Public Policy Impact on Internet Accessibility for Persons with Disabilities: The Effects of the Federal Regulation Section 508 of the Workforce Investment Act on Non-Governmental Entities



By

Ronald E. Milliman, Ph.D.

Professor of Marketing
Western Kentucky University
(270) 782-9325
E-mail: ron.Milliman@wku.edu




Public Policy Impact on Internet Accessibility for Persons with Disabilities: The Effects of the Federal Regulation Section 508 of the Workforce Investment Act on Non-Governmental Entities

Abstract

This study examines the impact of the W3C and Federal Regulation Section 508, as an expression of public policy, on non-government organizations. A total of 1080 profit-seeking (both B2C and B2B) and not-for-profit firms and organizations were examined in an attempt to uncover the answer to three key questions: 1) What proportion of firms, excluding government entities, has W3C and Section 508 compliant web sites? 2) Of those non-compliant firms, what reasons are given for their non-compliance? 3) What would it take to encourage non-compliant firms to make their web sites compliant? Among the findings, it was uncovered that over 98% of the firms investigated continue to support {inaccessible} web sites, ignoring the W3C standards and US Federal Section 508 regulations and thus, potentially locking out a sizable portion of the market.

Introduction

On June 21st of 2001, US Federal Regulation Section 508 of the Workforce Investment Act became effective, requiring all Federal Government web sites to be accessible to persons with various disabilities. Additionally, several states have adopted similar regulations, e.g. Maine, Minnesota, Missouri, Maryland, New York, Texas, Kentucky, Arkansas, Pennsylvania, Nebraska, Arizona, Oregon, and West Virginia. At this writing, several other states are also in various stages of passing similar legislation. These various government entities have, subsequently, begun including "accessibility" clauses in their contracts, thus, imposing, in effect, similar regulations on all firms and organizations doing business with them.

Over the strong objections of many business proponents and lobbyists arguing that persons with various disabilities are not a part of their target market, the regulations are still being passed. Opponents of the laws claim such legislation is placing an unfair burden on them. Johnson and Moxon States that "many companies, particularly those whose products are marketed across the full spectrum of the population, considered disabled customers as a niche market at best, and an unwanted intrusion at worst" (p.247).

To minimize the risk of being sued over this issue, all firms, large and small, private or public, for-profit or not-for-profit must deal with this issue of web accessibility. In some extreme instances of organizations refusing to comply with the federal statue, lawsuits have already been filed to force compliance. There is an arguable legal precedent for web services to be made accessible to disabled people within the spirit of the Americans with Disabilities Act of 1990 (ADA). AOL, H&R Block, Intuit, Bank of America, Wells Fargo, and the Sydney Organizing Committee for the Olympic Games, all have had legal action brought against them because of inaccessible web sites. Last year, for one example, the National Federation of the Blind filed and then, subsequently dropped a lawsuit against America Online after the company agreed to make version 6 of its software accessible to screen readers. However, more litigation based on the ADA and Section 508 standards is surely inevitable as Federal, State, and local governments and businesses resist and indeed, refuse, to comply with accessibility guidelines and the law.

The question arises: is it really that difficult to make web sites fully accessible? Many web design professionals assert that by making their web sites fully accessible to all persons with various disabilities forces them to use a very limited repertoire of the total array of tools and techniques available such as JAVA, flashing graphics, certain colors, etc. Yet, other designers that embrace the goal of total accessibility claim they can create attractive and very functional web sites that are also fully accessible to persons with disabilities with very little additional effort. For instance, designing compliant web sites requires the use of alternate descriptions for images, and invoking some limitations on the use of tables for layout purposes, i.e. tables can pose a problem for screen-reading software. Properly designed tables provide descriptive information that is otherwise hidden to the visual user. Other coding options allow much easier tabbing through links or navigation without a mouse, as well as the inclusion of full text explanations of acronyms that would otherwise be meaningless when put through a screen reader (Williamson 2001; p9).

Difficulties in designing for all users arise when features on a Web page can only be deciphered by one sense, such as sight or hearing. For example, a graphic element might convey important information. But for a blind person, the content is invisible if the graphic is not coded for accessibility. Screen-reader software allows a blind user to access content conveyed by the graphic. The text will be read out loud or accessed through the user's refreshable Braille display. By coding the graphic for accessibility, the content is revealed to two senses. The same is true for audio content and the hearing impaired. A system that permits captioning is a flexible and inclusive system. People with disabilities will use specialized software to enable their computers to talk to them in ways that suit their needs. There is access software that translates information on a screen into speech or Braille. Captioning for the hearing impaired is created through transcription or webcasts, and other technologies allow folks to manipulate computer commands without using their hands. All these adaptive technologies require friendly computer systems and flexible, accessible, Internet pages to work.

There is, of course, some investment in time required to familiarize oneself with the guidelines for accessible web design and the construction techniques. However, experts agree that it's far more economical and efficient to make an initial investment rather than retrofit a design. If retrofitting is necessary, even that task has been made much easier and much less costly by using a program such as the "A-Prompt Toolkit Version 1 web accessibility validation and repair utility," recently released and down loadable (free of charge) at: http://www.aprompt.ca/.

The fact is that with utilities like A-Prompt, bad web page design is a barrier which can be overcome relatively easily. Most of the leading web design programs (e.g. Dreamweaver, Front Page, and HomeSite) are now including accessibility features as a part of their regular repertoire of tools.

Another excellent service created to assist web designers and web masters to determine their site's level of compliance and what needs to be changed to bring it into compliance is a free service available to all called "Bobby." What is Bobby? Bobby is a tool for Web page authors. It helps them identify needed changes to their pages so users with disabilities can more easily use their Web pages. For example, a blind user will be aided by adding a sound track to a movie, and a written transcript will aid a hard-of-hearing user. Bobby will recommend that these be added if they do not already exist (http://www.bobby.com).

How large is the total disability market, and how many persons, considering various types of disabilities, are potentially effected? The American Association of People with Disabilities estimates that one in five Americans has some type of impairment, making access to technology, including the internet, absolutely vital. It is estimated that there are up to 54 million people in the United States with a disposable income of approximately $1 trillion, and another 750 million persons world-wide, with various disabilities whose full access to the web is limited because of poor, non-compliant web design (NCD, 2001).

According to a Harris poll conducted in 2000, disabled individuals use the Internet twice as much as the mainstream population, and it has made a major difference in their way of living. Half of disabled users surveyed said Internet access has substantially improved their lives while only one-quarter of the non-disabled users reported lifestyle enhancement (Harris, 2000).

While there is considerable variation in how people adapt to changes in their physical abilities such as the loss of their eyesight or hearing, more and more are turning to technology and the internet for solutions (Baker, Stephens, and Hill 2001). Though research shows serious sight loss, for instance, often reduces or even curtails independent mobility and the ability to get the information needed to participate fully in society (RNIB Campaign report 15), the World Wide Web provides opportunities for a disabled person to fulfill all types of otherwise very difficult to satisfy needs when made fully accessible, e.g. distance learning, shopping, and communicating with other people. Further, it can substantially reduce dependence on others and it could give a segment of our population that is often excluded, a voice in the 'information society'.

Therefore, making web sites accessible to blind and partially sighted people, and all disabled people, makes good marketing sense because, as the data shows, collectively, it is a large, fairly captive segment of the total market. By making their products easier to use and their sites easier to navigate, companies cater to all of their customers, not just those with handicaps (W3C/WAI's Education and Outreach Working Group, 2001).

Whether driven by market influences or the belief that web sites and intranets are required to be accessible under Titles II and III of the Americans with Disabilities Act, it would seem reasonable to expect that most firms would be designing their web sites to comply with the guidelines set forth for universal accessibility. So, what is the obstacle standing in the way, keeping businesses and organizations from immediately jumping on board and embracing the concept and application of total web access for all? That is the key question to be examined by this study. If there are so many benefits and so much to be gained by making web sites fully accessible to everyone, then, why isn't total web access being fully adopted? In the past, one major obstacle, and it still exists to a large extent, is the lack of standards. Not only have there been substantial variations in HTML coding and other web design coding tools and programs, but also there has been an absence of standards in browser design, other critical pieces of software, as well as hardware development (Jolley, 2001).

Review of the Related Literature

Since this is not a theoretical or conceptual piece dealing with a more common topic such as, for an example, "strategies for marketing on the internet," or some other heavily published topic, and since the focus of this article is on web accessibility and the extent of compliance to the W3C and Section 508 regulations by non-governmental entities, the number of studies related to this focus is extremely limited. However, two closely related studies are especially germane and reviewed here. The Royal National Institute for the Blind (RNIB) tested the web sites of 17 high street stores and banks against a set of accessibility criteria, in order to establish the accessibility of their sites to blind and partially sighted people. They reported that the results were extremely disappointing. All companies in the study failed to meet the needs of blind and partially sighted people. No company achieved a 100 per cent pass rate against the five criteria employed. Some web sites even failed every aspect of the test (Howell, 2000).

Rowland (2000) reported BOBBY data (version 3.0) gathered on a random sample of 400 prominent colleges, universities, and online learning institutions from all 50 states and the District of Columbia between December 15, 1998 and January 8, 1999. Results indicated that fewer than 1 in 4 post secondary institutions (n = 90; 22%) had front pages (i.e., the institutional home page) that would receive BOBBY approval. In going one level beyond the front page, for the institutions that had accessible front pages, only 3% (n=10) received BOBBY approval. Less than 1% of sites that were accessible one level beyond the front page were accessible at all links two levels from the front page.

Thus, from these two related studies, it is clear that web site accessibility and W3C compliance is severely lacking in these specific sectors, but since both of these studies were conducted prior to the June 21st date when the Section 508 regulations took effect, the question remains "has Section 508 had any effect on web site accessibility?"

The Present Study

Questions to Be Answered

This research will examine the extent of compliance to the W3C Web Accessibility Guidelines and the Federal regulation Section 508 in an attempt to answer the following key questions:

  1. What proportion of firms, excluding government entities, has compliant web sites?
  2. Of those non-compliant firms, what reasons are given for their non-compliance?
  3. Of those non-compliant firms, what reasons are given for their non-compliance?

Purpose of the Study

The overall purpose of this research is to examine how many profit-seeking and non-profit type firms currently comply with the W3C and Section 508 standards for web site accessibility, including both business-to-business (B2B) and business-to-consumer (B2C)target market orientated organizations. There are three major objectives:

  1. To determine what proportion of organizations are supporting inaccessible web sites, and thus, are in a non-compliant condition, reducing their market penetration and making themselves subject to possible legal action.
  2. To determine the primary reasons for firms continuing to support web sites that are not compliant.
  3. To determine what incentives, if any, might be most effective in encouraging non-compliant organizations to create and/or retrofit web sites that are compliant with the W3C World- Wide Consortium standards and Section 508 Federal regulations.

The intentions of this paper are to determine and, if appropriate, delineate the extent of the problem, and to cause awareness. It is not the purpose or intention or objective of this study to present an exhaustive theoretical and scholarly treatise on the theory of persuasion. It is a descriptive study intended to simply uncover the extent of a conjectured problem, known to exist in some segments prior to the enactment of the Federal standards, but no objective evidence has been presented following the enforcement date for the Federal regulations.

Research Design

Type of Study, Definitions, and incentives to ParticipateThis research was executed as an empirical field study. It was intended to be primarily a descriptive, as opposed to an inferential type study. A three-step procedure was used in an attempt to maximize the total number of completed questionnaires returned:

  1. E-mailed letter of introduction and statement of research purpose
  2. Initial e-mailed questionnaire with letter of explanation and brief instructions
  3. Follow-up E-mailed questionnaire with letter of explanation and brief instructions.

The study examined profit seeking compared to not-for-profit organizations based on their legal, tax status as self-reported by each organization. In like manner, whether they were classified as either B2C (Business-to-Consumer) or B2B (Business-to-Business) was determined by their primary mission and market reach, again, as self-reported by each firm. The issue of size is always a difficult one to define, whether by gross sales, number of employees, etc. Given the specific type of study, it was determined that the total number of employees would be the best measure to use to define size. Thus, to obtain a really sharp dichotomy, small was defined as a firm having fewer than 50 employees, and large was defined as an organization having 500 employees or more, both as self-reported by the individual firms investigated. The "Opt-in" list broker from whom the organizations were selected maintained the self-reported data in a database.

As an added incentive to participate, each organization that returned the questionnaire was given a free CD that fully covered the issue of web site accessibility and the related law, and in addition, each firm participating in the survey would also receive a detailed "Web Site Compliance Analysis." That is, each firm that returned a fully completed questionnaire, could have their web site analyzed for its level of compliance to the W3C web accessibility standards and a detailed report returned to them, including what, if anything, they needed to do to make their site fully compliant with Section 508 and to the world-wide guidelines for accessibility.

The Survey Instrument. To answer the previously stated research questions, a multiple-item questionnaire was developed, pre-tested, and subsequently, administered to a sample of web masters/web designers during the Summer of 2001. The questionnaire consisted of a combination of several short, open-end and closed-end items based upon the primary questions investigated by the study as depicted above: what proportion of web sites are in a condition of non-compliance, of those non-compliant organizations, what reasons are given for their non-compliance, what incentives would be needed to move non-compliant organizations over to a compliant status? Because of the sensitive nature, and possible legal implications, of the subject matter, an indirect questioning approach was used, e.g. "Why do you think some marketers and web designers are not following the accessible web design guidelines established by the W3C Consortium", rather than "Why have you not made your web site accessible?" The "Web Page Accessibility Checklist" established by the Federal Government (a href="http://www.usdoj.gov/crt/508/webpage.html") was also used as a guide in the construction of the survey instrument.



The Sample.The questionnaire was sent by e-mail to the web master or web designer designated for the web sites of a sample of organizations taken from each of the categories mentioned herein, i.e. profit-seeking compared to not-for-profit, B2C compared to B2B, small compared to large? A sample was randomly selected from each of these categories. The categories and the related samples were provided by a major e-mailing list broker, specializing in providing opt-in lists to web marketers.

Based on the pre-test and the desire to have the sample accuracy within +-5%, a sample of 360 large and 360 small firms were selected making a total of 720 firms in total for this part of the study. Each of these categories was further broken into B2C and B2B sub-categories with 180 firms in each. A separate selection process was used for choosing the organizations making up the profit-seeking and not-for-profit entities, and a total of 180 organizations comprised each of these categories.

The sampling procedure was set up on a "non-replacement" basis. An additional 50 firms were randomly chosen from each category as replacements for the any of the initial selected firms that were, for whatever reason, unavailable, e.g. out of business, web site down, etc.



Statistical Test Employed. Given the nature of the data, the Chi Square statistical test was used for all tests, and the level of significance was held at 0.05.



Results of the Study

Response rate

A total of 1080 individual organizations were contacted using the three-step procedure outlined above. Of these organizations, a total of 453 participated by returning the completed e-mailed questionnaire for an overall response rate of 41.94%. Specifically, in the final tabulation, 155 large firms participated with 72 of them in the B2C group and 83 in the B2B sub-sample. In the small organization group, a total of 146 firms returned their fully completed questionnaires (69 B2C and 77 B2B). Of the 180 profit-seeking firms that were sent a questionnaire, 63 were returned fully completed, and 89 of the 180 organizations in the not-for-profit group returned their completed questionnaires (See Tables 1 and 2).

What proportion of firms has fully compliant web sites?

To answer this question first, all of the web sites of the firms selected in the initial sample of 1080 were tested using "Bobby." To reiterate, Bobby is a tool for Web page authors. It indicates the level of web site accessibility and helps designers identify needed changes to their pages so users with disabilities can more easily use their Web sites. Of the 1080 firms, only 19 or 1.76% had compliant web sites, meaning 1061 or 98.24% failed the Bobby test (see Table 3). Because of the large number of sites involved and the enormous number of total possible pages to run through the Bobby test, only the first three pages were tested on any given firm's web site. If every page of every site were tested, the percentage of fully compliant sites would have undoubtedly been even less.

To investigate this question more deeply, those firms that participated in the study by returning their fully completed questionnaires were examined more closely. The first analysis was conducted comparing those firms that were compliant versus not compliant. While the significance of this question appears to be intuitively obvious, the Chi Square value is 254.91, DF of 1 which is significant.

Next, the large firms were compared to the small firms to see if one size tended to be significantly more or less compliant than the other. Of the 155 large firms participating in the study, 8 or 5.16% had compliant sites. In contrast, of the 146 small firms, only 4 (2.74%) had compliant web sites (Chi Square = 1.33, DF = 1; NS). So, it makes no difference whether a firm is large or small in terms of their likelihood of being compliant.

The next comparison was an overall test between the B2B and B2C type firms. Within the large firm group, 72 were classified as B2C, and of these 5 tested as compliant, and of the 83 that were classified as B2B, only 3 were accepted as compliant (Chi Square = 0.5, DF = 1; NS). In comparison, within the small firm group, 69 were classified as B2C, and of these 2 tested compliant, and of the 77 that were classified as B2B, the same number, 2, tested compliant (Chi Square = 0.3334, DF = 1; NS) (see Table 4). Therefore, we can conclude that B2C are as likely or unlikely to be compliant as B2B type organizations.

Among the profit-seeking compared to the not-for-profit firms, 2 (out of 63) and 5 (out of 89) were compliant according to Bobby (Chi Square = 1.29, DF = 1; NS) (see Table 5). Thus, it makes no difference whether an organization is a profit-seeking or a not-for-profit type; one is as likely to be non-compliant as the other.



What reasons are given for non-compliance?

A pre-test of this issue using open-ended questions revealed a range of answers that could be focused around the following responses, and these were, then, the choices used on the final questionnaire. Again, the indirect questioning technique was used, i.e. "What do you think the primary reason is for firms having web sites that do not meet the W3C or Section 508 standards for accessibility to those persons with various disabilities?" They were instructed to select the one best answer as they perceived it to be. Here are the possible responses and their results:

Not familiar with the W3C accessibility standards

36

7.93

Some familiarity with the accessibility standards, but not knowledgeable of the exact techniques for making their site fully compliant

79

17.40

Insufficient funds to make their site compliant

58

12.78

Persons with disabilities are not a part of their target audience

191

42.07

They just have not had the time to deal with it

41

09.03

It is simply not a priority with them

44

09.69

Other

5

1.10

An overall test of significance was initially performed yielding a total Chi Square of 332.71, DF = 6, 0.05 cutoff = 12.59; significant! While the responses were scattered across all of the possible reasons for non-compliance, obviously some had a much higher frequency than others. For analytical purposes, the one most dominate response was removed from the analysis (Persons with disabilities are not a part of their target audience) and a test of significance was performed on the remaining six responses yielding a Chi Square of 68.7794, DF = 5, cutoff = 11.07; significant! Further analysis was, then, performed by dropping out the least often identified reason of "Other," and conducting a test of significance among the remaining five responses. This test yielded a Chi Square value of 80.9988, DF = 4, cutoff = 9.49, Significant! The next lowest response was also dropped from the analysis (Not familiar with the W3C accessibility standards) and the remaining four responses were tested for significance: Chi Square = 16.23, DF = 3, cutoff = 7.82; Significant! The next most frequently identified response was dropped (Some familiarity with the accessibility standards, but not knowledgeable of the exact techniques for making their site fully compliant) and the remaining three responses were tested yielding a Chi Square of 3.45, DF = 2, cutoff = 5.99; Significant! Obviously, there is nothing to be gained by any further analysis of this item since dropping the most frequently identified response would leave frequencies of 44 and 41, which are, intuitively, not significant.

Thus, we can conclude that the differences among these responses are clearly significantly different and could not have occurred merely by chance. However, two major reasons that tend to jump out above the others are the belief that while web masters and web designers may have some familiarity with the accessibility standards, they are not sufficiently knowledgeable about the exact techniques for making their site fully compliant, and the reason that really dominates the findings is the belief that persons with disabilities are not a part of their target audience. These two reasons together account for nearly 60% of the explanations given.

Since many web masters and designers were relatively unaware of the details of the accessibility standards, it would seem reasonable to assume they would also be unaware of the real costs involved or time that might be required or the opportunity losses associated with their non-compliance or risks of legal costs in both time and money that could be incurred by their non-compliance.

The "Other" response captured a variety of extraneous answers such as: "... can not figure out why our site is not checking out as being compliant," or "... we are currently working on it but it's not quite ready yet, or "... this is a PC issue, and we do not concern ourselves with politically correct issues."

What Incentive Would Work Best for Achieving Compliance to Accessibility Standards?

Again, from an open-ended pre-test, the range of possible responses fell into the following four answers, which were used, in the final questionnaire. Below are the possible choices and their respective rates of response:

Making non-compliance punishable by a substantial legal penalty

45

9.91

Only when it can be shown that compliance will result in noticeably larger web site traffic

247

54.41

A government matching fund to help offset the costs

96

21.15

More education on "accessibility issues and how to make sites compliant"

66

14.54

Again, an overall test of significance was initially performed yielding a total Chi Square of 220.9427, DF = 3, 0.05 cutoff = 7.82; significant. For analytical purposes, the one most dominate response was removed from the analysis (Only when it can be shown that compliance will result in noticeably larger web site traffic) and a test of significance was performed on the remaining three responses, yielding a Chi Square of 19.0435, DF = 2, cutoff = 5.99; significant! Further analysis was, then, performed by dropping out the least often identified reason of "Making non-compliance punishable by a substantial legal penalty," and conducting a test of significance among the remaining two responses. This test yielded a Chi Square value of 5.5555, DF = 1, cutoff = 3.84, Not Significant! Thus, we can conclude that the differences among these four possible responses are different beyond mere chance.

The one reason that was selected much more often was "Only when it can be shown that compliance will result in noticeably larger web site traffic." Obviously, this cannot be demonstrated unless the site is first made compliant, which creates a bit of a dilemma.

Conclusions

The conclusions are rather clear. To answer the first question posed by this study, that is, what proportion of firms support web sites that fully comply with the W3C World-Wide Consortium standards and Section 508 Federal regulations, the answer is that an alarming proportion do not. Specifically, 98.24% of the web sites examined in this particular study representing all types of firms, large and small, B2B and B2C, and profit-seeking and not-for-profit, failed the Bobby test for web site accessibility. While there appeared to be some variations among firm types, these differences were relatively small. The objective evidence is very plain: there is a long way to go before the web is considered fully accessible to all internet participants, whether they be persons with some form of disability or someone simply trying to use their wireless device.

As for the primary reasons given for the lack of web site accessibility, two were especially prominent: 1) web masters and designers are somewhat familiar with the accessibility standards, but not knowledgeable of the exact techniques for making their site fully compliant (17.40%), and 2) it was felt that persons with disabilities are not a part of their target audience (42.07%). These findings tended to jump out from among the other possible explanations given by the respondents. This would seem to indicate the overall lack of knowledge, knowledge about what is needed to make a web site accessible, and the lack of knowledge about the characteristics of the people who are being alienated by their site remaining inaccessible, but who might possibly be a client or customer if they could access the site. It seems to be a general attitude widely held that persons with disabilities are not a part of a firm's target audience. It is true that all persons with disabilities are certainly not a part of every firm's market, but every firm's target audience probably includes some, if not several, persons with some type of disability. In addition, maintaining an inaccessible web site also excludes most of the increasingly popular wireless and hand held devices. Finally, those firms doing business with various government bodies or desirous of doing business with those government entities, may have no choice; that is, they either make their web sites accessible to persons with disabilities or face legal action from either the government or from individuals or groups representing persons with various disabilities.

The question of what can be done to encourage more firms to create and/or retrofit their web sites to make them fully accessible is a very difficult question to answer, but this research has uncovered a couple of rather interesting responses: 1) web sites will be made compliant only when it can be shown that compliance will result in noticeably larger web site traffic (54.41%), and web sites would be made compliant if a government matching fund to help offset the costs was available (21.15%).

While these two responses were given most frequently, it is especially interesting to note the most often selected choice was to be able to demonstrate how traffic to the site will increase as a result of making the site compliant to the W3C and Section 508 standards. This becomes a kind of "catch 22" situation. Perhaps, the only way to get past this dilemma is to more widely publicize cases that can be documented where site traffic has gone up after making a firm's site compliant and to better educate web masters and web designers about how easily compliance can be achieved using readily available analysis and design tools. The other frequently chosen alternative to this question indicated that web masters and designers would comply if some form of government funding offset the cost. While this may be a very desirable solution from the business's point of view, minimizing its risk and financial exposure, this alternative solution is probably a bit unrealistic, given the many competing sources for the limited government funds.

Managerial Implications

From a managerial point of view, there are many reasons for making a firm's web site compliant with the W3C World Wide Consortium and Section 508 Federal Regulations. As stated earlier in this paper, a surprising percentage of persons have either some type of disability that limits their web access or they are trying to use some type of wireless device, which is text based and thus, cannot access the graphics and other design techniques that make sites inaccessible to those types of devices. Thus, firms may be alienating up to as much as 40% of their potential market by maintaining inaccessible web sites. This study reveals that over 98% of the firms that have web sites fall in this category. Thus, 98% of the firms that have web sites could expand their market penetration by making their web sites W3C and Section 508 compliant and fully accessible!

If managers elect to continue to ignore this issue, they are risking lawsuits, and increasing their risk of financial loss based upon the recent incidents of legal action taken by such groups as the National Federation of the Blind. Such legal entanglements are always expensive, either in direct costs or from a public relations standpoint. Further, any firm that goes counter to the evident public policy is risking creating an adverse relationship with the entity or entities that have set forth that public policy. The damaged public relations and adverse relationship with policy makers could prove to be even more costly, in an intangible sense, than the legal costs if sued and certainly more costly than designing or retrofitting a web site to be accessible.

It is always nice when managers can make decisions that not only help them reach their profitability goals or their organizational goals but that is also the socially responsible thing to do. However, sometimes, managers are faced with decisions that may not contribute directly to their profitability goals but may contribute to their positive public image or yield some other less tangible positive result. In some instances, making your web site fully accessible may not yield substantially to the firm's bottom line, but it will have many other less tangible benefits, and again, it is the socially responsible action to take. It is an accomplishment that can be clearly flagged by placing the "bobby approved" graphic on the front of the firm's site and perhaps, even make note of it in the organization's annual report. In other words, the firm that makes its web site compliant with the W3C guidelines and Section 508 regulations can gain some valuable, positive PR points.

Recommendations for Further Research

There is, of course, room for considerable additional research in this area. For instance, it would be interesting to examine the firms that advertise offline to see if they are any more or less likely to have an accessible web site than those organizations that concentrate their advertising efforts on the web and do no offline advertising. As a subset of this group and in view of the findings of this study pertaining to not-for-profit organizations, it would be interesting to look at the firms that support the public broadcasting system and compare them with organizations that advertise but do not support the public broadcasting system with the thought that support of the Public Broadcasting System might reflect more of a public consciousness or socially responsible mind set.

The question also arises: "Are European based firms or any specific country's firms' sites, for instance, any more or less likely to have accessible web sites than US based firms?" Are multi-national firms any more likely to have accessible web sites? Are certain industries more likely to have accessible site than others?

Then, there is also the issue of a web site being technically accessible, i.e. it passes the Bobby test, but it is not "functionally" accessible in terms of its "usability." Many site are technically "accessible," but they have some many links are so complex that they are not "useable." For an example, many blind persons may consider the Yahoo site as functionally inaccessible because it has on the front page alone over 250 links and is highly graphical, making it extremely difficult for persons with screen reader programs to successfully navigate that site. Thus, there is considerable opportunity for research in the question of what makes a site functionally accessible or functionally inaccessible.

It is quite plain that there is vast room for improvement in web site accessibility. For many firms, it will substantially increase their overall profitability by making their sites accessible. In other cases, by making their sites compliant with the W3C and Section 508 standards, firms will be avoiding possible lawsuits, thus, reducing their risk of loss. Regardless, of the reason, it would seem to be both the politically correct and the socially responsible decision to make!

References

Baker, Stacey Menzel, Debra Lynn Stephens, and Ronald Paul Hill (2001), Marketplace Experiences of Consumers with Visual Impairments: Beyond the Americans with Disabilities Act, Journal of Public Policy & Marketing, Fall(Vol. 20 Issue 2) p215-225

HARRIS POLL #34, July 05, 2000 (http://www.harrisinteractive.com/harris_poll/index.asp?PID=97)

Howell, Julie (2000), Get the message online: Making internet shopping accessible to blind and partially sighted people, Royal National Institute for the Blind (July), Campaign report 15.

Johnson, L. and E. Moxon, (1998). In whose service? Technology, care and disabled people: the case for a disability politics perspective. Disability & Society, 13, 2.

Jolley, William (2001), Information Access: The New Paradigm Through Convergence and Digitisation, Round Table on Information Access for People with Print Disabilities, Brisbane, (April)

NATIONAL COUNCIL ON DISABILITY, NCD #01-326 May 18, 2001 (http://www.ncd.gov)

Rowland, Cyndi (2000), Accessibility of the Internet In Postsecondary Education: Meeting The Challenge, UNIVERSAL WEB ACCESSIBILITY SYMPOSIUM 2000, San Antonio, Texas (October) http://www.webaim.org/articles/whitepaper.htm

W3C/WAI's Education and Outreach Working Group (2001), "Business Benefits of Accessible Web Design", Andrew Arch editor http://www.w3.org/WAI/bcase/benefits.html

US Department of Justice, (2000) Web Page Accessibility Checklist (http://www.usdoj.gov/crt/508/webpage.html)

Williamson, Kirsty, Louise Stockfeld, Steve Wright, Don Schauder, and Amanda Bow (2001), "The Internet for the Blind and Visually-Impaired," Journal of Computer Mediated Communication

World Wide Web Consortium (2000) http://www.w3.org/tr


Table 1

Summary of Response Rates

Total Firms
Large Small Totals
B2C B2B B2C B2B
Mailed 180 180 180 180 720
Completed  72  83  69  77 301

Table 2

Summary of Response Rates

Total Firms
Profit-seeking Not-for-profit Totals
Mailed 180 180 360
Completed  63  89 152