IT Essay: An analysis of difficulties experienced by visually impaired users on the web and how to d

The growth of the World Wide Web or the web has been phenomenal in the last decade and this medium has come to be widely accepted as well as extensively used all over the world.

A burgeoning amount of information is now available on the web and this information is quickly searchable. Efforts to enhance the utility of the web are ongoing, with changes in technology such as the authoring languages, communications technologies as well as digital hardware having an influence. A significant proportion of the people in all communities around the world are visual impaired. There are about 10 million visual impaired persons in the United States of America and more then 2 million in the United Kingdom alone. It is, therefore, important that efforts be made to make the web more accessible to these visual impaired persons so that they too can enjoy a higher degree of integration with the mainstream society. Laws in many countries now require that the websites that are designed should be accessible to the visual impaired. Depending on their disability, many visual impaired users of the web "hear" rather then "view" web pages through screen readers and have difficulties navigating the web. PHP, MySQL and style sheets using XHTML as well as Java Script are amongst the latest software tools that are used to design websites for the visual impaired and it is, therefore, important for many to get a feel for how these applications are used in the design of websites. This dissertation presents important findings from some of the latest published research and code for a visual impaired friendly website. This should be of interest to those who want to design such websites and who are interested in how the visual impaired interact with the web.

Visual impaired persons are greatly assisted if they can be provided with screen readers or built - in text magnifiers on the web pages which they may be browsing. Screen readers read out aloud text which has been presented on a web page and these may be activated by clicking a selection link on a website for the visual impaired or the screen readers may be used as separate software. Text magnifiers make it possible for a user to adjust the size of the text which has been presented on a web page to suite their liking, a size with which they are most comfortable. Such techniques can assist the moderately visual impaired to interact with the web through a computer. However, severely visual impaired persons or those who are approaching total blindness may need additional assistive features. Screen readers can enable either full reading of a page or simple web interactions. It is important that web pages be designed with accessibility guidelines in mind which meet the most fundamental requirements for assisting a visual impaired person. Visual impaired individuals resort to complex cognitive processes when interacting with the web through assistive audio and such mechanisms are poorly understood at present, requiring further research and feedback from visual impaired users. Screen readers do not present the implicit cues associated with typography or font size that are inherent in a web page. Hence, a web site that is being optimised for the visual impaired should also have a mechanism for feedback or data collection that will assist with the process of further enhancements. Some interesting questions which need to be answered when investigating the use of the web by visual impaired users relate to the objects or "landmarks" which such individuals use in order to navigate on the web and the manner in which such objects are used. Web pages may then be provided with such objects which assist the visual impaired. There is a need to have a mechanism to capture knowledge about what is useful for a large number of visual impaired persons and to transfer such knowledge to a computer database for analysis (Yesilada, 2005, Pp. 16 -21), (Ma, 2003, Pp. 1 - 20), (Annamalai, 2002, Pp. 1 - 15), (Colwell, 2001, Pp. 11 - 38) and (Leahy, 2004, Pp. 2 - 11).

This dissertation attempts to provide some solutions for the previously mentioned problems of having a screen reader, a text - magnifier and a mechanism for the generation of a database for the analysis of survey data collected from visual impaired users.

1.2 Aims and Objectives
The aims and objectives of the dissertation are to provide basic assistive tools and a mechanism for further investigation as well as design experimentation in an accessible website design that caters for the visual impaired. Additional assistive features can be incorporated and experiments in design conducted when the data collected from the users of the website has been analysed. The main objective is to build on the basic assistive technology that has become widely available today and to find out what is pleasing for the visual impaired users. For such an ongoing investigation to be possible, it is important that the website cater to the moderately visual impaired and hence attempts to provide Braille, tactile or complex audio assistance for the severely visual impaired have not been attempted. Trying to find remedies for the blind or the severely visual impaired who may be better catered for by providing additional assistive hardware will distract from the investigations associated with web design for the moderately impaired.

1.3 Purpose and Scope
As has been previously mentioned, the purpose of this project is to further investigate what can be done for enhancing the interaction of the moderately visual impaired by building in the best of the currently available assistive techniques into a website and then trying to experiment with the design in order to gain insights on what can be done to a web page in order to make it more pleasing for the visual impaired. Hence, the dissertation provides a look at the text -to- speech techniques for a web page, techniques for text magnification and techniques for data gathering and analysis. Blind or severely visual impaired users are not included in the scope of the study.

The next chapter of the dissertation takes a look at the nature of visual impairments that are to be found amongst the visual impaired web users.

2. Background and literature research on visual impairment
In order to better understand the nature of the problems that are faced by the visual impaired when they interact with the web and computers, it is appropriate to consider the nature of the different types of eye impairment that commonly occur in a population. This chapter takes a look at the nature of visual impairment and the historical attempts to understand or find solutions to these problems.

2.1 Introduction
There are many diseases of the eye which can result in impaired vision and the nature of the visual impairment can range from blurred vision, halos in vision, blind spots as well as the so called floaters in which the object being viewed appears to "float" or there are specks in the field of vision. Colour blindness, dyslexia as well as total blindness may also be regarded as a type of visual impairment (Medline Plus, 2005, Pp. 1). Although vision problems have been observed in all age groups, such problems are the most prevalent amongst the elderly who are 75 years or above. There is a significant increase in vision impairment with aging and figures indicate that individuals above the age of 50 are very much more likely to start experiencing vision problems (Department of Health, 2000, Pp. 3 - 5). Along with the purely vision related problems, visual impaired persons may also suffer from a combination of vision and cognitive problems such as poor eyesight and a deterioration of the short term memory. The lens of the human eye becomes yellow with aging and more opaque, reducing the ability to discriminate between colours. There is also a shrinking of the pupil with aging and less light enters the eye. Hence, many of the visual impaired may not be able to distinguish between colours and they may also have difficulty reading small text.

This section of the dissertation presents a literature review, definitions and a discussion of visual impairment.

Published literature related to visual impairment indicates that (Harris, 1971, "The Elderly at Home") was amongst the early investigators who first examined the requirements of visual impaired people without any age limit. However, the earlier work of (Gray, 1967, "Mobility and Reading Habits of the Blind.") was amongst the most significant attempts to try and understand the reading habits of the visual impaired. The work of (Hunt, 1978, "The Elderly at Home.") provided a comparison between the needs of the older visual impaired persons and the needs of the older persons in general. However, it was the published work of (Cullinan, 1977, "The Epidemiology of Visual Disability.") which provided information about the nature of visual disabilities which were prevalent in the United Kingdom. This was a local survey and it is not possible to use this work as an indicator of the nature of visual impairment on a global level. The Royal National Institute for the Blind or RNIB in 1981 attempted to conduct a nationwide survey of blind and partially sighted people in the United Kingdom in order to better understand their needs (RNIB, 1981, "Who are Britain's Blind People?"). The RNIB has since then continued to conduct further studies to better understand visual impairment. However, it was (Bruce, 1991, "Blind and partially sighted adults in Britain: The RNIB Needs Survey, Volume 1") who first provided a wealth of information about the blind and partially sighted in Great Britain. Since then, researchers have continued to try and better understand the needs of the visual impaired with the first attempts at enhancement of the electronic media being directed towards enhancing television for the visual impaired (Davis, 1996, "The needs and problems of older people with a visual impairment: historical, current and future perspectives") and (Pettit, 1996, "AUDETEL: Enhancing television for visually impaired people"). The research work on enhancing television for the visual impaired has had some impact on what is preferable for such individuals when they interact with the web. Serious content which was intricate did not appeal to the visual impaired, who also had difficulty reading subtitles and they preferred to interact with television through dialogue and narration. Lengthy gaps in the sound track were confusing for the visual impaired, but despite their disability about 56 % of those surveyed could distinguish outlines close up, with 55 % being able to distinguish between colours. About 85 % of the visual impaired could detect light changes on the screen. Hence, it can be concluded that with the use of contrasting colours, an ability to magnify text on a website and audio feedback as well as the ability to read aloud text there is the potential to make web more accessible for at least 50 % of the visual impaired.

In the relatively recent literature for accessible web design, it has been found that design guidelines for accessible websites and style guides are considered to be useful in the design of websites for the visual impaired. However, many developers do not apply these guidelines and there is a need to enhance the guidelines as well as a requirement for better presentation (Colwell, 2001, Chapters 1 & 2) and (Milne, 2005, Pp. 558 - 562). Hepatic feedback devices offer better possibilities for those users of the web who are rather severely visual impaired (Colwell, 2001, Chapters 6 and 7). Some research tools such as Dante have recently been developed which attempt to assist the visual impaired travel or browse the web by extracting travel objects which are to be found on a web page which they may be viewing. The page is then transformed into a new page which offers better understanding and navigation possibilities for the visual impaired (Yesilada, 2005, Chapter 5) and (Harper, 2001, Pp. 1). Intelligent screen reader systems which "guess" the desires of an impaired user also offer interesting possibilities (Ma, 2003, Pp. 2 - 20), (Hermsdorf, 2000, Pp. 1 - 6) and (Stefano, 2000, Pp. 1). Transformed web pages may then be converted to Voice XML and produce audio outputs (Annamalai, 2002, Pp. 21 - 41). Hence, web page transformation tools which modify the information presented to the visual impaired users of the web and rearrange it in such a manner that the web page is more easily understood by such users offer new possibilities for research.

The next few sections discuss the nature of visual impairment that commonly afflicts the visual impaired.

Some definitions of terms that are important in relation to the understanding of visual impairment are as follows (Leahy, 2004, Pp. 4 - 14) and (Department of Veterans Affairs, 2002, Pp. 3 - 26):

Dyslexia: This is a term which is used to describe a wide range of learning difficulties that vary in degree between different individuals. The deficiencies are of neurological origin and dyslexics are often unable to recognise words, comprehend what they are reading or speak correctly what they hear. Word recognition is a problem and such individuals may have difficulties associated with comprehending web pages.

Blind / Visual Impaired: This dissertation regards as blind, those persons who have severe vision problems and cannot read even with the assistance of magnifiers. Visually impaired persons are considered to be those who have a deficiency in their visual ability that is lower then that of an average healthy person. In practice, there are many different types of blindness such as red - green blindness, light receptive blindness, tunnel vision or having a narrow field of vision. Those suffering from light receptive blindness can distinguish between light and dark objects but have difficulty deciphering shapes. Obviously, visual impairment is one of the causes of dyslexia. Visual impairment may be measured by tests such as visual acuity, visual field, colour vision, and contrast sensitivity testing.

Literacy Problems: These are different from dyslexia or visual impairment and occur because individuals may be attempting to read a language that is not their mother tongue. Literacy problems may be the result of dyslexia, but such problems do not indicate that a person is necessarily a dyslexic.

Colour Contrast for the Visual Impaired: Selecting the right background and foreground colour contrast is important for the visually impaired because this makes dyslexics decipher text better. Some dyslexics suffer from seeing jumping text in which the text appears to move up or down the field of vision and this may be corrected by having the right background colour. The preferred background colour may be different for different dyslexic persons.

Screen Reader: This is software which reads out aloud words on a selected screen which may be from a source file or a web page.

Complexity of English: The complexity of the language presented on a web page may be determined by the Fog index which measures the length of words and sentences in content. Longer words and sentences result in more complex language which is difficult for the visual impaired to understand.

Blind - Browsers: These refer to special browsers for the visual impaired which present text - only depictions of a web page that is more readily understood by the blind.

Epileptic Users: These are those who are prone to have epileptic seizures if they see flickering screens between 2 and 55 Hz.

Visual acuity: Refers to an ability to distinguish details and shapes of objects with good contrast. A visual acuity of 20 / 200 refers to legal blindness.

Visual field: Refers to the entire area which can be seen by a person when the eye is looking straight ahead. A 20 degree restriction in the field of vision is referred to as legal blindness.

The next section presents some types of visual impairments.

Conditions which can result in visual impairments such as a lack of clarity of vision, restricted visual field, inability to differentiate colours, blurred vision, reduced contrast, flashes, jumping text or an inability to track motion are as follows (Medline, 2005, Pp. 1) and (Department of Veterans Affairs, 2002, Pp. 3 - 26):

Presbyopia: This refers to focusing difficulties with close objects and usually occurs as a result of aging.

Cataracts: Cloudiness in the lens of the eye which gradually impairs vision resulting initially in poor low light vision and ultimately deteriorating day time vision. Halos of light and a restricted visual field can also result from cataracts.

Glaucoma: This refers to increased pressure in the eye which may occur suddenly or gradually, resulting in poor night vision, blind spots or a loss of vision.

Diabetic retinopathy: Occurs as a result of complications from diabetes which results in bleeding from the retina, causing blindness.

Macular degeneration: This occurs as a result of aging and common symptoms of this disease are loss of vision, blurred vision, loss of colour discrimination or distorted vision.

Eye infection: Occur as a result of bacterial infections which can cause a temporary vision impairment that can turn into something more damaging if the infection is neglected.

Floaters: Particles in the eye which may be accompanied by flashes of light and a sensation of shade or hanging in the vision field.

Retinal detachment: This is a condition which arises from an acute case of floaters.
Optic neuritis: This refers to an inflammation of the optic nerve which may result in physical pain in the eve and vision problems.

Stroke or TIA: A bursting of an artery in the brain which can cause the vision areas of the brain to function poorly.

Brain tumour: Malignant growth in the brain causing damage to areas in the brain which decipher vision information.

Bleeding into the eye: Bleeding inside the eye possibly as a result of internal eye damage, due to a wound can impair vision can also result in the loss of vision.

Temporal arteritis: This refers to the inflammation of an artery in the brain which supplies the eye with blood. Such an inflammation may cause temporary or permanent damage to various parts of the eye.

Migraine headaches: This condition may or may not necessarily be accompanied with a headache and can be a passing condition. Spots of light, halos, or zigzag patterns often accompany migraines.

The next section discusses some of the adjustments which a visual impaired person may make and some problematic areas.

Positive mental attitude makes a difference in adjusting to visual impairment and this adjustment can be made easier through assistive aids which are provided to the visual impaired. Positive strategies for adjustment include the development of acceptance by others, trust, independence as well as the minimisation and control of the impairment. Negative strategies for adjustments can include denial, shame, helplessness and isolation. The visual impaired may apply a large number of strategies for adjustments including the use of larger print material, the use of internet with assistive aids, the use of Braille or tactile material and the use of bright pleasing colours. Loss of vision often means that dependence on hearing or touch is enhanced and a far acuter sense of hearing is developed. Obviously, the attitude of the community and the availability of assistive aids are important in helping the visual impaired to adjust to their lives. Thus, the work being conducted on developing websites for the visual impaired is important for the community as well as the visual impaired (Manchester Metropolitan University, 2005, "ADAPTING TO VISUAL IMPAIRMENT").

It may be concluded that a large variety of conditions can result in a person becoming visual impaired with a number of different types of visual impairments. However, attempting to present information to the visual impaired in a manner that is most pleasing to them and which is most acceptable is the basis of assisting such people. Having the capacity to view larger fonts, change colours or contrasts and depending on audio as well as having the most relevant information presented in the most acceptable manner are the best ways to assist the visual impaired. Websites for the visual impaired must, therefore, cater to such needs in addition to presenting information in a manner which is most acceptable to the visual impaired users. The next chapter discusses the manner in which websites may be made more accessible for the visual impaired.

This chapter takes a look at some of the more important themes that are to be found in published literature related to website design for the visual impaired users of the web. A number of design guidelines and style guides have been prepared as a result of the work which researchers have been conducting in order to produce better and more accessible web pages and these are discussed in this chapter.

There is a huge amount of published literature on what can be done to improve web access for the visual impaired person and to improve web accessibility in general (Accesify.com, 2005, Pp. 1). It has to be remembered that although web accessibility is about making websites more useable for the impaired, it is also about making the web accessible on all browsers and connections. Hence accessibility is about making the web more accessible for the able as well as the disabled users. In the year 2002, the Code of Practice for Part III of the Disability Discrimination Act of the United Kingdom made accessible websites to be a legal requirement (Webcredible, 2005, Pp.1).It is, therefore, important that web designers now be able to design accessible websites. Websites are also likely to be accessed by text - only Lynx browsers and screen readers for the visual impaired as well as handheld devices or the WebTV. Thus, a website should be future ready and work with all access technologies. It is widely believed that if a case makes it to the court, the W3C accessibility guidelines will be used to determine if a website is sufficiently accessible (W3C, 1999, Pp.1). Both priority 1 and priority 2 guidelines in the W3C set of accessibility guidelines must be satisfied because priority 2 guidelines are the European Union standard. However, it is best if all priority 1, priority 2 and priority 3 guidelines are satisfied for a website design. In order to make a website more accessible and usable by the visual impaired persons, the web pages, images, search engines and the forms should be considered for enhancement. Web users have successfully sued and have been awarded damages for inaccessible websites and examples of such legal action include the action against the Sydney Olympic Organising Committee, Ramada.com and Priceline.com (Webcredible, 2005, Pp. 7).

The next section of this dissertation takes a look at the most important guidelines as well as design checkpoints that must be considered when designing websites for the visual impaired users.

Creating a text - only equivalent is not sufficient to create an accessible website because the text - only version may itself be inaccessible. Alternatives must be provided for images in the form of ALT text, JavaScript using the <noscript> tag, written transcripts for audio and video as well as HTML equivalents for Flash. All information that is available with colour should also be available without colour. Forms and tables are especially important in a website for the visual impaired users because these can really confuse visual impaired users who are attempting to access websites using screen readers (Webcredible, 2005, Pp. 1 - 10) and (Roberts, 2002, Pp. 1 - 7). The forms need to be made accessible by having prompt text that is correctly positioned and assigned to form items. Information that is to be presented in forms needs to be broken down into manageable chunks and the forms need to be broken down using <fieldset> and <legend>. Visually impaired users who use screen readers listen to sub-headings on forms by grouping menus in dropdown boxes. Placeholding text must be placed inside the edit boxes and text should be positioned correctly. The structure and the presentation of the web pages should be completely separated by using cascading style sheets or CSS, not tables to lay out web pages. The website that is designed should be accessible without the use of a mouse and it should be checked with a Lynx browser. Tables on web pages can create a real mess if they are not properly designed when a visual impaired reader attempts to read them with a screen reader. Every table should include a caption and a summary, with the caption that describes the table coming under the opening table tag. The summary for a table guides the non - visual agents about what the table is for and it is usually placed after the &lt;table summary =" tags. The content of the tables is contained between the &lt; table&gt; &lt;/table&gt tags, while rows are placed using the &lt;tr&gt;&lt;/tr&gt tags. A number of cells in rows may be placed using &lt;td&gt;&lt;/td&gt, while header elements are placed using &lt;th&gt;&lt;/th&gt. (Roberts, 2002, Pp. 1).

There are very many articles which present useful tips for the designers of accessible websites (Accesify.com. 2005, "Articles and Tutorials on Web Accessibility"). However, it is still the W3C or the World Wide Web Consortium which has been the most successful in creating the most useful checklists and design guidelines for designers of websites for the visual impaired (W3C, 2005, Pp. 1) and (W3C, 1999, Pp.1). These guidelines are discussed in the next section.

The checklist for web content accessibility which has been prepared by the W3C consortium is a good reference document for those who are designing websites (W3C, 1999, "Checklist of Checkpoints for Web Content Accessibility Guidelines 1.0").  As has been previously mentioned, priority 1 and priority 2 guidelines are something which all developers should definitely adhere to and priority 3 guidelines should also be attempted to be included in the design. The documents on a website should be organised in such a manner that they are capable of being read without style sheets. Style sheet are preferred for controlling layout rather then tables (W3C, 2005, "Recommendations for Tables in HTML"). Websites for the visual impaired are required to clearly identify any changes in the natural language of a document's text as well as text equivalents such as captions. It is necessary that the design of accessible web pages prevents the screen from flickering, so that the epileptic are not caused any problems, while using the simplest and the clearest language to present content. In addition to being provided with redundant text links for each section of the server side - image map, it is also necessary that client - side image maps be provided rather then server - side image maps unless the image is too complex for its regions to be defined by geometric shapes. If frames have been used then they should be provided with a title in order to facilitate frame identification and navigation. Graphs and charts on accessible web pages should be summarised using the longdesc attribute. Alternative content should be provided for scripts, applets, & plug-ins as a contingency against active content not being supported or being inaccessible (W3C, 2001, "Quick Tips to Make Accessible Web Sites").

The W3C guidelines also require that the background and foreground colour combinations provide adequate colour contrast for the visual impaired. Relative, rather then absolute units should be used for mark-up language attribute values. Auto - redirection pages and pop - up windows can interfere with any screen readers that are being used. Hence, it is appropriate to use the server to perform the redirects. Meta data should be used to add semantic information to web pages (W3C, 1999, Pp. 1). The important consideration in designing websites for the visual impaired is that the browsers and screen readers etc should be able to function correctly without confusing the visual impaired user. If information on a website has been presented in PDF format then equivalent information should also be provided in text format. Captions and auditory descriptions of the visual track should be synchronised. Heading elements should be used to convey document structure and this should not be for visual formatting. Dynamic content should be constantly updated.  Blinking and constantly moving content should be avoided unless the user permits this content. Because many visual impaired users rely on the keyboard rather then the mouse which needs visual feedback, therefore, links and controls should be keyboard navigable. The target of each link should be clearly identified. A skip to main content link should be provided on each page. Any site timeouts should be explained to the website users (RNIB, 2005, Pp. 12 - 53) and (RNIB, 2006, Pp.1).

Different groups of designers use guidelines related to web accessibility differently and they have to find a constructive approach for their use which suits them. However, it is important to realise that guidelines are important as are checklists and they do produce improved designs for the visual impaired who will otherwise be seriously disadvantaged. It is best to refer to the W3C guidelines and checklists which are updated when appropriate (Colwell, 2001, Chapter 2). It is also important to have a web page validated using a suitable validation tool in order to ensure that the page does comply with the requirements of the visual impaired and the law (Schmetzke, 2005, "Validation").

The next chapter of the dissertation considers the methodology for the dissertation including the use of surveys which are likely to add to the knowledge about the ease of use of a website which has been designed for the visual impaired.

This chapter of the dissertation presents a discussion of the methodology for research that was used in order to generate information and to analyse this information for use in the design of a visual impaired - friendly website.

The research that has been presented in this dissertation can be divided into two parts. The first part of the research is related to collecting relevant as well as current literature associated with website design for the visual impaired, analysing this literature for existing design guidelines or techniques and succinctly presenting any conclusions. Such literature can be collected for evaluation by querying web search engines such as Google and querying the databases of reputable libraries such as the Library of Congress or British Libraries and selecting material for further study by considering the content or abstracts. A website containing web pages which incorporate these guidelines can then be constructed and further research can be conducted about the content as well as presentation of this website in order to arrive at better conclusions about the requirements for the visual impaired. It has to be remembered that some of the requirements of the visual impaired web users, especially those that are associated with browsing the web are poorly understood and more research consisting of the gathering of user preference data as well as its analysis is required. Different web accessibility concepts can be incorporated into a website and selected visual impaired users of the web can then be invited to present their opinions related to the concepts which have been incorporated into such a website (Colwell, 2001, Chapter 4 "Evaluation of Web Pages by Blind and Partially Sighted People") and (Yesilada, 2005, Pp. 122 - 127).

Those responding to surveys can fill an online form built into a page of the website which has been constructed for this purpose and this data can then be shifted into a SQL database, such as a MySQL database which is connected to the website so that the data may be readily analysed. Alternatively, some users may prefer to respond through the email with answers. However, the letter approach is likely to be more cumbersome because the responses will be required to be shifted into the database manually. It is important to develop a relevant questionnaire for those being surveyed so that relevant and important information which is likely to enhance the knowledge of the requirements of the visual impaired may be collected. This is discussed in the next section.

The questionnaire which is to be used to collect research information can be presented as a web form in which the users can enter their opinion by either clicking on the closest response or by writing a response in a text box of the form. PHP forms can be readily constructed and the PHP code for such forms is widely available, as in (Maketemplet.com, 2005, Pp.1). The W3C accessibility standards are based on the data related to the requirements of the visual impaired users which has already been collected and analysed. An attempt can be made to enhance the knowledge further by including questions related to what assists the visual impaired users in travelling the web. The methodology for storing data submitted through a form into a MySQL database has been presented in (New York PHP, 2005, Pp. 1) and (Leahy, 2004, Pp. 28 - 46). A website which has been developed for survey purposes and which also includes the web form can attempt to present various examples of accessible designs to the users and collect information about what design they found to be the most accessible by them (Colwell, 2001, Chapter 4) and (Colwell, 2001, Appendices).

The next section discusses the design of the questionnaire for a survey of the visual impaired.

The questionnaire which is designed should attempt to gather a number of distinct sets of information for analysis. It is important that data which sheds some light on the nature of visual disability which a user is suffering from, the frequency of web use, the length of time that such a user has been using the web for and the kinds of assistive aids which they have been using should be available. Different features associated with accessible design which have been presented and incorporated into the website which has been designed using elements such as accessible forms, tables and frames etc can be labelled as Example 1, Example 2… etc and the user can be asked about how accessible these examples are for them. The degree of accessibility can be measured to be from very accessible to not very accessible by analysing the information from the web form which the user being surveyed is required to fill (Colwell, 2001, Appendices). Further information may be collected about the ease of mobility which a visual impaired user experiences when interacting with the website (Yesilada, 2005, Pp. 122 - 125). An example questionnaire and a selection of possible responses which may be selected by those who are being surveyed have been presented in (Appendix B). A set of PHP scripts which make it possible for surveys to be generated quickly using forms and which can assist in the analysis of results are presented in (Gregg, 2005, Pp.1). After creating the survey, the survey URL can be embedded into the code for the website being created and the results of the survey can be viewed online.

It should be noted that research into the requirements of the visual impaired users is an ongoing effort and after data has been collected with one set of approaches to web design for the visual impaired, a different set of approaches may be investigated by constructing another website with a different set of elements which can then be gauged using the survey form. Those who are to be surveyed can then be asked to present their opinions about several different sets of approaches.

The next few sections discuss issues related to sending out the form and data analysis.

For a survey that is being conducted by using a form which has been linked to a MySQL or other appropriate database on a website, it is most appropriate for a user who is being surveyed to browse the website which has been prepared and then use the online form to fill in their comments. However, users must know that a new website for the visual impaired users exists and that a survey is required prior to any possible feedback being available.  Emails are, therefore, required to be sent out to those who are required to be surveyed, requesting that they respond to the survey. The emails that are sent should contain the URL of the website as well as the URL of the survey form page. An explanation should be included in the email about the nature of the research into the web accessibility for the visual impaired that is being conducted. Contact email, physical address and telephone numbers should also be included because some of the recipients are likely to check about the survey through a telephone contact. There are many free advertising lists and groups that are available on yahoo groups which may be used or some e-lists can be rented with the surveyor paying a modest fee for the email message being transmitted. It is also possible that organisations working for the blind may have lists for the visual impaired and they can be approached for sending out email requests. One email in two or three weeks for a noble purpose does not constitute spam. The email that is sent should reassure the recipient that their identity and particulars will be kept confidential and that only the overall results of the survey may be published. A follow up email is usually also required urging a recipient who has previously received an invitation to participate so that those who may have postponed their participation may be reminded to add their comments. The number of additions to the web database should be periodically checked to see if sufficient responses have been received. Those who have been sent the email should also be given the option of sending back their results by email if it is more convenient for them to do so due to severe impairment. However, this will mean that the responses will have to be manually added to the web database. It may be desirable to send invitation emails which are restricted to a geographic region only, such as the United Kingdom or Wales or even a locality in order to ascertain the situation of the visual impaired in that locality only (Colwell, 2001, Appendices).
4.5. Analysis
In order to analyse the results from the survey which have been collected on a MySQL database, the database needs to be queried using different SQL queries, depending on what needs to be investigated. Such queries can be incorporated into the PHP scripts for displaying the required information (Li, 2001, Pp. 1), (Vasquez, 2003, Pp. 1), (Gilfillan, 2002, Pp.1) and (Gowans, 2001, Pp1). However, a better way is to run multiple and complex queries on a database of collected data in order to analyse it from several different perspectives by the use of a tool such as a query analyser which can assist in performing analysis much more quickly (VSI Systems, 2005, Pp.1) and (XLINESOFT, 2005, Pp. 1). A survey form and useful code for interaction with the database is presented in (Appendix C,'A').

The user responses to the survey can be evaluated through an analysis of the data stored in the MySQL database. Responses can be analysed in order to generate statistics for the number of respondents who were satisfied or dissatisfied with the presentation of image maps, forms, images, frames, text enlarger etc and their specific responses can be queried to find out what is lacking. The mobility or ease of travel on the web can also be explored in a similar fashion. This approach coupled with the literature search offers the best methodology for an analysis of the types of visual impaired users who are likely to be found in a region of choice, the nature of their computing equipment, types of internet connections and their preferences (Colwell, 2001, Chapter 4).