=Paper=
{{Paper
|id=None
|storemode=property
|title=Experienced Barriers in Web Applications and Their Comparison to the WCAG Guidelines
|pdfUrl=https://ceur-ws.org/Vol-792/Ruth_Janneck.pdf
|volume=Vol-792
}}
==Experienced Barriers in Web Applications and Their Comparison to the WCAG Guidelines==
https://ceur-ws.org/Vol-792/Ruth_Janneck.pdf
Experienced Barriers in Web Applications and Their
Comparison to the WCAG Guidelines
Diana Ruth-Janneck
Technical University of Dresden, Private Lectureship Applied Computer Science
01062 Dresden, Germany
diana.ruth-janneck@pdai.de
Abstract. A German organization established the study “Web 2.0 Accessible”
in 2008 which asked Internet users with disabilities about the barriers they ex-
perience when using websites and web applications. This paper gives some in-
teresting facts of the study which are useful for the design and implementation
of accessible web applications. Therefore it raises classifications of barriers in
various dimensions we have registered in the German study regarding the use of
web applications by persons with disabilities and correlate these barriers with
the WCAG 2.0 Guidelines and Principles to support the facts of the study. Fur-
thermore the results will be supported by key results of other studies regarding
web accessibility. The objective of this paper is to identify in practise experi-
enced barriers for people with disabilities using websites and applications.
Keywords: Web Accessibility, Barriers, People with disabilities using the
Internet, WCAG 2.0, Study Web2.0 Accessible.
1 Introduction
In our today’s more and more digital world, it is necessary that web pages and web
applications are accessible for all users and especially for users with disabilities. More
than 8.6 million people with disabilities live in Germany. Their number represents
about 10.5% of the German population [1]. The European Commission assumes that
one fifth of the working age population have a disability and almost 60% of the popu-
lation would be likely to benefit from web accessibility [2]. And this number does not
include elderly people, which often experience similar problems to access web pages
and their interactions. Thus, accessible web applications can be an important step to
an inclusive web for all. But nowadays there are many problems and restrictions for
users with disabilities. A German study was established in 2008 for register the actual
state of the art regarding the Internet access by people with disabilities. The results
provide important and practically relevant aspects for designers, developers and
evaluators of accessible web applications. It outlines the most critical technical, de-
sign and editorial barriers for different user groups with disabilities. These results are
very useful in connection with the WCAG 2.0 Checkpoints and the statements of the
questioned user.
This paper gives an insight into the German study “Web2.0 Accessible” [3] at first.
The most important applications for Internet users with disabilities and the applica-
tions showing the highest problem rates compiled in the study are especially interest-
ing here. The second part of the paper consists of the comparison of the extracted
most critical issues of web applications with the actual WCAG 2.0 Checkpoints [4].
At the end of this paper, the most important aspects of accessible web applications are
summarized.
1.1 Methodology and Related Works
The following statements result from a qualitative analysis of the data from the online
survey in connection with the transcription of the statements given in the interviews.
The most important applications for Internet users with disabilities have been
identified due to the highest usage frequencies in the different user groups. These
applications have been compared to the recorded problem rates and to the
interviewees’ statements. These results have been the basis of developing the most
important accessibility issues and critical aspects for the different user groups with
disabilities and of connecting them with the corresponding WCAG Checkpoints.
The facts about the barriers which are experienced in practice by people with disabili-
ties are supported by the results of two other studies concerning web accessibility
with user involvement. One important study was established by the Disability Rights
Commission (DRC) in the UK [5] in 2004. It is a large-scale study which tested 1.000
popular British sites for technical compliance with the WCAG 1.0 Checkpoints.
Furthermore, 10% of these sites were tested with a group of 50 users with different
impairments and by accessibility experts. The user statements and experienced key
problems are the most interesting points for this work. Another helpful survey was a
study on screen readers with 100 blind users in the US in 2007 by Lazar et.al. [6].
This study, whose findings have supported our results, recorded frustrations for blind
users using the web. A lot of other studies and research look for e.g. the compliance
of government websites (e.g. MeAC [2] or Lopes et.al.[7]) or of popular websites
(e.g. Sullivan et.al. [8]) with the WCAG or with accessibility barriers for older people
(e.g. Sayago et.al. [9]).
2 The German Study “Web 2.0 Accessible”
The German organization “Aktion Mensch” established the study "Opportunities and
Risks of the Internet of the Future from the Perspective of People with Disabilities"
[3] regarding the use of web2.0 applications by disabled people in 2008 in whose
evaluation the author significantly participated. This study offers reliable statistical
data concerning the use of web applications by people with disabilities as well as
which barriers and problems of use occur. It forms the statistical and qualitative basis
for the statements and classifications made in this paper.
2.1 Methodology of the Study
The study involved three steps of the inclusion of data in order to collect both
quantitative and qualitative data. On the one hand, experts from science and self-help
organizations have been consulted to capture the current state of knowledge on
Internet use by people with disabilities. Additionally, experienced Internet users with
disabilities were questioned in group interviews about their experiences and habits
with web2.0 applications as well as about the barriers they experienced. The data
from these steps about possible barriers in web applications were used for the con-
crete realization of the online survey which measured the scope of these barriers.
Disabled Internet users were questioned about use habits and barriers with the help of
an accessible online survey including audio files and sign language videos. The online
survey was organized in multiple areas to acquire several data [3, 10]:
• Demographic characteristics including kinds of disabilities and frequency of use of
the web (participants without an impairment or with sporadic use were excluded
here)
• Data on the use of assistive technologies and technical equipment
• Free text answers to “What is the best thing on the Internet for you?” and “What is
the most annoying thing on the Internet for you?”
• Data on the use behavior and intentions of use by selection from given lists (“I use
websites for…” and “With the Internet I can …”)
• Prominence of different popular websites (“Which kinds of website do you know?”
If the participant does not know the website he will not get more questions about.)
• Questions about the prominence, the share of use (from a list) and experienced
barriers for the familiar websites; for every barrier noticed: free text answer for
more explanation and a question about the kind of barrier (not operable, not
perceivable, not understandable, no orientation)
• Graduated compliance for special advantages by using websites for disabled
Internet users in a list of statements.
The survey was supported with the help of special German websites for people with
disabilities and self-help organizations in January and February 2008. A total of 671
people with disabilities have completed the questionnaire. This allows us to make
precise statements on the test results from the perspective of the different kinds of
disabilities, but not on all Internet users in general or all users with disabilities
because of the non-representative basis [3].
2.2 Facts about the Study
People with different kinds of disabilities were interviewed during the study. In total,
10 people were interviewed as experts, 57 people were questioned in group-
interviews and 671 people have completed the online-survey. The following table
gives the total number of questioned people with the different types of disabilities and
the assistive technologies used most often in these groups [3].
table 1. Numbers of Questioned People and the Assistive Technologies Used Most
Often
Type of Disability Number of Assistive Technologies Used Most Often
Questioned
People 1
Visual Impairment 133 Screen magnifier (56%), audio response
(22%), Screen reader (21%)
Blindness 124 Screen reader (91%), Braille terminal
(85%), audio response (70%)
Hardness of Hearing 96 Audio response (16%), screen magnifier
(13%)
Deafness 260 Screen magnifier (7%), audio response
(6%)
Motor and Dexterity 75 Special scroll wheels or trackball mouse
Impairments (20%), special keyboard/on-screen key-
board (17%), voice recognition software
(16%)
Dyslexia 41 Screen magnifier (32%), audio response
(24%), spell assist programs and voice-
recognition facilities (20%)
Learning and Cognitive 35 and 13 Screen magnifier (22%), audio response
Impairments (20%), Screen reader (13%)
The number of respondents in the groups Dyslexia and Learning and Cognitive
Impairments was too small to make reliable statements about the barriers and strate-
gies, but only tendencies could be deduced from the answers. We can make some
statistical and qualitative statements about the experienced barriers and useful strate-
gies concerning the other groups. The information is complemented by free text an-
swers given in the online survey and statements from the interviews. This has given
us some interesting and concrete facts about the strategies of and barriers for Internet
users with disabilities.
These different kinds of disabilities are summarized for the following considerations
into groups because the Internet is used with the help of similar assistive technologies
(AT) or use strategies due to the respective disability-related restrictions. A definition
and differentiation of the types of disabilities cannot be given at this point. All in all,
it is remarkable that the magnification software was considered as of significant usage
share by all surveyed user groups which is probably due to multiple restrictions or
simplification of the perception of contents.
1 Because of 82 participants with multiple impairments, the total number of questioned people
is less then the sum over all user groups. These persons provide data for each affected user
group with disabilities.
The people questioned distinguished themselves by a high technical standard of the
Internet access so that it can be assumed that identified barriers are not caused by
lacking technical equipment. The results of the study proved furthermore that the
interviewees are very experienced in dealing with Internet applications, they show
high usage frequency and have above-average experiences with web2.0 applications
so that also factors like insecurity or low affinity to the Internet can be excluded [3].
A comparison with an annual German study (ARD/ZDF-Online-Study 2010) about
the use of the Internet, which takes the whole German population into account, shows
that Internet users with disabilities use the web about 6.5 days/week, internet users
without disabilities use it only 5.1 days/week [1]. This fact underlines another result
of the study “Web 2.0 Accessible”: the Internet is one of the most important things in
daily life for users with disabilities and it can help to live in a far more independent
way. It is a tool for information and communication. For more than 40% of the
questioned users with disabilities, Internet is a tool for the compensation of disability-
related disadvantages [3].
Web accessibility is a very important fact for all questioned Internet users,
particularly for blind users. 88% of them chose “accessibility is very important for
me”. Furthermore, 82% of the questioned people with blindness said that barriers
disturb their access to the content. But web accessibility is also very important for
70% of the users with deafness and 70% of the users with cognitive impairments. The
users with visual impairments have the smallest coincidence with 59% [3]. This is
explainable with the circumstance that these users can manage accessibility problems
in different ways. Accessibility problems are often invincible for users who are
dependent on assistive technologies.
2.4 The Most Important Applications
The study inquired, among other things, prominence, use and problems in dealing
with various web2.0 applications and their functions. The most significant shares of
utilization (more than 60% of use) and the highest problem rates (more than 20%)
should indicate the value of certain application classes for the different user groups in
the following comparison. Thus, it can be derived which applications are used very
often and in which of them problems are frequently noticed.
Applications with high usage and problem rates should increase the efforts with re-
spect to accessibility. The shares of use given in the table were determined during the
quantitative part of the study by the online survey. It indicates the proportion of those
who have used or tried to use the application. The problem rate is calculated from the
quotient of the problems and the use / attempt to use and provides a projection of the
anticipated problems in the use of the application by the user groups [3].
table 2. Maximum Share of Use and Rate of Problems 2 [modified according to [3,
11])
Visually Handi- Blind Hard of Deaf Motor Dis-
capped Hearing abilities
Share of Read wikis Read wikis Read Read wikis Read wikis
Use > (79%), (85%), wikis (61%), (84%),
60% (68%),
Make user regis- Make user View photos Make user
tration (75%), registration View (60%) registration
(80%), photos (71%),
View photos
(60%)
(70%), Write com- View pho-
ments tos (65%)
View videos
(60%),
(61%)
Listen to
podcasts
(60%)
Rate of Make user regis- Make user View Read wikis Write
Pro- tration (41%), registration videos (26%), comments
blems > (69%), (33%) (28%),
Edit user profile View videos
20%
(30%), Edit user (23%), Edit user
profile profile
View videos Write com-
(58%), (20%)
(28%), ments (21%)
View videos
Write comments
(31%),
(25%),
Write com-
Read weblogs
ments
(25%),
(30%)
View photos
(23%),
Listen to pod-
casts (21%)
Prob- Captchas, Captchas, Quality Quality of Navigation,
lematic of me- media files, orientation
Completion of Completion
Aspects dia files, and oper-
forms, of forms, Linguistic
ability of
Audi- problems,
Visual fields of Visual web appli-
tory
applications fields of Auditory cations
fields
applications fields
2 There are no reliable data about the groups Dyslexia, Learning disabilities and Cognitive Im-
pairments given because the basis of results is too small (see table 1).
Across all user groups, which have been questioned, the highest shares of use are to
be registered concerning the reading of wikis. This kind of website causes less prob-
lems (problem rate between 6% and 13%) for the most user groups. Deaf Internet
users show the highest problem rate here (26%) because of problems in understanding
the content 3 . The questioned participants mainly have problems with comprehensibil-
ity of the (user-generated) content (48% of all who have problems) and orientation on
the website (39% of all who have problems) because of numerous links.
High shares of use are also mentioned for user registration in all groups. This is be-
cause a lot of web applications and services require a registration for full access.
Problems here are based on inaccessible forms from the technical point of view and
incomprehensibility of the explanations for the required data from the editorial point
of view. Thus, 73% of those who have problems with a user registration chose “not
perceivable” and 66% chose for “not operable” [3]. Visually impaired, blind and
physically disabled persons are mostly confronted with these problems.
Much the same applies for editing user profiles and the use of forms. Forms and in
particular Captchas limit the independent use here. Especially users who use AT have
problems with orientation on the website (53% of all who have problems), with
operability of the form elements (50% of all who have problems) and with the
perceptibility of the elements (47% of all who have problems). Similar problems are
noticed on form-based and editor-based applications such as the writing in wikis or
weblogs. The users show great interest in writing comments (e.g. 60% of blind users),
but problems like those mentioned above limit the easy access and interaction [10].
Moreover, the table shows that partially sighted and blind users indicated problems
with the use of primarily visual media like photos and videos. Disability-related re-
strictions and bad media quality are the main cause of it. Even hearing impaired and
deaf Internet users stated problems with visual media which are caused by insufficient
media quality as well as with the unlimited operability and availability of appropriate
media players. Statements from free text answers [3] have given some more aspects:
too small images, too small video windows or poor resolution, audio streams which
are too noisy or added with background noise, unavailability of subtitles or sign
language videos. Despite problems in access and use of media content, show the high
share of use of visual and auditory content in all questioned user groups big interests
for these. This fact should emphasize the importance of making this content as
accessible as possible for all.
3 Comparison of the Results with WCAG
Various areas of accountability and contributors in the development and operation
process of a web application have to be determined for the prevention of barriers.
3 The linguistic barrier concerning reading and writing, for instance of wikis and comments,
applies to deaf users because the German sign language differs substantially from the spoken
and written language. Thus, deaf people experience a more difficult access to the written
language.
They should be responsible for ensuring accessibility in their respective field of ac-
tion. For the following comparison of the study results with the WCAG 2.0 Check-
points (CP), the facts have been organized in relation to the area of accountability.
Different areas of accountability have been defined for this purpose which correlates
to the different positions in the development process of web applications like devel-
opers, authors, designers and customers. According to the description of the “Essen-
tial Components of Web Accessibility” from the W3C Web Accessibility Initiative 4
(WAI) Group [12] which presents interdependencies between the components and
roles, the four different areas of accountability are not independent at all. It is for this
reason that some items in the tables are assigned to several areas. The four defined
areas are: technical aspects, editorial aspects, design aspects and organizational as-
pects of web accessibility. A description of each can be found in the following para-
graphs.
Furthermore, three tables concerning technical barriers, editorial barriers and design
barriers have been developed. In the planning and development of a web application,
the different person in charge can consult the specific table to look up the most critical
aspects and use cases compiled during the study und which WCAG Checkpoint corre-
lates to them. In connection with the guidelines and technical documents of the W3C
WAI Group, the responsible person can choose the relevant checkpoints and select the
corresponding success criterion and best practise.
For the comparison, the items of the classifications of barriers published in [11 and
13] were each collated with the WCAG 2.0 Checkpoints. For the better readability the
results have been spread into one table for each area of accountability. Quantitative
results cannot be given for the aspects in the tables because for one thing the items are
grouped according to the impairment groups and for another thing the items are not
exactly a part of the survey. The items in the tables are based on the qualitative analy-
sis of the survey and the interviews which figured out the most critical aspects in web
applications today.
3.1 Technical Aspects
The area of technical aspects includes all critical aspects based on technical restric-
tions, conditions or implementations: e.g. used techniques (e.g AJAX, JavaScript),
programming styles and restrictions in hard- and software because of assistive tech-
nologies (AT). Examples for these are Captchas, insufficient operability of flash-
players or missing semantics and markups in web forms. Web programmers, service
providers and producers of utilities and AT are responsible for these aspects [13].
Guidelines for these groups are e.g. the documents from the W3C WAI: WCAG,
UAAG and ATAG and evaluation tools like validators.
The following table shows the most important technical based accessibility issues of
the different user groups in connection with the corresponding WCAG Checkpoints.
All in all it is remarkable that technical barriers have a big influence on the operability
of the applications with the used AT. So it is conspicuous that the user groups which
4 http://www.w3.org/WAI/guid-tech.html
are reliant on AT recognized most barriers. In some cases, these barriers hinder the
independent use of the application by the affected user groups.
table 3. The Most Important Technical Based Accessibility Issues for the Different
User Groups in Connection with the Corresponding WCAG Checkpoints
Type of Disability Technical Aspects WCAG Check-
points
Operability, 2
Semantics of web forms & buttons, 1.3, 4.1
General Error messages, 3.1, 3.3
Semantics of media content, 1.3, 4.1
Operable & available player 1.1
Forms in PDF, 3.3
Captchas, 1.1
Visual Impairments Operable forms & editors, 2.1
Operability with AT (e.g. JavaScript,
2.1, 2.4, 4.1
flash, AJAX) & without mouse
Hearing Impairments Download & control of podcasts 2.2
Operability of: web forms, 2.1, 3.3, 4.1
buttons, 2.1, 2.4, 4.1
Motor and Dexterity drop-down-menus, 2.4, 4.1
Impairments players, 2.1, 2.2, 2.4, 4.1
activation of links, 2.4, 4.1
Operability without mouse 2.1, 4.1
It is not surprising that visual impaired, blind and physically impaired Internet users
have most problems resulting from technical problems with their AT. Most of the
problems occur during the interaction with interactive elements and forms without a
mouse.
3.1.1 Registration, Forms and Buttons
Forms and corresponding buttons e.g. for user registration or buying processes are a
relevant part of a web page and have to work for all interested users. But there are big
problems with forms and buttons in a technical and in a cognitive way. This is under-
line with the rate of problems in table 2 where e.g. 69% of blind users probably will
have problems with user registration forms. But there are also high problem rates for
this type of user interaction for the other user groups.
Problems in forms resulting from different aspects for different user groups, but the
most important aspect for all user groups is the using language in forms. Users have
to understand what kind of input and in which format it is expected, why this data is
necessary etc. (CP 3.3). Thus, explanations and labels are the most important aspects
(see also par. 3.2). But if the explanation is too long e.g. users with deafness and cog-
nitive impairments don’t understand it [3]. The DRC-study 5 support these results with
significant values e.g. in the group hearing impaired users for “complex terms/ lan-
guage” as a key problem [5].
Furthermore, the access to forms with assistive technologies is the second big prob-
lem. This includes the labels, input elements, selection elements, submit buttons etc.
Elements of forms which are not logically ordered (CP 2.4) or which are not de-
scribed with all necessary mark-up (CP 3.3) can cause problems for users which are
dependent on the keyboard or on the special AT-functions for forms. They can have
problems to reach the actual form element, to identify the label of the form element
and to reach the submit button. Especially the tab order and focus behaviour (CP 2.4
and 3.2) need attention for accessible forms. This fact was also supported by the study
of Lazar et. al. [6], where “poorly designed/unlabeled forms” is the third-highest point
of frustrating points in web pages.
Some participants said in the survey that there are problems with comment functions
when accessing with screen readers. The AT does not recognize the written text in the
forms and they can not check their comments before submitting [3]. In large forms,
e.g. for comments or for text in wikis, problems can occur with the reaction of the
application to user controls. Surprising action is the result. Another problem is the
correct structuring and formatting of text in editor forms with AT. Participants said
that they often do not format their comments or that they ask for help [3, 10].
Moreover the usability of forms in general is an important aspect which has to be well
researched and realized. This includes e.g. the perceptibility of the elements and pos-
sibilities (e.g. in sliders or selection bars; CP 1.1 and 1.4), the logic and programmed
sequences of the elements (especially the keyboard access and focus; CP 2.1 and 2.4)
and the wording (CP 3.3) as well. Because of these several impacts all programmers
and editors have to think very carefully about the realization of forms and buttons to
get them accessible in a perceptible, operable, understandable and robust way.
3.1.2 Operability with Assistive Technologies
The operability of web pages and their functions with assistive technologies is causal
for many problems for users who are dependent on AT. This key area is focus in two
of four principles of the WCAG 2.0: Principle 2 Operable and Principle 4 Robust.
These principles regard to keyboard navigation (CP 2.1), navigation and orientation
(CP 2.4) and compatibility with current and future user agents including assistive
technologies (CP 4.1). The German study find out that users often have problems with
the operability of:
5 There is no direct comparison possible between the statistical results of the German study and
the British study because they have different scopes. But the core results of the key problems
are comparable.
• Forms and buttons,
• Drop-down-menus,
• Players and editors,
• Multimedia components.
The core of all these aspects is the access without a mouse (CP 2.1). This is particu-
larly important for blind and physically impaired users. In the interviews and free text
answers of the study users said that they often can’t use players in multimedia sites
and editors in wikis or weblogs with only a keyboard. So they have no control of mul-
timedia content. These problems are formulated often in the interviews as well as in
the free text answers fields for the applications wikis, weblogs and media sharing
websites [3]. The DRC-study confirmed this with significant values for “incompatibil-
ity between accessibility software and web pages” for blind and partially sighted users
[5]. In the US-Study of Lazar et al. were “conflicts between screen reader and appli-
cation” the second-highest cause for frustration of blind Internet users [6].
3.1.3 Captchas
Captchas (CP 1.1) are a problem for 39% of the questioned blind people and for 5%
of the questioned visual impaired people. 50 persons said that this is the main barrier
for an independent user registration because a screen reader can not recognize these
figures. One way for solving this problem is the possibility to listen to an audio-file
which represents the Captcha but this is not a preferred way for most of blind and
visual impaired people [14]. Some of them use webvisum (www.webvisum.com/)
which is a firefox extension for solving graphical Captchas. The corresponding form
element for the solution has to be reachable too.
3.1.4 Conclusion for Technical Barriers
Summarizing the technical aspects we can see that the main problems are in connec-
tion with AT. If programmers do not use all necessary mark-up, no logical structure
for all elements and no alternative ways for access like audio-Captchas some users
with disabilities will have big problems to interact with the application. Another im-
portant point for programmers and utility producers is the ensurance of full access and
interaction with only a keyboard as input device. Furthermore, developers have to
ensure the accessibility of special features like applets in Java, PDF or flash. For these
objects the WCAG Checkpoints are fully applicable, e.g. keyboard access (CP 2.1),
focus highlighting and order (CP 2.4), alternative text (CP 1.1), scalability (CP 1.4)
etc. Central importance should be admitted also to the accessibility of web forms.
These measures are beneficial to all user groups because the readability, usability and
accessibility of form elements are crucial for the independent participation e.g. in
social networks.
3.2 Editorial Aspects
Editorial and content-related barriers contain insufficient editorial or structural con-
tent preparation for Internet requirements, e.g. difficult language, missing textual
structures or missing semantics of media content [13]. Guidelines for web editors are
e.g. “European standards for making information easy to read and understand” [15].
The table shows the main problems regarding the editorial aspects of a web applica-
tion. This includes the understandability of the content in general but also the under-
standability of all text elements of a webpage because these support the orientation,
the user guidance and the content reception. One of the main interesting points in the
table is the several listed issue of missing or unclear descriptions, semantics and
mark-ups of media content. This means for different user groups that they have no
access to the content of media because there is no alternative text (CP 1.1, 1,2, 1.3).
table 4. Critical Editorial Aspects for Different User Groups in Connection with
the Corresponding WCAG-Checkpoints
Type of Disability Editorial Aspects WCAG
Checkpoints
Understandability, 3
Orientation & clear arrangement, 1.3, 1.4, 3.2
General
Quality, size & contrasts of media content, 1.4
Descriptions of media content (alt text) 1.1, 1.2
Semantics of content, 1.3, 4.1
Descriptions of media content (alt text), 1.1, 1.2
Visual Impairments
Numerous links, 2.4
Names of links 2.4
Hearing Impair- Videos in sign language & with subtitles, 1.2, 1.4
ments
Quality of podcasts 1.4
Cognitive Impair-
ments Understandability 3.1
Motor and Dexterity
Impairments Semantics of content 1.3, 4.1
The most important critical point over all questioned user groups is the language in
the broadest sense. This means that a big part of the Internet users have small or big
problems with understanding the provided information. The problems result primarily
from the use of difficult language and foreign words. This is especially problematic in
explanations and forms and in error messages, because users often don’t understand
what they have to do or what is wrong.
This is amazing and a very important fact. This means, that editors and programmers
have to pay more attention to the wording and explanations. And this is an aspect
which we can not check in an automatic way but only with attention to this. One solu-
tion for this is a very careful text editing and structuring of the web content. Here we
have to pay more attention to the editors and their awareness of accessibility issues.
This important fact is underline with an own principle in the WCAG: Principle 3 Un-
derstandable. It emphasizes the understandability of the information and the operation
of user interfaces by users [4]. This includes in checkpoint 3.1 the content itself with
identifying the language of text and mechanisms for abbreviations and pronuncia-
tions. With checkpoint 3.2 the programmers have to make the web pages predictable
in operation and appearance which includes e.g. a consistent navigation and identifi-
cation. That means that for example all buttons have the same wording for the same
functions. In checkpoint 3.3 the avoiding of errors and mistakes is in the focus. This
includes the wording of labels, error messages and suggestions.
Almost all these criterion can checked automatically whether they exist or not e.g. if
there is a label element or if there is an identifier for the language of the text. But only
humans can check if the wording is understandable and formulate as easy as possible.
Problems with the language is not only critical for interaction but also for the user
guidance in a website with the navigation and for the understandability of the pro-
vided information in general. The British DRC-study underlines these facts with their
results. For all analysed user groups language can be an accessibility problem. For
blind users are “incorrect or non-existent labelling of links, form elements and
frames” key problems. For dyslexic users “complicated language or terminology” is a
key problem. All groups have significant values for “confusing and disorienting navi-
gation mechanisms” where the wording is a key factor for a predictable navigation
[5]. This supports our findings with the problems in orientation and navigation over
all questioned user groups.
The German Study, the British DRC-Study and the study from Lazar et.al. work out
that another main problem are missing or unhelpful alt text for pictures and videos
how it is required by CP 1.1. In all studies this is a focus issue particularly for blind
users [3, 5, 6]. For hearing impaired users are the lack of alternative media for audio-
based information and missing subtitles or captures key problems [3, 5]. A compari-
son of different accessibility studies in McEwan et.al. [16] underlines this by showing
that missing alternative text for images and objects is “the most fundamental accessi-
bility problem in commercial website development”.
In the evaluation of the interviews in connection with the survey it can extracted that
the different user groups with disabilities have different problems with the language.
Visual impaired and blind users have often problems if the wording in the navigation
is not clear and easy and when headlines not describe what the content represent [3,
6]. Users with hearing impairments, deafness and with cognitive impairments have
mostly problems with the understanding of the content especially in long text [3, 5].
Additional they have problems in writing text like in comments or wikis because they
are often not very confident with the written word and that’s why they are scared
about the reaction of other users. This is one reason why these user groups use func-
tions like comments with below average in relation to the other questioned user
groups [3].
For summarizing this point the following list of aspects is helpful to pay attention for
the critical issues concerning the language and understandability of text in websites
[11]:
• Difficult language and foreign words,
• Content in general,
• Alt-Text and descriptions of media content,
• Explanations and agreements,
• Error messages and suggestions,
• Expected inputs,
• Names of links and in navigation.
3.3 Design Aspects
In addition to technical and editorial barriers also design barriers can have a deep im-
pact to the accessibility of websites and applications. These aspects influence the user
guidance and perceptibility of functions and not at least the aesthetic impression to the
user. Design barriers based on inadequate accessible design of user interfaces, e.g.
insufficient contrast, background images or too small font sizes [13]. The following
table relate the most critical design based issues due to the accessibility of the web
content and functionality in connection with the corresponding WCAG Checkpoints.
table 5. Critical Design Related Aspects for Different User Groups and the Corre-
sponding WCAG Checkpoints
Type of Disability Design Aspects WCAG
Checkpoints
Perceptibility, 1
Orientation & clear arrangement, 1.3, 1.4, 3.2
General
Perceptibility of functions, 1.4, 2.4
Quality, size & contrasts of media content 1.4
Quality of pictures, 2.3
Visual Impairments Optimization for certain screen resolution, 1.4
Size of buttons & interactive elements 2.4
Cognitive Impair-
ments Orientation & clear arrangement 1.3, 1.4, 3.2
Motor and Dexterity
Arrangement of links 2.4
Impairments
One of the most important and impressive fact from the study is that almost all Inter-
net users have problems to access and interact with a web application because of less
orientation and arrangement (CP 1.3, 1.4, 3.2). The main problems are poor contrast
and too small font sizes (CP 1.4). The British DRC-study comes to similar results: all
groups except the blind users identify “graphics and text size too small” and
“inappropriate use of colors and poor contrast between content and background” as
key problems [5]. Furthermore, in our study as well as in the DRC-study almost all
user groups outline problems with the layout and orientation in the website: “unclear
and confusing layout of pages” and “confusing and disorienting navigation
mechanisms” are the most referred key problems.
Another important aspect is the perceptibility of functions especially in editors and
players (CP 1.4, 2.4). For almost all applications with media content people formulate
problems due to the non-perceptibility of functions. The impact is clear: if they can’t
detect or separate the functions they can’t use it.
Orientation, clear arrangement (CP 1.3, 1.4, 3.2) and problems with quality, size and
contrast of media content (CP 1.4) can be attributed as well to editorial as to designer
aspects. On the one hand, the design should intend suitable format templates and
place holders and on the other hand the editorial staff has to process contents and me-
dia for the Internet and appropriately integrate it into the format templates. Navigation
and contents must be offered well and identifiably structures (paragraphs, headings)
and with sufficient font size for the orientation and clear arrangement.
3.4 Organizational Aspects
Organizational barriers based on organizational circumstances and a lack of aware-
ness for accessibility issues. Examples are missing budget for videos in sign language
and alternative preparation. Orderers and customers are responsible for the realization
of the critical factors when they have a high awareness for accessibility issues. Actu-
ally there are a lot of policies and laws for public and government websites regarding
the accessibility of these for all users (compare e.g. http://www.w3.org/WAI/Policy/).
The DRC-Study [5] and the MeAC-study from the European Commission [2] contain
a lot of recommendations for website commissioners and organizations. The W3C
resource “Developing a Web Accessibility Business Case for Your Organization”
(http://www.w3.org/WAI/bcase/Overview.html) give a holistic business case for de-
veloping accessible websites and can use e.g. for orderer and customer to convince for
accessibility. It describes the social, technical, financial and legal factors and impacts
of accessibility.
5 Conclusion
Interestingly, the recorded results coincide very well with the four principles that
characterize the WCAG 2.0. So the requests for perceptibility, operability and
understandability have been articulated by the interviewed persons repeatedly. The
principle of robustness is primarily reflected concerning the performance of web
applications accessed with assistive technologies. The results presented here show the
high practical relevance of both: the results of the study and the WCAG 2.0 because
the problems identified in the study reflect barriers which still occur and the WCAG
2.0 documents can provide the answers for their accessible implementation.
The study has identified wiki applications, registrations and other forms and media
applications as the most important applications which are very interesting for people
with disabilities although they are often faced with accessibility problems. The most
important barriers for all user groups are understandability in the broadest sense, the
use of forms and the operability of multimedia components, especially with assistive
technologies.
Technical problems which are for the most part caused by insufficient operability of
the applications with assistive technologies are especially noticed by visually
impaired, blind and physically disabled persons. Hearing impaired and deaf Internet
users particularly encounter problems of understanding due to insufficient or
superficial treatment of content and media in formats they understand, e.g. videos in
sign language or with subtitles. Therefore, primarily organizational and editorial
aspects are perceived by this group. Even users with reading disabilities, such as
dyslexia, as well as with learning and intellectual disabilities are affected by editorial
aspects so that restrictions on account of the linguistic competence are experienced.
Therefore, there seems to be a need for more awareness not only for technical aspects
of accessibility but particularly for editorial aspects. All elements of a web application
such as links, labels, menus and the content itself have to be formulated very carefully
and as easy as possible with different user groups in mind. To check the
understandability of all contents, and in particular of elements for user interactions
such as forms, web applications have to be evaluated by users with a variety of
abilities because these in practice experienced problems will otherwise not be
detected.
Acknowledgement
I give thanks to ”Aktion Mensch“ and to ”Stiftung Digitale Chancen“ for providing
the study data and the opportunity to evaluate it in that detail.
References
[1] Eimeren, Birgit von ; Frees, Beate: Ergebnisse der ARD/ZDFOnlinestudie 2010: Fast 50
Millionen Deutsche sind online – Multimedia für alle? In: Media Perspektiven 7-8 (2010),
334-349. http://www.ard-zdf-onlinestudie.de/
[2] European Commission: Assessment of the Status of eAccessibility in Europe. MeAC –
Measuring Progress of eAccessibility in Europe. Bonn (2007)
[3] Berger, A., Caspers, T., Croll, J., Hofmann, J., Kubicek, H., Peter, U., Ruth-Janneck, D.,
Trump, T.: Web 2.0/Barrierefrei. Eine Studie zur Nutzung von Web2.0 Anwendungen durch
Menschen mit Behinderungen. Aktion Mensch e.V. Bonn. (2010)
[4] Caldwell, B., Cooper, M.: Guarino Reis L., Vanderheiden, G.: Web Content Accessibility
Guidelines (WCAG) 2.0. W3C Recommendation. World Wide Web Consortium (W3C).
December 2008. http://www.w3.org/TR/WCAG20/. (2008)
[5] Disability Rights Commission (DRC): The Web. Access and Inclusion for Disabled People.
London (2004)
[6] Lazar, J.; Allen, A.; Kleinman, J.; Malarkey, C.: What Frustrates Screen Reader Users on
the Web: A Study of 100 Blind Users. International Journal of Human-Computer Interac-
tion. 22:3, pp. 247 - 269 (2007)
[7] Lopes, R.; Gomes, D.; Carriço, L.: Web Not For All: A Large Scale Study of Web
Accessibility. In: Proceedings of the International Cross Disciplinary Conference on Web
Accessibility 2010. pp. 10:1-10:4 (2010)
[8] Sullivan, T. and Matson, R.: Barriers to Use: Usability and Content Accessibility on the
Web's Most Popular Sites. Proceedings of the ACM Conference on Universal Usability, pp.
139-144 (2000)
[9] Sayago, S. and Blat, J.: An Ethnographical Study of the Accessibility Barriers in the
Everyday Interactions of Older People with the Web. Universal Access in the Information
Society. pp. 1-13 (2011)
[10] Cornelssen, I., Schmitz, C.: Vorstellung der Ergebnisse der Studie „Chancen und
Risiken des Internets der Zukunft aus Sicht von Menschen mit Behinderungen“ auf der Ak-
tion Mensch-Fachtagung „Einfach für Alle – Konzepte und Zukunftsbilder für ein Barriere-
freies Internet“. http://www.einfach-fuer-alle.de/studie/. (2008)
[11] Ruth-Janneck, D.: Multidimensionale Klassifizierung von Barrieren in Webanwen-
dungen. Mensch&Computer 2009. pp. 13-22 (2009)
[12] Henry, S. L.: Essential Components of Web Accessibility.
http://www.w3.org/WAI/intro/components.php (2005)
[13] Ruth-Janneck, D.: An Integrative Accessibility Engineering Approach Using Multi-
dimensional Classifications of Barriers in the Web. In: Proceedings of the International
Cross-Disciplinary Conference on Web Accessibility 2011. pp. 10:1-10:4 (2011)
[14] Mayer, T.: 74,75 Punkte für die Facebook-Registrierung. BIK BITV-Test.
http://www.bitvtest.de/infothek/artikel/lesen/facebook-1.html (2011)
[15] Inclusion Europe: Information for all. European standards for making information
easy to read and understand, http://www.inclusion-
europe.org/LLL/documents/Information%20for%20all.pdf. Brussels. (1999)
[16] McEwan, T. and Weerts, B.: ALT Text and Basic Accessibility. Proceedings of the
21st British HCI Group Annual Conference on HCI 2008: People and Computers XXI:
HCI...but not as we know it. pp. 71-74 (2007)