Association of Assistive Technology Act Programs (ATAP)

HAVA - Comparision of Accessibility in Currently Available Equipment

Legend: (+) conforms (0) able to conform, may/may not depending on set up (-) does not conform TABLE 1. Required FEC 2002 Access Standards (2.2.7.2)	AutoMark	InkaVote	Populex	Sequoia	Diebold	AccuPoll	Hart	ES&S iVote	Unilect 5
(b.1-3) Audio output communicates complete content of ballot and provides instruction to voter in operation of voting machine and voter has same vote capabilities as non-audio voters.	+	+	+	+	+	+	+	+	+
(b.4-5) Audio output enables voter to review write-in input, edit that input, and confirm that the edits meet voter intent and allows voter to request repetition of any information provided by system.	+	+	+ 2	+	+	+	+	+	+
(b.6-7) Audio output delivered by headphones and provides audio signal through industry standard connector of 1/8 inch jack	+	+	+	+	+	+	+	+	+
(b.8) Audio output has volume control with user adjustable amplification up to a max of 105 dB SPL with auto reset to default after each voter.	+	+	+	+	+	+	- 4	0 4	-
(e.1) Electronic image display provides a high contrast display defined as contrast ratio of 6:1 or greater	+	-	+	+	+	+	+	+	+
(e.2) Electronic image display provides either a black and white only display that communicates all the content information of a color display or a color display that provides the voter with a means of adjusting the colors.	+	-	+	01	+	+	01	+	+
(e.3) Electronic image display provides at least two font sizes using a sans-serif or similar font, one with capitol letters of at least 6.3 millimeters and one with capital letters of between 3.0 and 4.0 millimeters.	01	-	+	0 1	+	+	0 1	0 1	-
(f) Touchscreen input has mechanically operated alternative input that is tactilely discernible, can be operated with one hand, requires less than 5lbs of force to operate, and has no key repeat function.	+	+	+	+	+	+	+	+	+

Notes : The first three systems are ballot-marking devices. In these systems, the one and only official vote is the paper ballot, which a voter interacts with electronically to provide accessibility. These systems were rated for accessibility of the electronic interface only. Ballot handling is rated in Table 2. The remaining systems are DRE’s in which at least one vote record is electronic with an option for a voter verified paper audit trail (VVPAT). The DRE’s were rated for conformance to the access standards for the electronic vote process only. VVPAT accessibility is rated separately in Table 2 using VVSG standards.

1) These systems were rated 0 for visual screen output requirements (e.g. two required font sizes or monochrome/color adjustment) because the system could deliver those outputs but depending on ballot layout the system is also capable of NOT delivering the required outputs.

2) This system used manual (paper/pencil) write in which delivered no accessibility for people with vision and motor disabilities. As of August 2005, the system has been upgraded to provide electronic write in capacity. It is unclear how the original version was ITA certified to the 2002 access standards.

3) This system did not provide any automatic default for the volume control audio ballot output. It is unclear how this system obtained ITA certification.

4) This system uses an in-line, add-on device to provide the auto reset of volume output to default that depends on poll workers to deploy properly.

5) This system was not ITA certified to the 2002 access standards at the time of review.

General Note about FEC 2002 Access Standard 2.2.7.1

It was assumed that the ITA would verifying conformance with the reach range standards found in 2.2.7.1 a through f. However, for those systems that had completed ITA certification, it was not clear if any ITA had evaluated for conformance to those standards. In the reports provided, those standards were marked “NA” indicating they had not be tested. No attempt was made during our review to evaluate for conformance to the reach ranges. If the ITA’s did not evaluate for conformance to these standards, it is unknown whether or not any of the systems conform.

General Note about FEC 2002 Access Standards 2.2.7.2

The 2.2.7.2 access standards not listed in the first table were either not applicable to any of the systems reviewed or otherwise were not discriminating between systems. As a result they were not included in the table to keep the data focused on those standards that did show differences between systems.

Legend: (+) provides this feature (0) can provide this feature via option or in ballot set up (-) does not provide this feature TABLE 2. VVSG standards Italics = should bold = shall * = other desirable features	AutoMark	InkaVote	Populex	Sequoia	Diebold	AccuPoll	Hart	ES&S iVote	UniLect
a) Uses digitized (human) speech instead of synthesized (computer).	0	+	-	+	+	+	+	+	+
b) Simultaneous audio output & visual display (large print) is available.	+	-	+	-	+	+	+	-	0
c) Speed of speech (audio) output is voter adjustable.	+	-	0	-	+	-	-	-	-
d) Tactile controls have distinguishing shape/color to support ease of use.	+	+	-	+	-	+	+	+	+
e) Switch input (usually dual) option is available.	+	-	-	0	-	+	+	-	-
f) VVPAT reviewable by voters with vision disabilities (audio and large print)	NA	NA	NA	-	-	- 2	-	-	-
g) Paper ballots can be submitted independently by individuals with motor disabilities.	-	-	-	NA	NA	-	NA	NA	NA
*h) Size of screen adequate for large font display and ease of use.	+	-	-	+	+	+	+	+	+
*i) Simultaneous touch-screen and tactile control input is available.	+	-	+	0	+	+	-	-	-
*j) Size and weight are portable to allow for curbside voting.	-	-	+	0 1	0 1	-	0 1	0 1	0 1

Notes : The above features (a) through (g) are in the current version of the VVSG and (h) through (j) are other desired features identified during system review. The first three systems are ballot-marking devices that utilize only a paper ballot, thus the VVPAT standard (f) does not apply. The other systems are all DRE’s that were reviewed with a VVPAT.

The three ballot marking devices and one DRE VVPAT (AccuPoll) utilize “voter handled” paper ballots that must be placed in a ballot box during the voting process, thus feature (g) applies to those. The other DRE’s utilize a VVPAT that is not “voter handled” and feature (g) does not apply.

Items (f) and (g) are the only two features that are absent from all products currently on the market.

1) These DRE’s with an attached VVPAT were cumbersome or impossible to use for curbside voting -- while most DRE’s without a VVPAT were portable.

2) This system is the only one that provided a scanner to deliver audio read back of the VVPAT. However, the VVPAT had to be manually positioned by a blind voter under the scanner to enable to bar code to be read – making it non-function. In addition, no large print output was available.

General Note about Review : Each of these systems was reviewed “hands on” in April or August of 2005 and every attempt was made to independently confirm conformance to the standards rather than relying on vendor provided information. The system current at the time was reviewed. The following are the specific systems reviewed:

ES&S AutoMARK Voter Assist Terminal, Version 1.0
ES&S/Unisyn InkaVote
Populex Digital Paper Ballot System
Sequoia AVC Edge with VeriVote
Diebold AccuVote TSX DRE with printer module
AccuPoll DRE AVS-1000 v 2.5
Hart Intercivic e-Slate
ES&S iVotronic with real time paper audit trail
UniLect PATRIOT with VVPAT