(Condensed Report)
By
Jean-Marie D' Amour
Project Co-ordinator
Adaptive Technologies
Quebec Foundation for the Blind
and Rehabilitation Counsellor for Visual Disabilities
Nazareth & Louis-Braille Institute
December 2nd, 2003
Study translated by
Catherine Roy
Advisor
AccessibilitéWeb
The ability to browse the Web is a condition for social participation in today's world. The digital divide largely excludes society's most vulnerable populations, whether for economic reasons or simply due to functional limitations that have not been taken into account by the masters of these technologies. Exclusion based on a functional limitation strikes persons with disabilities in particular, but also affects to different degrees a big proportion of the ageing population.
And yet, the Web allows persons with disabilities to overcome obstacles with regard to access to information or accessibility of the environment (for example, in the case of on-line transactions). Thus, despite its enormous potential as a tool for social participation, the Web is sowed with obstacles making it difficult or impossible to use.
During the summer of 2003, the Quebec Foundation for the Blind took the initiative of conducting an evaluation of the level of accessibility of 200 French-language Web sites in Quebec and Canada . In total, 800 pages were evaluated to provide indicators concerning the level of accessibility of the Web. Through this initiative, we wish to create awareness among decision-makers and Web designers concerning the importance and the urgency to improve the accessibility of Web sites for persons with functional limitations, whatever their nature and their severity. We call on all stakeholders, the Canadian and Quebec governments, corporations, private and public organisations as well as the community of persons with disabilities to contribute to this objective.
Two important partners collaborated on this project:
For a list of persons having contributed to this study, please consult Appendix 1.
Return to the beginning of the page
A deliberate sample of 200 French-language Web sites was selected with the objective to establish a representative list of today's Web by identifying important or significant sites in different areas of activity. Three groups and seventeen subgroups were defined:
For every site, four pages were chosen according to the following criteria:
The objective of this selection was not to establish a representative sample of the contents (i.e. information) of the site, but to verify if accessibility guidelines were known and applied. That is why we tried to evaluate at least one data table and one form from each site, elements that represent particular difficulties specifically addressed by accessibility guidelines.
The study is based on version 1.0 of the Web Content Accessibility Guidelines - WCAG (1999) . These guidelines, defined by the Web Accessibility Initiative (WAI) , an international program of the World Wide Web Consortium (W3C) , are widely recognised and implemented by the international community. Although the evaluation was conducted by the Quebec Foundation for the Blind , a non-profit service organisation for persons with a visual disability, the accessibility guidelines address the needs of all types of functional limitations and the evaluation respected this orientation.
The WAI classifies the guidelines into a set of priority levels in accordance with their impact on accessibility. The priorities are as follows:
Our study used Priority 1 and 2 level checkpoints, which " must " or " should " be applied by all Web sites. Priority 3 level checkpoints, being of a more "optional" nature, were not considered within the framework of this evaluation.
Version 2.0 of WCAG, still in the works at the time of this study, classifies guidelines according to four criteria:
We thus took into account classification by priorities as well as classification by criteria.
Two semi-automatic evaluation tools were used:
To verify the application of Priority 1 and 2 level guidelines, we also used a questionnaire of 59 questions (26 for priority 1 and 33 for priority 2), developed specifically for this study and which can be found at the following address: http://www.aveugles.org/accesweb/questionnaire.php
DISCLAIMER: Any methodology has its limits. Our evaluation is one of conformity with standards based on the application of technical criteria; a practical evaluation would probably bring some nuances. During the next evaluation, if we have the means, it would therefore be interesting to complete this technical evaluation with a practical one, taking further into account the usability dimension.
In spite of this methodological limit, we believe that the overall profile as well as subgroup profiles accurately portray the level of accessibility of French-language Web sites in Quebec and Canada . Considering the limited number of pages audited per site, it would be unreasonable to attempt to establish through this overview, a precise profile of the particular strengths and weaknesses of each site.
Return to the beginning of the page
The results are compiled in various ways:
Non of the subgroups obtain the mention of "very good" or "excellent". Within the 17 subgroups, two receive the "good" mention while 7 deserve the "very poor" mention.
Subgroups with the best results are, by order of rank : services for persons with disabilities, rehabilitation centres and organisations for persons with disabilities.
The Canadian government sub-group ranks 4rth while the Quebec government ranks 10th out of 17.
The three least accessible sub-groups (from best to worst) : sports and leisure, arts and culture, careers and employment. They are proceeded closely by the finance subgroup
| Subgroups | Score on 10 |
|---|---|
| Service for persons with disabilities | 5,8 |
| Rehabilitation Centres | 5,5 |
| Organisations for persons with disabilities | 4,8 |
| Government of Canada | 4,6 |
| Computer and Internet | 4,4 |
| Municipal services | 4,3 |
| Transportation | 3,4 |
| Education | 3,4 |
| Products and services | 3,3 |
| Government of Quebec | 3,2 |
| Finance | 3,1 |
| Current events and media | 2,7 |
| Communication | 2,7 |
| Food | 2,6 |
| Sports and leisure | 2,4 |
| Arts and culture | 2,3 |
| Careers and employment | 1,6 |
Thirty-three sites are ranked as "excellent", "very good" or "good". They are the following:
| Identification of the site | Mention |
|---|---|
| Quebec Foundation for the Blind | Excellent |
| Nazareth & Louis- Braille Institute | Excellent |
| W3Quebec | Excellent |
| Visuaide Inc . | Excellent |
| Public Service Commission of Canada (French version) | Very good |
| Supreme Court of Canada (French version) | Very good |
| Conf. des organismes de personnes handicapées du Québec | Very good |
| Lucie-Bruneau Rehabilitation Center | Very good |
| Government of Canada Site (French version) | Good |
| Regroupement des aveugles et amblyopes du Québec | Good |
| Treasury Board of Canada Secretariat (French version) | Good |
| CAMO for persons with disabilities | Good |
| MIRA Foundation | Good |
| Heritage Canada (French version) | Good |
| Industry Canada (French version) | Good |
| Montreal Transportation Board (French site) | Good |
| Association québécoise des parents d'enfants handicapés visuels | Good |
| Montreal Public Libraries (French site) | Good |
| Avocats Desjardins, Ducharme, Stein & Monast | Good |
| Montreal Association for the Blind (French version) | Good |
| Relations avec les citoyens et l'immigration du Québec | Good |
| Human Resources Development Canada (French version) | Good |
| Service de développement de l'employabilité de la Montérégie | Good |
| Desjardins | Good |
| Ministère du conseil exécutif du Québec | Good |
| Canadian Human Rights Commission (French version) | Good |
| Raymond-Dewar Institute | Good |
| Québec Ombudsman (French version) | Good |
| Sympatico (French version) | Good |
| Canadian National Institute for the Blind (French version) | Good |
| Allo Prof. | Good |
| Montreal Rehabilitation Institute | Good |
| La surdité au Quebec | Good |
Only 16% of the evaluated sites offer a level of accessibility qualified as " good ", " very good " or " excellent ". Consequently, 84% of sites offer a very disappointing level of accessibility (" poor ", " very poor " or " null ").
| Mention | Result |
|---|---|
| Very poor or Null | 39,0 % |
| Poor | 44,5 % |
| Good | 12,5 % |
| Very good | 2,0 % |
| Excellent | 2,0 % |
For Priority 1 , the most frequent errors are:
For Priority 2 , the most frequent errors are:
Finally, although this problem is present in only 45% of the total number of audited pages, it is present in a very high percentage of the pages containing a form:
| Checkpoint | Corresponding pages |
|---|---|
| P1 Perceivable (1.1a, 1.1c and 1.1h) - Images, buttons or image maps without alt text | 77,1 % |
| P1 Robust ( 6.3 ) - Unusable content or feature without Javascript | 72,4 % |
| P1 Understandable ( 4.1 ) - Unidentified changes in language | 41,4 % |
| P1 Operable ( 5.1 ) - Data table without headers | 12,2 % |
| P2 Robust ( 3.2 ) - HTML or CSS coding errors | 94,6 % |
| P2 Operable ( 3.5 ) - Absent or misused headings | 76,9 % |
| P2 Perceivable ( NNG) - Insufficient font sizes | 75,8 % |
| P2 Robust ( 11.2 ) - Deprecated or obsolete coding | 65,9 % |
| P2 Operable (6.4 and U9.3) - Inaccessible scripts with keyboard | 57,8 % |
| P2 Robust 3.4b - Fixed font sizes | 55,1 % |
| P2 Understandable ( 10.1 ) - Links open Popup window without warning | 46,6 % |
| P2 Operable ( 13.3 ) - Absent site map | 46,1 % |
| P2 Perceivable (10.2 and 12.4) - Missing or inexplicitly associated label | 44,9 % |
| P2 U ( 13.1 ) - Links lacking meaning | 37,9 % |
| P2 Perceivable ( 2.2 ) - Insufficient contrast | 24,9 % |
| P2 Perceivable ( 3.3 ) - Absent CSS style sheet | 17,6 % |
| P2 Robust 3.4a - Use of fixed rather than relative values | 8,1 % |
| P2 Operable ( 12.3 ) - Form fields to be grouped together | 7,2 % |
| P2 Operable (7.1 and 7.3) - Distracting movement | 7,1 % |
| P2 Operable ( 13.2 ) - Insignificant page title | 6,8 % |
Return to the beginning of the page
The plan of action that we propose has two main objectives:
All Web sites are encouraged to apply all accessibility guidelines, beginning with those of priority 1. A systematic and continuous action is the best way to ensure success.
However, it is possible to significantly improve the level of accessibility of Web sites by applying 7 recommendations classified by order of priority. To establish this order, we took into account the frequency of the problem, its impact on navigation for users with a functional limitation and the cost of correction.
To get a better idea of how close or how far we are from our objectives, the following table presents the level of implementation of each recommendation for every group and subgroup.
Briefly, the 7 recommendations are::
| Groups | 1 | 2 | 3 | 4 | 5 | 6 | 7 |
|---|---|---|---|---|---|---|---|
| Government | 7 % | 21 % | 22 % | 16 % | 61 % | 21 % | 4 % |
| Private or public | 25 % | 17 % | 23 % | 21 % | 49 % | 17 % | 5 % |
| Persons with disabilities | 52 % | 45 % | 62 % | 50 % | 73 % | 45 % | 11 % |
| Subgroups | 1 | 2 | 3 | 4 | 5 | 6 | 7 |
|---|---|---|---|---|---|---|---|
| Government of Canada | 8 % | 48 % | 35 % | 31 % | 56 % | 48 % | 11 % |
| Government of Quebec | 7 % | 7 % | 15 % | 9 % | 63 % | 7 % | 1 % |
| Current events and media | 27 % | 5 % | 28 % | 23 % | 52 % | 5 % | 0 % |
| Food | 12 % | 16 % | 6 % | 37 % | 31 % | 16 % | 3 % |
| Arts and culture | 22 % | 12 % | 25 % | 9 % | 47 % | 12 % | 12 % |
| Communication | 21 % | 39 % | 17 % | 25 % | 21 % | 39 % | 0 % |
| Education | 30 % | 25 % | 10 % | 20 % | 65 % | 25 % | 7 % |
| Careers and Employment | 42 % | 17 % | 8 % | 8 % | 17 % | 17 % | 0 % |
| Finance | 39 % | 21 % | 29 % | 29 % | 54 % | 21 % | 0 % |
| Computer and Internet | 25 % | 33 % | 46 % | 46 % | 33 % | 33 % | 17 % |
| Products and services. | 25 % | 17 % | 31 % | 3 % | 47 % | 17 % | 0 % |
| Municipal services | 5 % | 50 % | 30 % | 5 % | 80 % | 50 % | 0 % |
| Sports and leisure | 21 % | 7 % | 21 % | 14 % | 46 % | 7 % | 14 % |
| Transportation | 27 % | 14 % | 20 % | 29 % | 46 % | 14 % | 0 % |
| Organisations for persons with disabilities | 79 % | 50 % | 79 % | 62 % | 71 % | 50 % | 4 % |
| Services for persons with disabilities | 50 % | 50 % | 50 % | 52 % | 65 % | 50 % | 15 % |
| Rehabilitation Centres | 36 % | 36 % | 64 % | 41 % | 83 % | 36 % | 11 % |
Different countries have adopted measures to ensure Web accessibility.
In 1990, the US adopted the A mericans with Disabilities Act (A DA) which prohibits discrimination based on disability and strives to ensure equal opportunities for persons with disabilities in the area of employment as well as access to government services, to public equipment, to commercial installations and to transportation.
By virtue of this law, a private or public company of 15 employees or more must offer reasonable accommodation to every employee with a disability, including access to computers and software.
In December 2000, the US federal government enacted Section 508 of the Rehabilitation Act which allotted six months to all federal government entities to adapt computers, software, telephones, telecommunication equipment, videos and multimedia products and, naturally, Web sites. Section 508 mandates that Priority 1 level guidelines be observed, as well as 5 other Priority 2 and 3 level guidelines.
There is presently no Web accessibility legislation in France .
However, on October 7th 1999 , the French Prime Minister signed a circular "pertaining to public Internet sites of the State" .
Notably, this circular mentions the necessity of ensuring that public sites are accessible to persons with disabilities and, in particular, to those with a visual or auditory limitation.
This circular explicitly refers to WAI's Priority 1 level guidelines.
Following the Descargues Report , the Interministerial Committee for State Reform adopted a resolution on October 12th 2000 , stipulating that all State Web sites must be accessible to the blind, according to specifications defined by the Prime Minister (circular of October 7th, 1999 ).
In May 2000, the Treasury Board of Canada Council approved the Common Look and Feel Standards (CLF) of the Internet . These standards apply to all federal institutions listed in Schedule 1, 1.1 and 2 of the Financial Administration Act and these institutions must conform by December 31st 2002 .
This document refers explicitly to WAI's Priority 1 and 2 level guidelines.
Our study demonstrates however the limits of such a policy.
In 2001, the Ontario government adopted the Ontarians With Disabilities Act . Article 6 of said Act stipulates:
"6. The Government of Ontario shall provide its internet sites in a format that is accessible to persons with disabilities, unless it is not technically feasible to do so."
The Government of Quebec has an Internet Broadcasting Reference Guide which mentions four accessibility checkpoints. Only one of these checkpoints is covered by our recommendations. The document also states:
"On this subject, government sites should conform to accessibility standards established by WAI ( Web Accessibility Initiative ). This organisation is the recognised reference in the field of Web accessibility and proposes minimum standards to ensure that as many people as possible may browse the Web."
Our evaluation demonstrates without a doubt that this intent has had no effect on the poor level of accessibility of Web sites maintained by the Government of Quebec.
We therefore call on the Government of Quebec to include an article in the revision of its disability legislation, compelling the provincial government to provide accessible Web sites for persons with disabilities.
In February 2003, during consultations held by the Quebec Socials Affairs Commission on the revision of Quebec disability legislation (bill 155), CAMO for persons with disabilities submitted the following article for inclusion in the new law:
"The government must ensure that its electronic network is accessible to persons with disabilities. All new electronic content and any update of electronic content produced by and\or for the government of Quebec must conform to international standards developed by the W3C/WAI, specifically, Priority 1 and 2 level checkpoints of the Web Content Accessibility Guidelines.
All existing electronic content produced by the government of Quebec and any update of electronic content supplied by a third party, must conform to Priority 1 and 2 level standards two years following the adoption of the present law.
The government of Quebec will revise used standards every 3 years to take into account updates made to the Web accessibility standards promoted by the W3C/WAI."
Return to the beginning of the page
Our study highlights the degree to which French-language Web sites are behind with regards to accessibility for persons with disabilities. This is partly due to a lack of information and training for technical resources as well as a lack of commitment and leadership at the political level, creating important barriers for the social participation of persons with disabilities, who represent over a million people in Quebec .
The extremely low level of accessibility for most of the 200 evaluated Web sites (84% were ranked as poor, very poor or null) has important repercussions on the social and economic participation of persons with disabilities and the ageing population. It is not merely about obstacles to Web navigation in an optional context. As the sampling demonstrates, it is a question of access to consumerism, work, education and training, civil (civic ?) participation, access to information, etc.
In a society where technologies are quickly becoming the principal, if not unique, means of communication and interaction in several sectors, the consequences of the very poor state of Web accessibility in Quebec cannot be underestimated. While information technologies present considerable potential for inclusion, in terms of access to information and services, the reality is that situations of exclusion are already upon us and, without greater awareness and clear commitments, initiatives striving to eliminate the digital divide will continue to ignore the needs of an important part of society.
"The power of the Web is in its universality. Access by everyone regardless of disability is an essential aspect." Note 2
We call upon decision-makers and Web designers to improve the level of accessibility of this important tool.
We particularly call upon the Canadian and Quebec governments to assume their social responsibility and serve as examples in this field.
Return to the beginning of the page
The author wishes to thank the following people for their contribution to this project: