193.43(e)(2)(e)
HFDS – 8.18.4.2
74
Speech output is difficult to understand due to poor sound quality or interference.
Detailed Description: Voice output may be difficult for some users to understand because the
device’s speakers are not capable of reproducing the voice output clearly and without
distortion, particularly at higher volumes. Similarly, static and distortion can be included in
recorded audio, making it difficult to understand. Other sounds that accompany voice output,
such as background music, may also make it more difficult for users to understand the content
of voice messages.
Populations Impacted: Voters who are blind; voters with low vision; voters who are hard of
hearing; voters with cognitive impairments.
Potential Solutions: Ensure that the device’s speakers are capable of reproducing voice output
legibly through the full range of output volumes. The range of volume available should be
implemented as described in Section 508 guideline 1194.25(f). The speakers used in the device
should be tested to ensure that they are capable of reproducing voice output clearly and without
distortion through the entire range of volume adjustments.
Avoid background music or other sounds that may reduce the legibility of voice output.
Background music or other sound effects that are played back at the same time as voice output
messages may be distracting, and may reduce the legibility of the voice output.
Ensure that information conveyed by voice is accompanied by a redundant visual presentation.
Providing redundant visual information (for example, via on-screen text or graphics) may help
users who are hard of hearing or who routinely use a combination of visual and audio to obtain
information.
Applicable Guidelines:
ADA-ABA – 707.5
Section 508 – 1194.25(f), 1194.31(a), 1194.31(b), 1194.31(c), 1194.31(d)
Section 255 – 1193.43(e)(2)(a), 1193.43(e)(2)(e)
HFDS – 7.3.2.1, 7.3.2.2, 8.18.4.1, 8.18.4.2
75
Speech output is not repeated.
Detailed Description: Users may fail to hear or understand voice output messages when they
are first presented. If the messages do not repeat, either automatically or under user control,
then users may be unsure how to proceed with their interactions with the device.
Populations Impacted: Voters with cognitive impairments; voters who are hard of hearing;
voters who are blind; voters with low vision.
Potential Solutions: Repeat voice messages automatically if the user does not respond within
some period of time. If the user fails to respond or take action within a reasonable amount of
time (a few seconds) after a voice message is presented, repeat the voice message.
Provide a control that allows the user to request that voice messages be repeated. This will
allow users to listen to messages again if they did not hear or understand the message initially.
Ideally, the repeat control should be a physical control (as opposed to an on-screen control) and
should be tactilely discernible to increase the accessibility of the control for users with visual
impairments.
Applicable Guidelines:
ADA-ABA – 707.5.1
Section 508 – 1194.25(e), 1194.31(a), 1194.31(b), 1194.31(c)
Section 255 – 1193.43(e)(2)(d)
HFDS – 8.18.4.2
VSS – 2.2.7.2(b)(5)
ICT – 402.2.1
76
Speech output requires users to listen to large blocks of text without the ability to pause, start
over, or skip ahead.
Detailed Description: Users may have to listen to large blocks of text without the ability to
start over, pause, or skip text. If users want to listen to a specific section within the text, they
may have to listen to the text in its entirety a second time starting at the beginning, which can
be time consuming and can raise their level of stress and frustration. Moreover, repetition can
be useful, but it can become annoying and time consuming if used excessively. Unnecessary
repetition (for example, reading through the entire list of candidates for a contest, without the
ability to skip ahead, before allowing users to navigate through the list to select their desired
candidate) should be avoided.
Populations Impacted: Voters with cognitive impairments; voters who are hard of hearing;
voters who are blind; voters with low vision.
Potential solutions: Provide users the ability to “rewind” and “forward” the audio in small
increments, for example 10 seconds at a time. Allowing voters the ability to go backward or
forward gives them the ability to skim through sections they may have already heard or read
about and provides them the opportunity to hear a single section of text without having to hear
several paragraphs at a time. Ideally, the repeat control should be a physical control (as
opposed to an on-screen control) and should be tactilely discernible to increase the accessibility
of the control for users with visual impairments. This could reduce voter error if they need
only repeat a portion of the instructions. This solution would likely reduce the amount of time
to complete the voting process for those using the speech output technologies.
Applicable Guidelines:
VSS – 2.2.7.2(b)(5)
Section 508 – 1194.25(e)
ICT – 402.2, 402.2.1
77
Speech output is inaccurate.
Detailed Description: Computer generated speech output can be difficult to understand if the
computer generated voice does not mimic human speech well enough. Computer generated
speech may mispronounce names, contain unnecessary pauses, and/or speak with an odd
cadence, making the output difficult to parse and understand. Transitions between recorded
human speech and computer-generated speech can be jarring due to differences in voice,
volume and cadence.
Populations Impacted: Voters who are blind; voters with low vision; voters who are hard of
hearing; voters with cognitive impairments.
Potential Solution: Ensure that computer generated speech output passes a certain standard.
While newer computer generated speech programs have remarkably human-like speech output,
older systems may present a problem. If an older program is being used, the quality of the
speech output should be evaluated. There are multiple tests for evaluating different aspects of
speech quality, but the most common problems involve prosody (rhythm, stress, and intonation
of the speech). This aspect of synthetic speech can be evaluated with a test such as SAM
Prosodic Form Test (van Heuven & van Bezooijen, 1995).
Applicable Guidelines:
ICT – 402.2, 402.2.1
78
Non-verbal audio output is not meaningful.
Detailed Description: In addition to verbal output, some voting devices use simple non-verbal
audio output (i.e., beeps or tones) to communicate information to users. These tones may not
be meaningful in the absence of accompanying verbal information, such as an on-screen
message or graphic. Users with visual impairments may not have access to visual information,
and may therefore have difficulty interpreting the meaning of non-verbal audio output. Users
with cognitive impairments may also have difficulty understanding the meaning of non-verbal
audio output.
Populations Impacted: Voters who are blind; voters with low vision; voters with cognitive
impairments.
Potential solutions: When possible, select sounds with characteristics that convey meaning. It
is difficult to convey meaning through simple sounds in a way that is universally understood,
but it may be possible in some situations. For example, a “positive” sound (a “ding” or a rising
tone) could be used to indicate that input has been accepted, and a “negative” sound (a “buzz”
or a descending tone) could be used to indicate that input has been rejected.
Consider using non-verbal sounds only to convey very simple information. For example, a
“click” sound could be used to indicate that a key-press has been accepted. The information is
conveyed by the presence or absence of the sound, and not by the characteristics of the sound.
When using sounds in this way, ensure that the sounds are temporally matched with the
associated event, so that the relationship between the event (for example, a key-press) and the
sound is clear.
Use verbal messages to convey information. When detailed information needs to be conveyed
via sound, use verbal information to explicitly and unambiguously convey that information.
This eliminates the requirement for users to interpret the meaning of the sound, and also serves
as a redundant means of providing the information, which benefits users with visual
impairments.
Applicable Guidelines:
ADA-ABA – 707.5
Section 508 – 1194.31(a), 1194.31(b)
HFDS – 7.2.1.2, 7.2.1.4, 7.2.1.7, 7.2.1.8
79
Issues with Ballot Design and Presentation
This section addresses accessibility issues specific to presentation of information on an
absentee ballot. Navigation controls, presentation of instructions, and other graphical user interface
issues are described.
Instructions and Help
The system does not provide sufficient instructions.
Detailed Description: Before users begin voting with a device, the device should provide
instructions that give users with the basic information they need to operate the voting device
and to interact with the ballot. More detailed information (help) should be available on request
throughout the voting process as needed.
Population Impacted: All voters, especially those with cognitive impairments.
Potential Solutions: Instructions and help should be provided at appropriate times. Before
users begin voting with a device, the device should provide general instructions that give users
the basic information they need to operate the voting device and to interact with the ballot.
More detailed information (help) should be available on request throughout the voting process
as needed.
Ballot instructions should be clear and direct. The use of double negatives should be avoided
(“If the oval is not marked, your vote cannot be counted”) as well as messages that are unlikely
to help voters understand and detect errors. Voters should be warned of the consequences of an
action before they have an opportunity to take that action (Laskowski & Redish , 2006). Other
best practices include, but are not limited to the following: state consequences before users are
likely to act, state context (consequence) before action, avoid the use of technical language,
state instructions and logical order.
Applicable Guidelines:
VVSG – 3.2.4-E.4
HFDS – 8.1.3.5, 15.2.1.3
80
Timing of instructions is inappropriate.
Detailed Description: Information in instructions that is presented too early might be
forgotten. If all instructions are presented at the beginning of the ballot, then many voters are
likely to forget the instructions as they reach the end of the ballot. Users should be provided
with the basic information they need to use the device and navigate the ballot at the beginning
of the voting process, and additional instructions should be provided later in the context in
which they are needed.
Population impacted: All voters, especially those with cognitive impairments.
Potential Solutions: Separate steps in a sequence into individual instructions. Break
instructions into individual steps that users can read and execute before proceeding to the next
step. General instructions should be placed at the beginning of the ballot, and instructions that
are specific to individual sections should immediately precede their respective sections.
Present the steps in the order in which they must be performed, and use bullets or numbering to
distinguish individual steps and reinforce the order of steps. White space should be used to
separate steps.
Provide an advance organizer. An advance organizer is an overview of subsequent
information. Advance organizers have been shown to improve retention of information.
Present an overview of the ballot in the general instructions (for example, a table of contents
could include, “(1) Vote in partisan elections, (2) Vote in non-partisan elections, (3) Vote for
ballot initiatives, (4) Review and verify your votes, (5) Submit your ballot.”
Applicable Guidelines:
VVSG – 3.2.4-E.4
HFDS – 8.1.3.5, 15.2.1.3
81
Instructions may be difficult for users with cognitive impairments to understand.
Detailed Description: Instructions that are poorly written, overly complex, or that are written
at too high a reading level may lead to comprehension problems for users with cognitive
impairments. Instructions describing a series of sequential actions may also be confusing, if the
instructions are poorly structured.
Population Impacted: Voters with cognitive impairments.
Potential Solutions: Use simple language in instructions. Use the simplest language that
conveys the required information. Use familiar words, and use short sentences and paragraphs.
Use the imperative form of verbs (e.g., “Press the down arrow” instead of “The down arrow
should be pressed”). Use appropriate illustrations to reinforce the contents of the text.
Ensure that instructions are grammatically correct and free from typographical errors.
Instructions should be reviewed by a technical writer or a proficient editor who is familiar with
the ballot for which the instructions were written.
Provide structure to the steps by grouping them into sub-goals. A long sequence of actions
may appear to lack structure, making it difficult for users to understand. Provide structure by
using headings to group sets of related steps. Use of indentation can help delineate sets of substeps. However, do not rely solely on indentation, because it is inaccessible to blind voters.
Ensure that the sequence of actions is presented unambiguously and in the proper order,
especially if the instructions are presented via speech output. Instructions for actions should be
presented in the order in which they are to be executed (e.g., “Insert coins, then insert bills”
instead of “Before inserting bills, insert coins”).
Place CONDITION before ACTION. Goals or conditions for action should be presented before the
action, especially if the instructions are presented via speech output. For example, use the form
“To select the 3 PM showing, press 1” rather than “Press 1 to select the 3 PM showing.”
Test instructions with members of the user population to ensure that they are easy to
understand and to follow. User testing will help identify potential sources of errors or
confusion in instructions. Care should be taken to sample a representative portion of the
targeted user population, including users with disabilities.
Use unambiguous words and phrases. Ambiguous words and phrases will confuse voters. For
example, the instructions, “Continue voting next page” might be interpreted to mean (a)
continue to the next page for the next race or (b) continue to the next page to see more
candidates for this race.
Applicable Guidelines:
Section 255 – 1193.41(i)(2)(e), 1193.41(i)(2)(f)
HFDS – 2.6.3, 4.3.5.4.2, 4.3.5.4.3, 8.2.5.2.2, 15.5.1.1.1, 15.5.1.1.2, 15.6.2.3
ISO/IEC 71 – 8.7.2, 8.7.3, 8.7.4
VVSG – 3.2.4-C, 3.2.4-E.4
82
User Interface Design
Users should be able to freely navigate the ballot so that they can revisit and skip items as
desired. Users with disabilities need special affordances to allow navigation. The voting interface
should provide users with a clear indication of their current location within the ballot. The
considerations below apply to all forms of electronic ballots.
The system does not provide clear and consistent page titles.
Detailed Description: Each page of the user interface should have a title (typically the name of
the contest, but also including pages like Help or Ballot Review), and the title should be
presented in the same format and the same location on each page.
Population Impacted: All voters, especially those with cognitive impairments.
Potential Solution: Each page of the user interface should have a title. Typically the title
should display the name of the contests appearing on that page, or “Help” for help pages and
“Ballot Review” for the review pages. The titles should be presented in the same format and
the same location on each page.
83
Display elements do not receive focus in the correct reading order.
Detailed Description: Display elements, such as navigation controls and the list of candidates
on a ballot screen, should receive keyboard focus in the appropriate reading order. Most screen
readers will adhere to the same focus order rules. All necessary content should appear in the
tab order. The tab order should wrap at the bottom of the screen back to the top, and vice
versa.
Population Impacted: Voters who are blind; voters with low vision.
Potential Solution: Ensure that keyboard controls and screen readers advance the cursor and
read information in the correct order.
Applicable Guidelines:
ANSI/HFES 200 – 8.5.10
84
The system does not provide clear and consistent navigation controls.
Detailed Description: Navigation controls (for example, “Previous Contest,” “Next Contest,”
“Help,” etc.) should be provided in a consistent location on every page and should be easy to
locate, identify, and operate.
Population Impacted: All voters, especially those with cognitive impairments.
Potential Solution: Place common controls consistently throughout the user interface. If there
are controls that appear on multiple screens, such as navigation controls (e.g., Help, Back,
Next), ensure that the placement of those controls is the same on every screen.
85
The mapping of hardware controls to user interface functions is inconsistent.
Detailed Description: Hardware controls (e.g., physical buttons) are sometimes linked to
display elements on different screens. If these display elements represent different categories
of actions (e.g., “submit” vs. “select”), users are likely to make mistakes.
Population Impacted: All voters, especially those with cognitive impairments.
Potential Solution: A consistent control scheme should be used throughout the user interface.
For example, the same hardware control should always be used to activate software controls,
rather than using different controls to activate different types of items (e.g., buttons versus
checkboxes).
Applicable Guidelines:
HFDS – 6.3.2.11, 6.4.1.21, 6.4.1.22, 6.4.1.30
86
When scrolling is required (and it should be used sparingly and consistently), the user is not
notified that scrolling is needed.
Detailed Description: When there is too much information to fit on a single screen (for
example, if the list of candidates is very long), a clear indication that scrolling is available
should be provided to users.
Population Impacted: Voters with cognitive impairments.
Potential Solutions: A clear indication that scrolling is available should be provided to users.
This can be accomplished by providing obvious scrolling controls as part of the visual
interface, (e.g., providing a scroll bar) and by announcing the total number of candidates
available in the audio interface. When a list of candidates is long enough to require scrolling,
the presentation order of candidates should be randomized to offset presentation order effects.
Provide multiple columns to list candidates, so that all information can be seen on one page.
This would eliminate the need for page scrolling. If voter needs to enlarge font, scrolling may
still be required, but would be greatly reduced.
Applicable Guidelines:
VVSG – 3.2.6-A
87
Onscreen controls that are small and close together may be difficult to activate without
activating adjacent controls.
Detailed Description: If onscreen controls are small or are placed too close together, they may
be difficult for users without fine motor control to activate without accidentally activating
adjacent controls. Because no tactile feedback is provided by touchscreen controls, if
redundant visual and auditory feedback is not provided when controls are activated, these
accidental activations may go unnoticed.
Population Impacted: Voters with upper mobility impairments.
Potential Solutions: Redundant visual and auditory feedback should be provided. If the user
makes a mistake, the redundant feedback will alert the user to the mistake.
Tactically discernible controls should be provided for each function. Key surfaces should be
raised above surrounding surfaces or otherwise tactically discernible.
On touchscreens, controls should have an appropriate size and minimum distance from other
controls. Controls should have a minimum height of 0.5 inches and minimum width of 0.7
inches. The vertical distance between the centers of adjacent areas should be at least 0.6
inches, and the horizontal distance at least 0.8 inches.
Applicable Guidelines:
VSS – 2.2.7.2(g)
VVSG – 3.2.6-C.1
ICT – 407.3
88
Complex or inconsistent user interface screens may be difficult for users to understand.
Detailed Description: User interfaces that are complex (for example, displays that contain
many controls associated with multiple tasks) can be difficult for users with cognitive
impairments to navigate. Inconsistencies in displays, such as changes in control placement
from screen to screen or inconsistent use of terminology, can be confusing to users with
cognitive impairments. Similarly, inconsistent or excessive use of abbreviations can decrease
reading comprehension for users with cognitive impairments. Maintaining consistency and
keeping the interface as simple as possible are important usability considerations, and will
improve the accessibility of the machine to all users.
Population Impacted: All voters, especially those with cognitive impairments.
Potential Solutions: Reduce the complexity of user interface screens where possible. Design
screens around individual user tasks (for example, a voting interface might have one screen
dedicated to each contest). Avoid complex displays that contain a large number of options and
controls.
Place common controls consistently throughout the user interface. If there are controls that
appear on multiple screens, such as navigation controls, ensure that the placement of those
controls is the same on every screen.
Use consistent terminology throughout the user interface. Ensure that names and abbreviations
are applied consistently throughout the user interface.
Limit the use of abbreviations. Abbreviations (especially those that may be unfamiliar to users)
should be used sparingly in the user interface.
Applicable Guidelines:
Section 255 – 1193.41(i)(2)(a)
HFDS – 2.3.1, 4.3.5.4.1, 8.2.5.4.4, 8.2.5.4.8, 8.2.11.1.2, 8.14.1.10
VVSG – 3.2.4-C
89
System time-outs may cause problems for some users.
Detailed Description: The user interface for voting devices may include system time-outs,
which are situations where the user interface automatically changes states or resets if user input
is not received within a certain time period. People with disabilities often require more time to
respond than non-disabled users, so system time-outs can disrupt their transactions if they are
not notified that a time-out has occurred and allowed to request more time.
Populations Impacted: Voters with cognitive impairments; voters with upper mobility
impairments; voters who are blind; voters with low vision.
Potential Solution: Alert users when a time-out occurs, and allow them to request more time.
When a system time-out occurs while the user is performing a task, the user should be alerted
that the time-out has occurred, and given the option to request more time or cancel the task.
The user should be given sufficient time to respond to the alert before it expires. A good rule
of thumb for what constitutes sufficient time is 10 times the amount of time it would take an
average user respond.
Applicable Guidelines:
Section 508 – 1194.25(b)
Section 255 – 1193.41(g), 1193.41(i)(2)(f)
HFDS – 8.18.2.2
VSS – 2.2.7.2(g)
90
The voting device changes the ballot automatically based on assumptions about user intent.
Detailed Description: Voting devices may attempt to streamline the voting process by
providing default selections or by attempting to resolve errors automatically. Although these
actions may sometimes be appropriate, care must be taken to ensure that final control over
voting selections remains with the user. For some electronic voting devices, the first option
listed for a particular office is automatically selected by default. This may be confusing for
users with cognitive disabilities, because it may appear that they have already made a selection.
They may proceed to the next screen without realizing that they voted for a candidate.
Some voting devices automatically deselect options when voters attempt to make more than the
maximum number of selections for a contest. This could be appropriate when there are
mutually exclusive options for a contest, but should not occur when users can select more than
one option for a particular contest. In those situations, users should be notified that they have
exceeded the allowable number of selections, so that they can deselect candidates as desired
before making additional selections.
Population Impacted: Voters with cognitive impairments.
Potential Solutions: Options should not be pre-selected. None of the options on the ballot
should be selected without the user actively selecting those items. If the user does not actively
select one of the items, an undervote should occur.
Users should be notified that they have exceeded the allowable number of selections. When
more than one selection is allowable, users should be notified if they have exceeded the
allowable number of selections so that they can deselect candidates as desired before making
additional selections.
Applicable Guidelines:
HFDS – 8.2.11.7, 8.12.1.13, 8.12.1.14
91
The voting device does not provide sufficient feedback to the user.
Detailed Description: Users with different disabilities have different needs for the amount and
type of feedback provided by a voting device. The feedback that is provided may be
insufficient because it does not cover the full range of events for which feedback is required, or
it may be insufficient because it is provided in a format that is not useful to a user with a
particular disability (for example, visual feedback indicating that input has been accepted is not
useful to a user who is blind). Providing feedback for a wide range of events and user actions
in a variety of sensory modalities is beneficial for all users.
Populations Impacted: Voters who are blind; voters with low vision; voters who are deaf;
voters who are hard of hearing; voters with upper mobility impairments.
Potential Solutions: Ensure that feedback is provided for all relevant events. Provide
feedback for all user inputs, system status changes, user or system errors, and other events that
are relevant to the user’s interaction with the machine. Feedback for different events should be
distinct from one another and appropriate to the events represented. For example, a simple
click may be sufficient to acknowledge a keypress, but a more prominent tone may be
necessary to indicate that an error has occurred.
Provide feedback in a visual format. Visual feedback is necessary for users with hearing
impairments, but it can also be helpful for users with low vision (if the feedback is sufficiently
large or if it also makes use of color or other visual cues), and for users with upper mobility
impairments (to help the user determine when unintentional inputs have been made).
Provide feedback in an auditory format. Auditory feedback is necessary for users who are
blind, and it can also be helpful for users with low vision and for users with upper mobility
impairments. Beeps and other sounds help users know that input was accepted, and also serve
to alert users if an unintentional input was made (e.g., a button was pressed twice instead of
once). Voice output of more complex data (such as the name of a candidate that was selected,
or indications of required user actions) helps users with visual impairments verify transactions
and allows them to operate voting devices more effectively.
Applicable Guidelines:
ADA-ABA – 707.5
Section 508 – 1194.31(a), 1194.31(b), 1194.31(c), 1194.31(d)
Section 255 – 1193.43(a)(2)(a)
HFDS – 2.6.1, 5.11.1, 5.11.2, 8.15.8.3, 8.18.3.2, 8.18.4.1
VSS – 2.2.5.2.2, 2.2.7.2(h)
92
The system does not provide clear and consistent status information.
Detailed Description: Status information (i.e., voting progress) does not give the user
knowledge about his or her progress through the ballot.
Populations Impacted: All voters, especially those with cognitive impairments.
Potential Solution: The status information needed to interact with the ballot should be
provided in a simple and consistent format. Status should be available both visually and nonvisually. This includes the total number of contests and the current contest, the number of
selections allowed for the current contest and the number of selections already made, the
number of options available for the current contest, and so forth.
93
Information Presentation
The formatting and visual layout of the ballot can greatly affect users ability to comprehend the
ballot and cast their votes as they intended. Formatting and visual layout refers to design aspects such
as indentation, alignment, column breaks, and page breaks. Poor ballot formatting can cause problems
for all users (for example, consider the butterfly ballot used in the Florida Presidential election of
2000), and can be especially problematic for users with cognitive impairments.
Icons used in place of text labels may be ambiguous.
Detailed Description: Graphical icons may be used to identify controls in place of text labels
for various reasons: to produce a device that is language-independent, to reduce the amount of
space required to label controls and components, or to achieve a desired aesthetic effect in the
design of the device. However, understanding icons, especially if the icons are very abstract
and/or are unfamiliar to the user, can be difficult, especially for users with cognitive
impairments.
Population Impacted: All voters with vision, especially those with cognitive impairments.
Potential Solutions: Limit the use of icons. Consider the use of icons carefully, and avoid
using icons instead of text labels for the sake of aesthetics or other considerations to the
detriment of usability. Consider supplementing icons with text labels for clarity, but avoid
crowding.
Use standard or familiar icons. Icons representing certain functions or controls (such as stop,
play, headphone output, volume, etc.) are widely used and are likely to be familiar to most
users. Avoid deviating too far from the common appearance of those types of icons. Other
icons (such as a depiction of a card next to a smart card slot) may be clear in the context in
which they appear. The use of new icons to represent abstract concepts should be avoided.
When new icons must be developed, test the icons with members of the user population to
ensure that the meaning of the icons is clear. User testing may provide insight into how to
design clearer, more meaningful icons. Care should be taken to sample a representative portion
of the targeted user population, including users with disabilities.
Applicable Guidelines:
HFDS – 8.13.3.1, 8.13.3.9, 8.13.3.10
VVSG – 3.2.4-C.3
94
Small text and icons are difficult for users with low vision to perceive.
Detailed Description: Text and icons on GUI displays may be difficult for some users to read
if they are too small.
Population Impacted: Voters who are blind; voters with low vision.
Potential Solutions: Ensure that the font size used for text is sufficiently large. Characters
should be at least 3 millimeters. Where possible, allow font size to be magnified or adjusted to
a larger size.
Ensure that icons are large enough for low vision users to see. The largest dimension (height
or width) of icons should be at least as large as the character height calculated above. Icons
should be made as large as possible, given the space available.
Provide contrast adjustment for the display. Although users with low vision prefer and
generally require larger fonts, they may be able to read smaller fonts if the contrast is
sufficiently high. Provide a range of contrast settings for the user to adjust through a hardware
control. (Software controls are problematic, because if the contrast is insufficient for the user,
the user may not be able to read the display in order to find the contrast adjustment controls.)
Provide an alternate display mode with larger fonts and high contrast options. A userselectable alternate display mode that uses larger fonts and provides high contrast options, even
if it contains only the most important information and controls, will be useful for users with
low vision.
Provide alternatives to the visual display to facilitate interaction by users with low vision. A
voice display should be integrated into the machine, so that visual content is presented in an
auditory fashion as well. For example, using a set of hardware controls, the user could navigate
through configuration menus that are voiced, without having to read the menus on the display.
Applicable Guidelines:
VSS – 2.2.7.2(b)
ADA-ABA – 707.5
Section 508 – 1194.31(b)
Section 255 – 1193.43(b)
HFDS – 5.11.1, 5.11.7, 8.2.5.6.5, 8.2.5.6.6, 8.2.5.6.9, 8.18.3.1
95
Non-text elements are not accessible to some users with visual impairments.
Detailed Description: Non-text elements include pictures, icons, arrows, and other graphical
elements. The information presented by these elements is not accessible to users who are blind
or have low vision.
Populations Impacted: Voters who are blind; voters with low vision.
Potential Solution: Provide image tags that can be read by assistive technologies. The image
tags should describe the purpose of the image or the information conveyed in the image,
including operating instructions, visual prompts, error messages, and user input verification.
The audio information should provide sufficient instructions to allow vision impaired users the
same capabilities as other users.
Applicable Guidelines:
VSS – 2.2.7.2(b)(1-3)
ICT – 402.2
96
Control names are not descriptive.
Detailed Description: The user interface may be designed primarily for sighted users, with
visual context working in conjunction with control labels to provide the appropriate meaning.
However, visual context is not available for users with visual impairments, and users who have
recently become visually impaired may lack experience interpreting interfaces without visual
cues. Control labels should be descriptive enough that their purpose is evident even without
visual cues.
Population Impacted: Voters who are blind; voters with low vision.
Potential Solution: Control labels should be descriptive enough that their purpose is evident
even without visual cues. For example, a control should be visually and auditorily labeled
“Next Contest” rather than simply “Next”.
Applicable Guidelines:
VSS – 2.2.7.2(b)(1-3)
VVSG – 3.2.4-G
97
Candidates are listed on multiple columns or pages.
Detailed Description: Ballot designers sometimes split candidates for a given race across two
columns on a page or across multiple pages. Voters may assume that the two groups (e.g., first
and second columns) represent candidates for separate races. This results in a high rate of over
voting.
Population Impacted: Voters with cognitive impairments.
Potential Solutions: Display all candidates for single race in a single column. Space
constraints might necessitate some changes in format. For example, consider widening the
columns and placing more text on each line.
Include salient and clear instructions on both columns/pages. If the number of candidates
precludes displaying the candidates in a single column, then include appropriate instructions.
For example, instructions and at the bottom of the first column could state, “See next column
for more candidates,” and the second column could begin with the words, “candidates
continued.” Do not restate the name of the race at the top of the second column, because it
might be misinterpreted as the heading for another race.
Applicable Guidelines:
VVSG – VVSG 3.1.1(2), 3.2.6-A
VSS – 2.3.1.1(f)
HFDS – 8.1.1.5
98
Multiple races are displayed on one page of an electronic ballot.
Detailed Description: Some electronic ballots follow the design of traditional paper ballots,
and present multiple races on a single large electronic “page.” This design can make it hard for
users to navigate the ballot, and may induce users to skip contests because they don’t realize
that multiple contests are presented on a single page. Presenting a single contest on each page
allows for simplicity and consistency of navigation.
Population Impacted: Voters with cognitive impairments.
Potential Solutions: List only one race on each page of an electronic ballot. Users will be
very unlikely to skip a race if it is the only race displayed on a page. This applies to electronic
ballots only, because the cost of displaying one race per page on paper ballots would be
prohibitive.
Use visual cues such as bold text and shading to direct users’ attention. If multiple races are
listed on one page, salient visual cues can direct users’ attention appropriately. For example, a
horizontal line could separate the races, and the title of each race could be bolded and
highlighted.
Applicable Guidelines:
VVSG – 3.2.6-A
HFDS – 8.1.1.3
99
Write-In Voting
Voting for a write-in candidate is considerably more complicated than voting for a candidate
that is listed on the ballot. It requires additional physical and cognitive activities that may be
challenging for many voters.
Use of the write-in option is confusing.
Detailed Description: Many ballots require two actions for casting a write-in vote: filling in a
circle or punching a hole next to the “write-in” option, and writing the name of the candidate in
the blank. Many state laws dictate that the ballot is spoiled if both actions are not performed.
A typical voting error is the failure to mark the bubble/hole next to the “write-in” option. It is
likely that many voters do not read the instructions or do not understand the instructions, and
fail to see the option before they write-in a candidate’s name.
Population Impacted: Voters with cognitive impairments; voters with upper mobility
impairments.
Potential solution: Place bubble with instructions immediately adjacent to write-in blank. The
bubble should be followed immediately by brief and clear instructions, such as “Mark this
bubble for write-in.” Use visual cues to group the write in bubble, instructions, and write-in
blank.
If using an electronic voting system, allow write-in via touchscreen. If voter chooses to writein, provide a pop-up keyboard to allow insertion of candidate’s name.
Applicable Guidelines:
HFDS – 6.3.2.4, 6.3.2.7
VVSG – 3.1.1(2)
100
Typing the name of a write-in candidate is difficult.
Detailed Description: Although touchscreens can make it easy to select candidates, the act of
typing a candidate name can be difficult. Typing on touchscreens is typically accomplished via
an onscreen qwerty keyboard. The size and proximity of the keys can make them difficult to
use accurately, particularly for users with tremors and poor motor control.
Populations Impacted: Voters with upper mobility impairments; voters with low vision.
Potential solutions: Provide a speech recognition interface for entering write-in candidates.
The voting device should be equipped with a microphone and speech recognition software.
Make the onscreen keyboard as large as possible while providing dead space between the keys
to reduce accidental activations.
Enable predictive typing. The system could use a predictive algorithm to offer a list of
completed, common names from which the user could select. For example, after the user types
in the letters “Dre” the system could offer the options, “Drew, Drews, Drexel.” This solution
has the potential drawback of giving some write-in candidates an unfair advantage, and legal
challenges might arise.
Applicable Guidelines:
VVSG – 3.2.3.1
VSS – 3.2.4.1
Section 508 – 1194.23(k)(1)
ICT– 302.7
101
Ballot Review
Many electronic ballots provide voters with no review screen at the end of the ballot. The primary
purpose of the review screen is to assist voters in detecting errors, which include unintentional
selections, overvotes, and undervotes.
Users are not given the opportunity to review their options before submitting their vote.
Detailed Description: Research suggests that voting errors are quite common in the general
population, and they may be even more common in populations with cognitive impairments. If
voters are not given the opportunity to review their choices, many mistakes may go
uncorrected.
Populations Impacted: Voters with cognitive impairments.
Potential solution: Provide a review screen at the end of the ballot. The screen should list all
the races on the ballot paired with each selection made by the voter. Providing additional
information about undervotes (and overvotes) may inform voters that they either forgot to place
a vote and encourage them to correct the error. Occasionally these are not errors, but instead
are intentional choices to abstain from voting in a particular race. Therefore, voters should not
be forced to “correct the mistake”.
Applicable Guidelines:
VVSG – 3.1.1(1)
VSS – 2.4.3.3(h), 3.2(a)(1)(A)(i-ii)
VVSG – 3.2.2-A
102
Users do not notice when the review screen differs from their intended selections.
Detailed Description: users might make unintentional selections of candidates. Then they
may fail to notice their error when viewing their selections on the review screen. The potential
solutions to this problem rely on the voter remembering who he or she intended to vote for, or
at least the party affiliation.
Populations Impacted: Voters with cognitive impairments.
Potential Solutions: Use color and icon codes for party affiliations. Each political party could
be coded by a color and icon. The color and icon would appear with each selection on the
review screen. Clear instructions and a color/icon legend should appear at the top of the review
screen.
Use instructions to stress the importance of review. With a brief statement at the top of the
screen, urge voters to carefully review their selections. For example, “Voting errors are
common. Carefully review your selections below.”
Place hyperlinks on the review screen. Hyperlinks to review data would make it easier for
voters to see a summary of their voting actions. Voters should be able to click on each race
shown on the review screen. The system should make it easy for users to navigate directly to a
specific contest from the review screen, make desired changes, and then return to the review
screen after changes have been made.
Applicable Guidelines:
VVSG – 3.1.1(1), 3.1.1(2), 3.2(a)(1)(A)(i)
VSS – 2.4.3.3(h)
103
The system does not provide adequate notification of or protection against overvoting and
undervoting.
Detailed Description: Some electronic ballots simply allow users to mark the ballot but do not
provide any error detection capabilities. Electronic ballots should, at a minimum, notify users
when undervotes or overvotes have occurred, and allow them to correct these situations if
desired. Research shows that a substantial portion of voters do not notice undervotes and
overvotes on review screens (Campbell & Byrne, 2009). Occasionally these are not errors, but
instead are intentional choices to abstain from voting in a particular race. Therefore, voters
should not be forced to “correct the mistake”.
Population Impacted: All voters, especially those with cognitive impairments.
Potential Solution: Highlight undervotes and overvotes. Use a salient visual cue such as
highlighting to draw attention to undervotes and overvotes on the review screen. This will give
the voter an opportunity to go back and adjust, if they so choose. Undervotes/overvotes may be
the result of an inadvertent action and thus should be corrected in order for the vote to count as
expected.
Applicable Guidelines:
HFDS – 8.15.8.18, 8.15.11.2.1, 8.15.8.3
VVSG – 3.2.2-A, 3.1.1(2), 3.2(a)(1)(A)(i)
104
Revising a ballot is excessively complex or difficult.
Detailed Description: Some electronic voting systems make it unnecessarily difficult for users
to revise their ballots if they find mistakes on the review screen. For example, a system could
force users to start over and vote the entire ballot again in order to correct a mistake, or a
system could require users to back up through multiple pages to reach the page of the contest to
be corrected. The system should make it easy for users to navigate directly to a specific contest
from the review screen, make desired changes, and then return to the review screen after
changes have been made.
Population Impacted: All voters, especially those with cognitive impairments.
Potential Solutions: Place hyperlinks on the review screen. Hyperlinks to review data would
make it easier for voters to see a summary of their voting actions. Voters should be able to
click on each race shown on the review screen. The system should make it easy for users to
navigate directly to a specific contest from the review screen, make desired changes, and then
return to the review screen after changes have been made.
Voters should be warned of the consequences of an action before they have an opportunity to
take that action. Ballot instructions should be direct and concise to avoid errors, but should
assist the voter in correcting errors.
Applicable Guidelines:
ANSI/HFES 200 – 8.4.12
VSS – 2.4.3.3(h), 2.4.3.3(i)
VVSG 3.1.1(2), 3.2(a)(1)(A)(i-ii), 3.2.2.1-E
105
Policy-Related Issues
Voting issues for the target population that are related to policy (and are not specifically
accessibility related) are discussed below.
Obtaining accurate information on voter registration is difficult as service members transition
out of active duty status.
Detailed Description: Many states maintain special projects or best practices that were
implemented either for UOCAVA voters or for disabled civilian voters in general, without a
special emphasis on the target population (transitioning to veteran status). Because states rarely
provide additional services for this population, new veterans that are injured and possibly living
in a rehab facility or other veteran hospital facility may not know where to find election and
voter information. Only Indiana reported a best practice specifically for this population (MHI,
2012).
Populations Impacted: All voters transitioning out of active duty status.
Potential Solution: Voting assistance could be provided more extensively to Veterans
Administration facilities. Mandate that state offices participate in Warrior Transition Unit
meetings to ensure that injured service personnel preparing to transition from military service
to civilian life have current information about voting. The state offices could work directly
with Department of Veterans Affairs (VA) facilities, or this could be initiated at the local level.
106
Registering to vote is different from state to state.
Detailed Description: Registration request forms and other requests for information may vary
widely across states. Registered voters must provide a current valid mailing address to which
the ballot may be sent. Disabled active duty military personnel who are currently residing
away from their legal residence may have a Voting Assistance Officer (VAO) from the Federal
Voting Assistance Program (FVAP) or other assisting liaison. However, those who have
separated or retired from the military must have other assistance provided as available if no
family member or caregiver is present. The blank ballot may be sent by email, postal mail, or
picked up by a representative of the voter designated in writing. Obtaining the necessary
information may require assistance for those with certain disabilities and impairments.
Populations Impacted: All voters, especially those transitioning out of active duty status.
Potential Solutions: Standardize the registration request and allow absentee voters to register
permanently for absentee ballots. Some states already allow permanent applications, but only a
small percentage. The form, if standard across states, would not have to be filled out
completely every time a voter moves. Only the address information would need to change.
This would result in a reduction in voter errors, absentee ballots sent to incorrect addresses, and
less burden on the voter.
Provide voter assistance to recently transitioned veterans. Active duty service members are
provided with Voting Assistance Officers. Once they transition to veteran status, assign a
voting assistance officer (possibly civilian) to “transition” them into absentee voting as a
civilian.
Applicable Guidelines:
 HFDS 2003 – 2.4.1, 2.4.2, 2.4.5
 HAVA 2002 – Title III(a)(3)(C)
o Note: This guideline is applicable to the voting system, which should include the
process of registering to vote. Currently, no guideline is in place for the
population of transitioning service members to veteran status.
107
References
1. Brook, J. (1996). SUS - A quick and dirty usability scale. In P. W. Jordan, B. Thomas, B. A.
Weerdmeester & A. L. McClelland (Eds.), Usability Evaluation in Industry. London: Taylor
and Francis.
2. Campbell, B. A., & Byrne, M. D. (2009). Now do voters notice review screen anomalies? A
look at voting system usability. Paper presented at the Electronic Voting Technology
Workshop / Workshop on Trustworthy Elections, Montreal, Canada.
3. Drury, C. G. (1983). Task analysis methods in industry. Applied Ergonomics 14(1): 19-28.
4. Drury, C. G., Paramore, B., Van Cott, H. P., Grey, S. M., & Corlett, E. M. (1987). Task
analysis. In G. Salvendy (Ed) Handbook of Human Factors, Chapter 3.4. New York: J. Wiley
and Sons, 370-401.
5. Dumas, J. S., & Redish, J. C. (1994). A Practical Guide to Usability Testing: Revised Edition.
Intellect Books, Portland, Oregon.
6. Fain, W. B., & Folds, D. J. (2002). An approach to measuring accessibility of office automation
equipment. Proceedings of the Human Factors and Ergonomics Society Annual Meeting
46(24): 1982-1986.
7. Greene, K. K., Byrne, M. D., & Everett, S. P. (2006). A Comparison of Usability Between
Voting Methods. Paper presented at the USENIX/ACCURATE Electronic Voting Technology
Workshop, Vancouver, Canada.
8. Higginbotham, D. J., Shane, H., Russell, S., & Caves, K. (2007). "Access to AAC: Present,
past, and future". Augmentative and Alternative Communication 23 (3): 243–257.
DOI:10.1080/07434610701571058. PMID 17701743.
9. Laskowski, S. J., Autry, M., Cugini, J., Killam, W., & Yen, J. (2004). Improving the usability
and accessibility of voting systems and products. NIST Special Publication 500-256.
10. Laskowski, S. L., & Redish, J. (2006). Making ballot language understandable to voters. Paper
presented at the 2006 USENIX/ACCURATE Electronic Voting Technology Workshop,
Vancouver, Canada.
11. Nielsen, J. (1994b). Heuristic evaluation. In Nielsen, J., & Mack, R. L. (Eds.), Usability
Inspection Methods, John Wiley & Sons, New York, NY.
12. National Institute of Standards and Technology (2007). Usability Performance Benchmarks for
the Voluntary Voting System Guidelines.
13. Norden, L., Kimball, D. C., Quesenbery, W., & Chen, M. (2008). Ballot Design Checklist.
Retrieved June 7, 2012 from
http://www.brennancenter.org/index.php/ballots/pages/design_recommendations/
14. Rubin, J. (1994). Handbook of Usability Testing: How to Plan, Design, and Conduct Effective
Tests. John Wiley & Sons, New York, NY.
15. Sanders, M. S. and McCormick, E. J. (1993). Human Factors in Engineering and Design (7th
Edition). New York, McGraw-Hill
16. United States Department of Defense (2012). U.S. Casualty Status for OIF, OND, and OEF.
Retrieved July 10, 2012, from http://www.defense.gov/news/casualty.pdf.
17. United States Department of Justice (2004). ADA Checklist for Polling Places. Retrieved June
7, 2012, from http://www.ada.gov/votingchecklist.htm
18. United States Federal Election Commission (2002). Voting System Standards, Volume I Performance Standards.
19. United States Election Assistance Commission (2005). Voluntary Voting System Guidelines,
Version 1.0.
108
20. United States Election Assistance Commission (2009). Draft Voluntary Voting System
Guidelines, Version 1.1.
21. United States Election Assistance Commission (2011). A Survey of Internet Voting. Testing and
Certification Technical Paper #2. Retrieved July 23, 2012 from
http://www.eac.gov/assets/1/Documents/SIV-FINAL.pdf
22. van Heuven, V. J., & van Bezooijen, R. (1995). Quality Evaluation of Synthesized Speech. In
Kleijn, W. B., & Oaliwal, K. K. (Eds.), Speech Coding and Synthesis, Elsevier Science,
Amsterdam.
23. Wisconsin, D. R. (2011). Election day accessibility checklist. Retrieved May 25, 2011, from
http://gab.wi.gov/sites/default/files/publication/65/election_day_accessibility_checklist_03_201
1_pdf_16274.pdf
109