Introduction

2

Pre-Print Copy: A Longitudinal Study of State Government

Home Page Accessibility in Maryland

and the Role of Web Page Templates for Improving Accessibility

Jonathan Lazar,

Radcliffe Institute for Advanced Study, Harvard University and

Department of Computer and Information Sciences, Towson University

8000 York Road, Towson, MD 21252

jlazar@towson.edu, 410-704-2255

Brian Wentz (corresponding author)

Department of Computer Science and Information Technologies

Frostburg State University

101 Braddock Rd, Frostburg, MD 21532

bjwentz@frostburg.edu, 301-687-4023

And

Abdulelah Almalhem

Alexander Catinella

Catalin Antonescu

Yeveniy Aynbinder

Michael Bands

Edward Bastress

Brandon Chan

Brian Chelden

Darin Feustel

Nabin Gautam

Whitney Gregg

Michael Heppding

Cory Householder

Alex Libby

Corey Melton

Jack Olgren

Loren Palestino

Morgan Ricks

Scott Rinebold

Matthew Seidel

(all students from Towson University)

Abstract

It is well documented that government, at all levels, continues to have problems ensuring that government web sites follow laws related to web accessibility for people with disabilities. Although there are a number of published studies on government web accessibility that are point-in-time, there are no published studies consisting of a longitudinal analysis of state-level government web site accessibility. This paper contributes to the research literature in three ways: 1) an accessibility inspection of 25 Maryland state government homepages in 2012 which involved 150 human inspections of web pages, 2) a comparison of the results from 2012 to a similar accessibility evaluation in 2009, and 3) a discussion of the role of a web page template, which was introduced in Maryland state government shortly after the 2009 evaluation. The data from this longitudinal evaluation leads to the conclusion that web page templates do tend to result in more accessible sites within state government.

Abstract

Web accessibility, policy, disability, compliance, Section 508, templates

1. Introduction

Since the mid-1990s, many government agencies, both at the state and federal level, have been distributing information to the public via web sites (Bertot, Jaeger, Shuler, Simmons, & Grimes, 2009). Due to recent budget cuts and the increased diffusion and presence of smart phones, government agencies often do not answer the phone or provide the level of citizen interaction as much as they did in the past, instead pointing people to their web sites (Bertot & Jaeger, 2008). In this context, the accessibility and usability of public information on government web sites is of increased importance. When a web site is designed using accessibility guidelines, it meets the needs of a broad range of users. Accessible web sites typically meet the needs of people with perceptual impairments (low vision or blind, deaf or hard of hearing), motor impairments (limited or no use of hands for pointing or typing), and some cognitive impairments. Accessible web sites (which are essentially web sites that are flexible to the user’s technology and environment) also tend to increase usability for users of mobile devices (Henry & Arch, n.d.).

US state and federal laws clearly require that government web sites follow design regulations to ensure that the widest range of people are able to access government information online. At the US federal level, Section 508 of the Rehabilitation Act requires that all federal technology developed or procured, be accessible for people with disabilities, including federal web sites. The technical standards for Section 508 at the federal level have been in effect since 2001, and they were based on the Web Content Accessibility Guidelines (WCAG) 1.0, an international standard since 1999. Since 2008, Web Content Accessibility Guidelines (WCAG) 2.0 has been the international standard. The US Access Board is in the process of updating the Section 508 standards to meet the newest international standards (WCAG 2.0), and in the most recent draft, the U.S. Access Board has indicated simply that WCAG 2.0 would become the new standard for web sites covered under the Federal Section 508. The US Access Board recently indicated that they expect to release a notice of proposed rulemaking on the new Section 508 standards, in June 2013.

At the state level, most US states have a state-level equivalent to Section 508, often adopting the same web accessibility technical standards as Section 508. Furthermore, the US Department of Justice has indicated, as early as 2003, that since state and local governments are covered under the Americans with Disabilities Act (ADA), state and local government web sites must be accessible for people with disabilities. In a recent notice, the Department of Justice stated that the rulemaking for state and local government websites (Title II of the ADA) will be separated out from the rulemaking for Public Accommodations (Title III) of the ADA, and the next notice of proposed rulemaking for state and local government web sites will be issued in July 2013.

While a number of data collection efforts have occurred relating to a point-in-time evaluation of state-level accessibility (see section 1.1 of this paper), there have not been any longitudinal studies of state-level web accessibility. Furthermore, no studies at either the US Federal or state level have examined the impact of introducing a web page template into governmental use. The purpose of this project was to compare the accessibility of Maryland State web site homepages in 2009 and 2012, to determine how accessibility levels have changed over time, and whether the introduction of web site templates by the state of Maryland in 2009 had any impact on accessibility levels.

1.1 Inaccessible Government Web Sites

Despite the existence of web accessibility requirements in the US for federal and state web sites, government web sites continue to have inaccessibility problems and violate laws that, in some cases, have been in existence for more than a decade (Fagen & Fagen, 2004; Goette, Collier, & Whilte, 2006; Jaeger, 2006; Olalere & Lazar, 2011; Rubaii-Barrett & Wise, 2008). There are many reasons often given for why government web sites continue to be inaccessible. Some potential reasons cited include: no ongoing compliance activities by government agencies, accessibility responsibility and activity distributed throughout state and federal government (i.e. no clear “accessibility czar” or similar concept), no openness or transparency requirements of how/if agencies perform the necessary accessibility activities, accessibility responsibilities are often added to the job responsibilities of someone who already has full-time job responsibilities, and there are no existing government guidelines on process and policy related to accessibility, only technical specifications. (Olalere & Lazar, 2011; Lazar & Jaeger, 2011). There is also the ongoing issue that web developers are simply not familiar with accessibility guidelines, because their formal and on-the-job training has not included a comprehensive exposure towards developing interfaces with accessibility in mind (Jaeger, 2011).

One important type of data that is generally missing from the discussion of government web site accessibility is a longitudinal study. While longitudinal studies of web accessibility have taken place for the web in general (Hackett, Parmento, & Zeng, 2004) and metropolitan areas (Lazar & Greenidge, 2006), longitudinal studies have not been published related to the accessibility of government web sites. Having longitudinal studies would be helpful in understanding how web site accessibility changes over time, and for understanding what factors help to move the accessibility needle in the right direction. It would seem that this type of data should be available from government agencies, but compliance activities at the U.S. federal level had simply stopped between 2003-2010 (so no data exists), and no data is yet publicly available from 2010-2012 for comparison purposes. At the state level, no such data is publicly available, and it’s unknown if such longitudinal data even exists for any states. Limited published research exists about state-level web site accessibility: Yu and Parmanto found that no state web sites met the WCAG 1.0 criteria, but that state government web sites generally had fewer accessibility barriers than Federal web sites (Yu & Parmanto, 2011). Youngblood and Mackiewicz found that many of the municipal web sites within Alabama had accessibility problems (Youngblood & Mackiewicz, 2012) . Rubaii-Barrett and Wise found that US states with stronger web site accessibility statements also tend to have more accessible web sites (Rubaii-Barrett & Wise, 2008). Goette, Collier and Whilte found that, in 2004 approximately a third of the state government home pages had at least one WCAG priority level 1 accessibility violation, and therefore were not accessible (Goette, Collier, & Whilte, 2006). Fagen and Fagen, found that in 2002, only 3/50 state legislature web sites were fully accessible (Fagen & Fagen, 2004).

1.2 Background of 2009 Study

There is a unique opportunity for a longitudinal study within the state of Maryland. An evaluation study of 15 Maryland state government web sites was performed in the spring of 2009, right before a new web page template was introduced throughout state government (Lazar et al., 2010). Maryland has had a regulation in effect since 2005, the Maryland IT Non-Visual Access Regulations, requiring the accessibility of state web sites, using technical standards identical to the existing Section 508 web guidelines (1194.22) at the US federal Level (which have been public since 2001), which are loosely based on the WCAG 1.0 standards (which have been public since 1999). The Maryland state web page template was introduced around the time of the 2009 evaluation study, and a number of state agencies have adopted the template since then. While the template is not required, it is highly suggested that when state agencies re-design their site, that they use the template. Figure 1 shows the template from the Maryland Department of Information Technology. The template instructions state that the design, functionality, and content must comply with the Maryland regulations for non-visual accessibility. A statement of accessibility must be displayed on the web page footer, best practices of universal usability must be followed, sites should undergo usability testing, sites should be evaluated for code compliance via the W3C, color contrast should be tested, and the site should be functional without images.

Figure 1.Screenshot of the Maryland State Web Site Template

During the 2009 study, a set of 15 Maryland government homepages were selected for evaluation: five web sites that focused specifically on people with disabilities and 10 general state web sites. By comparing the accessibility of Maryland State web sites in 2009 and 2012, it is possible to determine how accessibility levels have changed over time, and whether the introduction of web site templates by the state of Maryland in 2009 has had any impact on accessibility levels.

In the 2009 study, multiple human evaluators individually evaluated each agency home page for accessibility, using an expert inspection method (only the homepage was inspected, not the rest of the site). This method involves a three-step approach, checking a web page using a screen reader, navigating using a keyboard-only approach, and then inspecting the code for compliance with each paragraph of the Maryland state regulations. Only homepages were evaluated, because they are often the most accessible, and if users with disabilities encounter major problems on the homepage, it becomes less likely that they will be able to even access much of the content on other pages of the site, so homepages are of the utmost importance (Lazar & Greenidge, 2006). Furthermore, the number of guidelines (or paragraphs of the law) violated is considered to be the most accurate metric for determining accessibility (Lazar & Greenidge, 2006; Loiacono, McCoy, & Chin, 2005). This is because it is more challenging to fix three violations of three different paragraphs of the law, than it is to fix three violations of one guideline.

In contrast to human evaluations, evaluations conducted with automated software tools (like Deque Worldspace, Odellus ComplyFirst, and SSB InFocus) may be able to point out where potential violations exist, but automated evaluations are often not as accurate as human evaluations, because the automated tools can determine only presence of code, but not appropriateness or usefulness of labeling in context (e.g. “alt text” is not descriptive alternative text for a graphic). Multi-stage human inspections of web pages, involving screen readers (such as JAWS or VoiceOver, the technologies actually utilized by people with blindness, low vision, or some print-related disabilities) are considered to be the most accurate form of accessibility evaluation (Mankoff et al., 2005), and that accuracy increases when multiple individuals evaluate the same interfaces (Lazar et al., 2010). Screen readers, while geared towards blind users, are very helpful in identifying accessibility violations because they help point out where web page components are not accessible for keyboard-only use, which also impacts on how people with motor impairments interact with web pages. JAWS is the most popular screen reader in the Windows operating system, and VoiceOver is the screen reader that comes built into all recent Macs (WebAim, 2012). Therefore, they are the leading screen readers in their respective environments. The functionality of the two screen readers is similar, although the specific key combinations may differ, and therefore, it would take an adjustment for an expert-level JAWS user to switch to VoiceOver.

2. Research Methods

For the 2012 evaluation, the researchers took the 15 homepages evaluated in spring 2009, re-evaluated those 15 homepages using the same research methods and also selected 10 additional home pages. Those 10 additional sites were chosen by taking the links from the “find an agency” pull-down menu on the Maryland.gov homepage (with over 90 choices), and using a random number generator to select which sites to evaluate from that list. The home pages that were evaluated in both 2009 and 2012 are described in Table 1.

Table 1.Site Names and URLs for Home Pages Evaluated in Both 2009 and 2012

Name of Agency, Department or Service

Web Site URL

Governor’s Office for Children (new name since 2009)

http://goc.maryland.gov/

Governor’s Office of Minority Affairs

http://www.mdminoritybusiness.com/

Governor’s Office on Service and Volunteerism

http://www.gosv.maryland.gov/

MD Department of Aging

http://www.aging.maryland.gov/

MD Department of Assessments and Taxation

http://dat.state.md.us/

MD Department of Disabilities

http://www.mdod.maryland.gov/

MD Division of Rehabilitation Services

http://www.dors.state.md.us/dors

MD State Library for the Blind and Physically Handicapped

http://www.lbph.lib.md.us/

MD Motor Vehicle Administration

http://www.mva.maryland.gov/

MD Office of the People’s Counsel

http://www.opc.state.md.us/

MD State Treasurer

http://www.treasurer.state.md.us/

MD Supplemental Retirement Plan

http://msrp.maryland.gov/

MTA-Mobility

http://mta.maryland.gov/mobility/

Office of Lt. Governor

http://www.governor.maryland.gov/ltgovernor/

Rural Maryland Council

http://www.rural.state.md.us/

The approach for the 2012 accessibility evaluations remained the same as the 2009 study: multiple human evaluators individually evaluated each agency home page for accessibility, using a three-stage expert inspection method (only the homepage was inspected, not the rest of the site). Two different screen readers, JAWS (which is the most popular Windows-based screen reader) and VoiceOver (which comes installed with Mac OS X) were used for the evaluations, and violations were only reported in this study if the violation appeared using both JAWS and VoiceOver (since, in some instances, a problem appeared using JAWS but not VoiceOver). A combination of Internet Explorer, Safari, Google Chrome, and Firefox were used for the evaluations, to give the broadest view of potential accessibility problems. Multi-stage human inspections of web pages, involving screen readers (such as JAWS or VoiceOver) are considered to be the most accurate form of accessibility evaluation (Mankoff et al., 2005), and that accuracy increases when multiple individuals evaluate the same interfaces (Lazar et al., 2010).

The 25 home pages were inspected by six evaluators per page, resulting in a total of 150 human inspections of the homepages. Except for the additional homepages evaluated, the additional web browsers used for the evaluations, and the use of newer versions of JAWS (screen reader) and VoiceOver, this study was a replication of the 2009 study.

When each individual evaluator had completed their evaluation, all of the evaluators met and compared their results. This approach of using multiple human evaluators to come up with one meta-evaluation increases the reliability and accuracy of the inspection. This approach of a three-stage human inspection and multiple individual human inspections of the same page has been previously utilized in many web accessibility evaluations (Lazar et al., 2010; Lazar, Biggers et al., 2010; Lazar et al., 2012; Lazar et al., 2011; Wentz et al., 2012). It is important to note that these web page inspections took place in April and May 2012. Because web page content changes often, a follow-up inspection of homepages took place in July 2012, to confirm that any accessibility violations reported in May were still present. A number of sites had been modified by July 2012, and those confirmations are reflected in the results presented in this Article. All finidngs of this project were accurate as of July 2012, while the authors acknowledge that some of the sites may have been modified since.

The Maryland state guidelines, which were implemented in 2005, are identical to US Section 508 (Maryland DOIT, 2005). Because of this, these evaluations were based on the 16 guidelines set forth in Section 508 of the US Rehabilitation Act (1194.22), identified as paragraphs A through paragraph P that focus on web site accessibility. Table 2 lists the 16 paragraphs of the Section 508 guidelines and provides a short description of each guideline (note that the descriptions are from Lazar et al., 2010, not from the law).

Table 2.Description of Each of the 16 Paragraphs of the Section 508 Web Accessibility Guidelines

(A) Text Equivalent (have a text equivalent for any graphical elements)

(B) Synchronized Equivalent Alternatives (have captioned video, transcripts of any audio, or other alternatives for multimedia)

(C) Use of Color (color should not be used as the only method for identifying elements of the web page or any data)

(D) Organization (style sheets are encouraged, but users should still be able to utilize a web page when style sheets are turned off)

(E) Redundant Text Links on Server-Side Image Map and (F) Client-Side Image Maps (redundant clickable links for server-side image maps, and accessible client-side image maps are preferred)

(G) and (H) Row and Column Headers (use appropriate headers and markup to allow easy navigation of a table)

(I) Frames (title all frames and label all frames for easy identification and navigation, e.g., use “navigation” “main content” and “search” rather than “top” or “bottom”)

(J) Screen Flicker Frequency (limit or eliminate the use of flickering, which can provoke seizures)

(K) Text-Only Page Default (if a web page cannot be made accessible, provide an equivalent text-only page, and make sure it is kept up to date)

(L) Scripting Languages (make sure that equivalents for any non-accessible scripting are included, e.g., for those who are not using pointing devices)

(M) Linked Plug-In or Applet (if any plug-ins are required, make sure to provide a link to an accessible version of the plug-in)

(N) Online Electronic Forms (all forms must be properly labeled and accessible)

(O) Method to Skip Repetitive Navigation Links (all web pages should have a link which allows a user to skip directly to the main content, bypassing any site navigation information)

(P) Alerts on Timed Responses (if any page responses are timed, the user should be given the opportunity to indicate that more time is needed)

3. Results

3.1 New sites evaluated and sites with templates

The 10 additional home pages which were evaluated in 2012 (but not 2009) are described in Table 3.

Table 3.Site Names and URLs for Home Pages Evaluated Only in 2012

Name of Agency, Department or Service

Web Site URL

Governor’s Grants Office

http://grants.maryland.gov/Pages/grantshome.aspx

Injured Workers’ Insurance Fund

https://www.iwif.com/

MD Department of Housing and Community Development

http://www.mdhousing.org/Website/Default.aspx

MD Emergency Management Agency

http://mema.maryland.gov/

MD Environmental Service

http://www.menv.com/

MD Lottery

http://www.mdlottery.com/

MD Office of the Public Defender

http://www.opd.state.md.us/

MD Public School Construction Program

http://www.pscp.state.md.us/

MD State Retirement and Pension System

http://www.sra.state.md.us/

MD Tax Court

http://www.txcrt.state.md.us/

As mentioned previously, the Maryland state government introduced a standard web site template in 2009. By 2012, 11 of the home pages (as follows) were clearly using the template:

· Governor’s Grants Office

· Governor’s Office for Children

· Governor’s Office on Service and Volunteerism

· MD Department of Aging

· MD Department of Assessments and Taxation

· MD Department of Disabilities

· MD Department of Housing and Community Development

· MD Emergency Management Agency

· MD Motor Vehicle Agency

· MD Public School Construction Program

· Office of the Lieutenant Governor

Three other home pages (MTA-Mobility, Office of the People’s Counsel, and the Maryland State Treasurer) looked like they were using some features of the template, but not other features, which made the HTML code not similar enough to consider it to be a clear use of the state web template for this paper.

3.2 Accessibility violations

During the 2012 data collection, most of the Maryland State government homepages (23/25, or 92%) evaluated contained one or more violations of the Maryland IT Non-Visual Access Guidelines. The only two homepages evaluated, which had no accessibility violations, were the Maryland Division of Rehabilitation Services and the Maryland Library for the Blind and Physically Handicapped. For the 25 home pages evaluated in 2012, on average, they had violated 2.7 paragraphs of the law.

Only 15 of the sites evaluated in 2012 were also evaluated in 2009. For those 15 sites evaluated in both years, there were, on average, 2.1 paragraphs of the law violated in 2012, compared to the 2.5 paragraphs of the law violated in 2009, which is a slight improvement. Table 4 contains a matrix of the paragraphs with violations by each site. There were violations in paragraphs A, B, C, D, F, L, M, N, and O. Paragraph A (lack of text equivalent for graphical elements) was the most violated paragraph (15 out of 25 web sites), with paragraph M (linked plug-in or applet), paragraph N (properly labeled forms), and paragraph O (lack of method to skip repetitive navigational links) following closely behind.

Table 4.Matrix of Section 508 Paragraphs Violated by Each Home Page

Section 508

Paragraph:

A

B

C

D

E

F

G

H

I

J

K

L

M

N

O

P

Assessments/Taxation

X

Children

X

X

Dept of Aging

X

Dept of Disabilities

X

DORS

Emergency Mgmt

X

X

X

X

Enviro. Service

X

X

X

X

X

Governor’s Grants

X

X

Housing and Community Dev.

X

X

X

X

Injured Workers Ins.

X

X

X

X

Library for Blind and PH

Lottery

X

X

X

X

X

Lt. Governor

X

X

X

Minority Affairs

X

X

X

X

X

X

MTA-Mobility

X

X

X

MVA

X

People’s Counsel

X

X

X

X

Public Defender

X

X

Public School Construction

X

X

X

Retirement and Pension

X

X

X

Rural MD Council

X

X

X

Service/Volunteerism

X

Supplemental Retirement Plan

X

X

X

X

Tax Court

X

X

X

Treasurer

X

X

Examples of violations in paragraph A included an “eServices” sign-in link on the Injured Worker’s Insurance Fund web site, which was a graphic without alternate text. Without an accessible link, users who are blind would not be able to log in and report claims, pay bills, and print certificates of insurance. Figure 2 shows a screenshot of this violation.

Figure 2.The Sign-in Button Does Not Have a Text Equivalent

Paragraph B (violated on five of the web sites) requires synchronized alternatives for multimedia, such as audio and video clips on web sites. That means that if there is an audio clip, there must be a transcript. If there is a video clip, there must be captioning. An example of a home page which violated this guideline was the “Realtime Transit Report” on the MTA-Mobility web site. There is no transcript or captioning available for the report, which would prevent users who are deaf or hard of hearing from accessing this report. A screenshot of the link to the transit report is shown in Figure 3. Under the accessibility guidelines, there would also need to be a link to a transcript.

Figure 3.The “Realtime Transit Report” Does Not Have a Transcript

Paragraph C (information with color should be available without color) was violated on the Maryland Tax Court home page. An example of this can be seen in Figure 4, which shows the blue on blue menu navigation area of the Maryland Tax Court home page, which is insufficient color contrast.

Figure 4.The Navigation Menu Area on the Maryland Tax Court home page Has Poor Contrast

The Maryland Lottery and the Governor’s Office of Minority Affairs had violations of paragraph D (page unusable without the style sheet enabled). This is important because some users with disabilities use their own style sheets (often to change the color contrast and/or size of the text), and if the content is not designed to be usable without a particular style sheet, it is likely to create accessibility problems for such users. The Office of Minority affairs violated paragraph F because it had a client-side image map that was not accessible via keyboard. Paragraph L requires that content and functionality provided by scripting languages is also available to users of assistive technology or in some equivalent form on the page. Four of the home pages evaluated violated this guideline, and an example of this is illustrated by Figure 5, showing the menus on the homepage of the Maryland Environmental Service web site, which has a drop-down menu for navigation which is inaccessible to keyboard-only users. This JavaScript menu is missing code which would allow for users who are unable to use pointing devices to access the content. Essentially, if you cannot use a pointing device (which blind people and some people with motor impairments cannot do), you cannot use the navigation.

Figure 5.Inaccessible JavaScript Menu (Works Only with Pointing Devices)

When web page content requires a plug-in or applet, paragraph M requires that a link to an accessible version of that plug-in be provided (such as an Adobe Acrobat reader or Microsoft Word reader). Paragraph M was violated by 13 of the home pages evaluated. Paragraph N requires that all online forms be accessible and properly labeled, and this guideline was violated by 12 of the 25 home pages evaluated. Figure 6 depicts the Maryland Office of People’s Counsel home page, which is one of the state home pages that use the Maryland.gov web site template. The search form field is not properly labeled in the HTML template, and this problem exists on the home pages that strictly used the Maryland.gov web site template (note that the template itself has one violation—that the search form field is not properly labeled). In the visual scan of the screenshot, it is evident that the search field is associated with the search button, but for those using screen readers, the field should also have an HTML label. Note that some of the home pages that utilize the state template have fixed the problem with the search form label. A number of state homepages (such as the Maryland Department of Disabilities) which use the template have only one accessibility violation, and the violation is the search form field not being properly labeled.

Figure 6.The Search Form Field Does Not Have an HTML Label

Paragraph O requires a method for users to skip repetitive navigational links, specifically in the form of a link to skip to content. An example of paragraph O correctly implemented is shown on Figure 7, which shows a skip to content link at the top of the Maryland State Library for the Blind and Physically Handicapped home page.

Figure 7.Example of a Correct Link to Skip Repetitive Navigational Content

3.3 Accessibility statements

While it is not a legal requirement to provide an accessibility statement on a web page, it is suggested in the Maryland state government web page template, and good web practice also encourages a web accessibility statement. Of the 25 home pages evaluated in 2012, 12/25 home pages had a link to an accessibility policy. It is interesting to note that nine of the 11 web pages that clearly used the state template also had links to accessibility statements (Office of Assessments and Taxation, and the Motor Vehicle Administration did not). Of the three sites that used some components of the template, two had a link to an accessibility statement (MTA-Mobility and the Office of the Treasurer), and one did not (Office of the People’s Counsel). As noted earlier, the Office of the Treasurer home page was a page which had some similarities to the state web page template, but the code did not match the template in many places. Of the remaining 11 sites that did not use the template, only one had a link to an accessibility statement (The Division of Rehabilitation Services). Using the template clearly impacted on whether a site had a link to an accessibility statement. Of course, it’s also important to examine the content of the accessibility statement. Some of the accessibility statements were simple:

“The Maryland State Treasurer's Office is committed to making this web site compliant with the State's nonvisual accessibility requirements. (Code of Maryland Regulations (COMAR) 14.33.02.01-.12). For more information on these regulations and guidance, please visit the Maryland Information Technology Nonvisual Access Website.”

Other web pages (such as the Maryland Public School Construction Program) simply linked to the state regulations without being specific about the accessibility of the site. Two web pages (Office of the Lt. Governor and the Governor’s Office on Grants) linked directly to the Governor’s statement on web accessibility which provided more detail about accessibility than any other accessibility statement within the state government. And, in one confusing instance, the Governor’s Office on Service and Volunteerism actually has a link to an accessibility statement on a different site (it links to the accessibility statement for the Governor’s Office on Community Initiatives).

3.3 Comparison between 2009 and 2012

A comparison of the 2009 data and the 2012 data can provide interesting insights for discussion (see Table 5 for a comparison of the sites which were evaluated in both 2009 and 2012). As mentioned earlier in the paper, for the 15 sites evaluated in both 2009 and 2012, there were, on average, 2.1 paragraphs of the law violated in 2012, compared to the 2.5 paragraph of the law violated in 2009, which is a slight improvement. Some homepages had the same paragraph violations in 2009 and 2012, such as the Maryland Department of Disabilities (which violated paragraph N in both years) and the Office of the People’s Counsel. The Division of Rehabilitation Services was in full compliance in both 2009 and 2012. Some homepages improved from 2009 to 2012, such as the Maryland State Library for the Blind and Physically Handicapped, which had two paragraph violations in 2009 but none in 2012, or the Office of Assessments and Taxation, which had two paragraph violations in 2009 but only one violation in 2012. The Office of the Lieutenant Governor homepage had decreased accessibility, with two paragraph violations in 2009 but three paragraph violations in 2012.

Table 5. Comparative Results of the 15 Sites Evaluated in both 2009 and 2012

Paragraph:

2012 violations

2009 violations

Assessment/

Taxation

A

C,L

Children

N, O

A, I

Dept of Aging

N

A, O

Dept of Disabilities

N

N

DORS

None

None

Library for Blind and PH

None

A, M

Lt. Governor

B, M, N

A, L

Minority Affairs

A, D, F,

M, N, O

A, E, F,

L, M, O

MTA-Mobility

A, B, M

L, M, N

MVA

A

L

People’s Counsel

A, L, N, O

A, L, N, O

Rural MD Council

A, M, O

A, M, O

Service/Volunt.

N

A, M

Supplemental Retirement Plan

A, L, M, O

A, D, M, O

Treasurer

A, L

A, L, N, O

Discussion

While there are some improvements between 2009 and 2012, accessibility compliance on Maryland state home pages continues to be a problem. Most state home pages do have some form of accessibility violation, and for some state agencies, the number of accessibility violations have increased since 2009. However, for the 15 home pages that were evaluated in both 2009 and 2012, there was a slight improvement in accessibility. There is another positive trend: the use of the Maryland state web page template has improved the accessibility of some home pages. When analyzing the results of this study, there was a clear accessibility benefit from using the template: on average, the state homepages using the template had better accessibility than the homepages not using the template. The 11 home pages which clearly used the Maryland.gov template had violations in an average of 2.1 paragraphs per homepage. In contrast, the 14 web sites which did not clearly use the Maryland.gov template had violations in an average of 3.1 paragraphs per homepage (and, when removing the two agencies that focus on disability but didn’t use the template, the number of violations increases to 3.7 paragraphs violated per homepage. While an increase of only one new type of violation may sound insignificant, the presence of an additional area of accessibility violation could significantly decrease the likelihood that a user with a particular disability can fully use the web page. For example, if the Maryland web page template was fully fixed to be accessible (by modifying the search box), this could now enable a user who is blind to submit information via an electronic form. Another example would be a site that has a paragraph A violation. The difference between a site that has text equivalents for graphical components, and one that does not, could mean the difference between a site that is primarily usable and one that is entirely unusable for an individual without sight.

Two of the state agencies that did not use the template include the Maryland Division of Rehabilitation Services and the State Library for the Blind and Physically Handicapped, but both home pages were fully accessible. If these two accessibility-conscious organizations, which had no accessibility violations, are removed from the analysis, the average for the rest of the non-template-using agencies rose to 3.7 paragraphs violated per homepage. If the three web sites that had partial use of the template are also removed, the average rises to 3.9. This illustrates the value of a template and also underscores the impact that a few minor modifications to the template could have on many web sites. For example, providing standard links to plug-ins and readers on the template as well as fixing the search form field label would remove two major categories of violations from many of these web sites (See Figure 9 for a suggestion on where some of these improvements to the template could go).

When a larger entity (such as state government or a large organization with many web sites) implements a standard web site template across its web pages, this often does improve accessibility. While there were still some accessibility violations on some pages that used the template, many of the major and repetitive violations that are often found on web sites were minimized on the sites using the template. For instance, the Maryland Department of Disabilities only has one accessibility violation, and it is the violation that is included in the state template (the search field that is not properly labeled). Several home pages evaluated used the Maryland state template and only had one error, otherwise being fully accessible, including the Department of Aging, Department of Disabilities, and the Governor’s Office on Service and Volunteerism. Figure 8 shows a screenshot of the Department of Aging home page, which was otherwise fully accessible.

Figure 8.Screenshot of the MD Department of Aging Home Page, Which Uses the MD State Web Site Template, and Is Accessible Except for the Search Form Field

It should be noted that the Maryland Department of Assessments and Taxation, the Motor Vehicle Administration, and MTA-Mobility, while using the Maryland State web site template, added the “label for” element to their search form code, making their search field accessible. The Maryland State Treasurer home page uses the “title” attribute to achieve a similar result. The appearance on those web pages is visually identical to Figure 8, and this underscores that such an accessibility fix has no impact to the overall visual appearance of the web site, but makes the form field more accessible to users of screen readers.

Figure 9.Screenshot of the MD State Web Site Template as Suggested by the Department of Information Technology, with Indications of Suggestions for Where Some of the Corrections Could Be Implemented

The Governor’s Office of Minority Affairs home page violated six paragraphs of Section 508 in the 2012 evaluation. The home page had many linked images without alternate text, hyperlinks that were empty, and link text that does not make sense to a non-sighted user which resulted in violations of paragraph A. There was an image map with links such as “Secretary Welcome,” “Business Startup,” and “Business Growth.” This image map violated paragraph F since it is missing alternate text or an alternate means of accessing the same content on the page. Paragraph M was violated because there are many linked PDF and Word documents on the home page that do not also have a link to an accessible viewer as required. Paragraph N requires accessible form labels (which generally require HTML label elements). The search form field at the upper right-hand corner of the home page had a search field without the “label for” element. Also, there is no method to skip repetitive navigational links on the home page (provided by a skip navigation link). Figure 10 shows a screenshot of the home page illustrating where some of these violations would be corrected.

Figure 10. Screenshot of the Governor’s Office of Minority Affairs Home Page, Illustrating some of the Violations from the 2012 Evaluation

In summary, the data from this evaluation of Maryland state homepages shows: 1) that web page accessibility within Maryland state government continues to be a problem, 2) that use of a web page template in state government can improve accessibility, but 3) only when the web page template itself is fully accessible.

4. Conclusion and Policy Recommendations

Web accessibility among Maryland state agency homepages continues to be a problem, despite clear laws in effect since 2005. It is likely that other state government web pages have accessibility problems, and it is well-documented that federal web pages continue to have accessibility problems (Olalere & Lazar, 2011). Among the 15 sites evaluated in both 2009 and 2012, there were slight improvements in accessibility. Furthermore, the impact of the new state web page template was remarkable. For homepages that clearly utilized the state web page template, there were much higher levels of accessibility and a greater likelihood that a web accessibility statement was present. The existing template has minor flaws related to accessibility, and it is likely that if the web page template was re-designed to include improved accessibility, and the web sites were updated to use the new template, it would result in even further improvement for a number of Maryland state agency web sites. We therefore offer three policy recommendations:

1. Develop web page templates, and ensure that those templates are fully accessible. There was a clear improvement in accessibility for state homepages that adopted the web page template since 2009. The page template helps to fill in the gap between wanting to improve accessibility and knowing what to do to improve accessibility. Often, public policies provide design guidelines for accessibility, but not advice on implementation or maintenance or ongoing compliance. A web page template helps bridge that gap, to some extent, by providing a starting point for developers, and a starting point which already incorporates accessibility features. Of course, it is important to ensure that the page templates incorporate ALL accessibility features required under state or federal law. It is likely that, if the Maryland web page template had incorporated other accessibility features (such as a properly-coded search box or links to accessible plug-ins) that the accessibility levels for homepages utilizing the template would be even higher. Most web developers have limited time to learn everything that they need to learn about accessibility. Page templates help by showing developers how those accessibility features are required to be implemented under applicable laws. If developers forget about specific aspects of accessibility, the features listed on the web page template can serve as a reminder. Of course, the downside is that, when the web page templates are assumed to be accessible, the web developers often investigate no further, and do not check to see if any accessibility features are missing.

2. Require use of web page templates when web sites are re-designed. There was a clear relationship in this study between improved accessibility and use of the web page template. The question is then, why aren’t all of the Maryland state government web sites using the web page template? Often, web sites are re-designed every 2-3 years, but this timeline is often extended for state government web sites, which may only see a web site re-design once every 5-7 years or even longer. It is unrealistic to require all state web sites to implement a web page template in a short time period (a year or so) because often, their development lifecycles and planning are longer-term than that. Most state accessibility policies recognize that fact. For instance, there was nearly a five year window in Maryland, between the implementation of the IT Non-Visual Access Regulatory Standards (2000), and the date that all state agencies were required to have accessible web sites (2005). Once a web page template is approved and implemented, it would be reasonable to require template usage by state web sites in one of two ways: either requiring that the web page template be implemented in a 2-3 year timeframe, or requiring that the page template be implemented the next time that a re-design of the site takes place. If no site re-design is currently planned, then perhaps requiring implementation in a 2-3 year timeframe is the better approach.

3. Require that ongoing compliance activities take place, and that accessibility efforts are clearly posted. One of the challenges of ongoing compliance is that, simply put, most federal and state governments have no idea if their web sites are currently compliant, or what steps have taken place (if any) towards accessibility compliance. Often, when advocacy groups file complaints about inaccessible government web sites, those advocacy groups are asked to explain why a site is inaccessible. Most federal and state government agencies really are not monitoring accessibility compliance, and do not collect ongoing data about site compliance. As mentioned earlier in the paper, there is very limited data available about state government web site compliance. Government compliance activities need to document, at least semi-annually, how effectively (or not) government web sites are complying with the relevant accessibility laws. With ongoing compliance activities, not only will there be better data available for policymakers, but also, the compliance activities themselves bring attention and awareness to the topic of accessibility, also potentially increasing the likelihood of compliance. Without ongoing compliance activities, government agencies can actually decrease in compliance. For instance, the homepage of the Office of the Lieutenant Governor decreased in accessibility from 2009-2012 with the addition of inaccessible multimedia components such as video without captions, and by not providing links to the required plug-ins.

Each government web site should also clearly state in an accessibility policy, what steps are being taken to ensure accessibility compliance. While most accessibility policies on government web sites simply state that they are in compliance with the relevant laws, a statement should provide the following information: (1) a description of the accessibility features built into the site, (2) a description of the user testing, expert inspection, and automated review utilized to check for accessibility before new information or pages get posted on the site, (3) a description of how often the site is checked for accessibility compliance, (4) a contact person for accessibility compliance within the agency (where complaints or problems can get sent), and (5) a link to where results of accessibility testing are posted (Olalere & Lazar, 2011).

The results of this study illustrate the ongoing necessity for government agencies to continually evaluate the accessibility of their web sites. The longitudinal analysis of homepage accessibility validated the benefits that can be achieved through implementing agency-wide or company wide web templates that are designed and evaluated for accessibility. It is clear that more frequent use of accessible web site templates can result in improved accessibility for the users of those web sites. When web sites are re-designed, they should incorporate the use of accessible web site templates, and those templates (as well as the content added to the template structure) should be regularly evaluated for accessibility. Not only do the templates help enforce accessibility, but they also provide for consistent branding and usability across a state government.

One final thought: it will be important to monitor the rulemaking process for web sites of state and local governments under Title II of the Americans with Disabilities Act. In the next few years, there may be requirements from the US Federal Government that are more standardized throughout all of the states. Currently, each state government is free to define web accessibility in their own way, although most just adopt the Federal Section 508 technical guidelines. In the future, there may be more strict requirements, either for adopting a specific technical standard, for reporting accessibility evaluations, or both.

References

Bertot, J. C., Jaeger, P. T., Shuler, J. A., Simmons, S. N., & Grimes, J. M. (2009). Reconciling government documents and e-government: Government information in policy, librarianship, and education. Government Information Quarterly, 26, 433-436.

Bertot, J. C., & Jaeger, P. T. (2008). The e-government paradox: Better customer service doesn’t necessarily cost less. Government Information Quarterly, 25, 149-154.

Fagen, J., & Fagen, B. (2004). An accessibility study of state legislative web sites. Government Information Quarterly, 21(1), 65-85.

Goette, T., Collier, C., & Whilte, J. (2006). An exploratory study of the accessibility of state government web sites. Universal Access in the Information Society, 5(1), 41-50.

Hackett, S., Parmento, B., & Zeng, X. (2004). Accessibility of Internet Websites through Time. Proceedings of the ACM Conference on Assistive Technology (ASSETS), 32-39.

Henry, S. & Arch, A. (n.d.). Developing a web accessibility business case for your organization: Overview. W3C WAI, http://www.w3.org/WAI/bcase/.

Jaeger, P. (2006). Assessing Section 508 compliance on federal e-government Web sites: A multi-method, user-centered evaluation of accessibility for persons with disabilities. Government Information Quarterly, 23(2), 169-190.

Jaeger, P. (2011). Disability and the Internet: Confronting a Digital Divide. Boulder: Lynne Rienner Publishers.

Lazar, J., Beavan, P, Brown, J., Coffey, D., Nolf, B., Poole, R., Turk, R., Waith, V., Wall, T., Weber, K., & Wenger, B. (2010). Investigating the accessibility of state government websites in Maryland. In P. Langdon, P. Clarkson and P. Robinson (Eds.), Designing inclusive interactions--Proceedings of the 2010 Cambridge Workshop on Universal Access and Assistive Technology (pp. 69-78). London: Springer-Verlag.

Lazar, J.,Wentz, J., Biggers, D., DeLair, J., Donnelly, M., Eludoyin, K., Henin, A., Markakis, J., Matos, A., McNichol, A., Nixon, J., Osbourne, R., Postnova, T., Raja, J., Roberts, R., Rothbard, J., Serra, H., Sfakianoudis, V., Tyler, V., & Yun, J. (2011) Societal Inclusion: Evaluating the Accessibility of Job Placement and Travel Web Sites. Proceedings of the INCLUDE 2011 conference.

Lazar, J., & Greenidge, K. (2006). One year older, but not necessarily wiser: An Evaluation of Homepage Accessibility Problems Over Time. Universal Access in the Information Society, 4(4), 285-291.

Lazar, J., & Jaeger, P. (2011). Reducing Barriers to Online Access for People With Disabilities. Issues in Science and Technology, 27(2), 68-82.

Lazar, J., Wentz, B., Akeley, C., Almulhim, M., Barmoy, S., Beavan, P., C. Beck, Blair, Bortz, A. Bradley, B., Carter, M., Crouch, D., Dehmer, G., Gorman, M., Gregory, C., Lanier, E., McIntee, A., Nelson Jr., R., Ritgert, D., Rogers Jr., R., Rosenwald, S., Sullivan, S., Wells, J., Willis, C., Wingo-Jones, K., & Yatto, T. (2012). Equal Access to Information? Evaluating the Accessibility of Public Library Websites in the State of Maryland. In P. Langdon, J. Clarkson, J. Robinson, J. Lazar & A. Heylighen (Eds.), Designing Inclusive Systems: Designing Inclusion for Real-world Applications (London: Springer-Verlag), 185-194.

Loiacono, E., McCoy, S., & Chin, W. (2005). Federal web site accessibility for people with disabilities. Information Technology Professional, 7(1), 27-31.

Olalere, A., & Lazar, J. (2011). Accessibility of U.S. federal government home pages: Section 508 compliance and site accessibility statements Government Information Quarterly, 28(3), 303-309.

Rubaii-Barrett, N., & Wise, L. (2008). Disability Access and e-government: an empirical analysis of state practices. Journal of Disability Policy Studies, 19(1), 52-64.

WebAim (2012). Screen reader use survey #4 results. Available at: http://webaim.org/projects/screenreadersurvey4/.

Wentz, B., Cirba, M., Kharal, N., Moran, J., & Slate, M. (2012). Evaluating the Accessibility and Usability of Blogging Platforms for Blind Users. In P. Langdon, J. Clarkson, P. Robinson, J. Lazar & A. Heylighen (Eds.), Designing Inclusive Systems (London: Springer, 43-52).

Youngblood, N., & Mackiewicz, J. (2012). A usability analysis of municipal government website home pages in Alabama. Government Information Quarterly, 29(4), 582-588.

Yu, D., & Parmanto, B. (2011). U.S. state government websites demonstrate better in terms of accessibility compared to federal government and commercial websites. Government Information Quarterly, 28(4), 484-490.

A link should be in the page code here to skip to main content

The code for this form field needs a “label for” element

Link to PDF reader could go here

Example of one of many images missing alternate text

Search form field is missing label

Image map is missing alternate text or redundant links elsewhere on the page

� http://www.w3.org/TR/WCAG20/

� http://www.access-board.gov/508.htm

� http://accessibility.gtri.gatech.edu/sitid/stateLawAtGlance.php contains a list of US state-level web accessibility laws

� http://www.ada.gov/websites2.htm

� http://www.reginfo.gov/public/do/eAgendaViewRule?pubId=201210&RIN=1190-AA65

� Information about the Maryland Department of Information Technology page template is available at: http://doit.maryland.gov/webcom/Pages/Standards.aspx

� http://www.governor.maryland.gov/accessibility.asp