A dissertation submitted in partial fulfilment of the requirements of the Master of Science degree in Information and Library Studies.
The Robert Gordon University.
Aberdeen Business School.
Department of Information Management.
2004
This report outlines research about blind and partially sighted people using electronic information services in real-life situations, and specifically how their experiences might inform the planning of these services in UK public libraries. This research does not aim to evaluate the extent or quality of individual services.
The reason for conducting this research was a perceived lack of evidence from users about the usefulness of real-life implementations of these services, including the interpretation of guidelines and standards.
Based upon a full literature review, a survey was conducted of blind and partially sighted people who have some experience of using electronic information services, suitable for, but not restricted to use in, public libraries. The methodology outlines the creation of two electronic survey formats, designed to be accessible and usable for blind and visually impaired people, and the related issues of delivery for the study sample. The data collected is included, and a full analysis of the findings is presented. Correlation was found with other wider surveys in the specific fields of blind and visually impaired people, and electronic information services.
The report concludes that the visually impaired and especially blind people surveyed feel that libraries are out of touch with their needs relating to electronic information services, that there is correlation between users personal preferences and the usability of services, and that diverse and varying services act as a barrier to use. Evidence was also found to support the idea that computer access technologies aimed at blind and visually impaired people may attract very little use in libraries, without significant investment in staff training.
A case is made for greater investigation of remote access to electronic services that can be accessed by blind people with their own equipment. This is felt to have usability benefits such as users familiarity with their own systems, and the reduction of unnecessary travel. It was also concluded that there is a need for specific consultation with people who are blind, separate from those who are partially sighted, and early involvement of both groups in testing of new services.
Limitations of the conclusions.
Appendix A - Final e-mail questionnaire.
Appendix B - User comments from the survey.
Please note Appendix C does not appears in this html text version, as it was a graphic representation.
Appendix D - Data about use of the web.
Appendix E - Pilot survey and invitation.
The author would like to thank:
All the people who took part with candour, in what is probably not the first or last survey they have had to do. Without their involvement, this research would be nothing.
The RNIB for providing the main source of contacts.
Laura Muir, my supervisor, who responded to my many, often demanding requests for information, and was a good sounding board for my thoughts.
Dan Jellinek at Headstar for permission to base the email survey upon the TEN standard.
David Potts for supplying unpublished data from the 2003 NETbase survey.
My family for bearing with another mildly worrying degree of obsession with a project.
Electronic delivery of information in public libraries has been increasing steadily in recent years, and the pace of change is accelerating reflecting the growing extent of use in society in general. The Internet has become established globally, as arguably the most pervasive information and communication infrastructure ever developed. The web in particular has become a hugely important medium for commerce, broadcasting and publishing, and there is now a national infrastructure of Internet access in public libraries established by the Peoples Network.
Despite the huge potential for developing electronic information services, there is little evidence to suggest that the needs of blind and partially sighted people are widely understood, and this risks excluding access. For example the web protocol HTTP was the development that made Internet use mainstream and as a result established a platform for almost unlimited amounts of information. Ironically for blind people, its popularity was due to its ability to add visual media, which also introduced an explosion of inaccessible content for people with limited sight.
Although guidelines have been developed, these have not always been followed, and even where they have, they have been interpreted in many different ways.
Staff in public libraries are currently in the position where they are having to develop new policies for stock and services in a hybrid world of print and electronic formats, where the quality and format of content, and the method of delivering it can vary enormously depending on the subject area.
Within the specific area of electronic access to information for people who are blind or partially sighted, public library staff have difficult choices to make. A significant proportion may be doing so with very limited or even no direct experience of implementing these services, or of the needs of visually impaired people in using them.
This research sets out with the simple aim of asking blind and visually impaired people what they think about using available electronically delivered information services, of the type currently being introduced in UK public libraries. By gathering feedback directly from real users of real services, it is intended to increase understanding to help inform service development in this area.
Whilst there are useful guidelines and standards to help, these can be confusing without user feedback to give insight into the thinking that went into producing them, and to test that they are effective when put into practise. This is especially important for developing services, where attention to these issues now, is an investment in the future of access for many people.
However for staff trying to develop services, there are several factors that make it unlikely that they have access to sufficient user feedback.
The literature indicates that many visually impaired and especially blind people do not choose to use public libraries because they feel they do not provide for their needs. This is compounded by the fact that the developing electronic information services, including the access technologies provided for visually impaired people, are so new that there may be too few users locally to get sufficient breadth of opinion from feedback.
This research is an attempt to provide information that might go some way to helping public library staff understand what blind and visually impaired users think about typical electronic information services, based upon what they themselves say about accessing them. It aims to collect from a wider pool of user experience than is likely to be available to individual services, and to gather data that is as specific as possible about individual types of services. The questions have where possible been aligned with other existing studies so that data is comparable.
It is not an attempt to assess how good libraries are performing, but to get critical insight that might help.
The research was intended to offer the following benefits:
Knowledge is based upon an ongoing cycle of theory and observation. Within the literature about electronic services for blind and visually impaired people, it was felt that there is considerably more theoretical benefits or claimed potential, than there was corresponding evidence of real impact for the end user, post-implementation.
Within the specific sphere of public libraries, there was even less. What evidence that exists was felt to be dated, due to the rapid pace of development. This research was intended to make a contribution towards redressing the balance.
The research required a qualitative methodology to collect testimonial evidence.
A case study was rejected. This was partly because it would be difficult to achieve impartially, as the sensitivity of the subject might make it politically risky for subjects to participate. It was also likely to have a limited use for a wider audience, because there was expected to be wide variation between individual library authorities services in terms of extent of implementation, degree of development, and corresponding usage by blind and visually impaired people.
It was decided to conduct a survey of individual blind and partially sighted users of common electronic services currently in use in UK public libraries.
To understand what users felt about real services required identifying the ways in which electronic information can be delivered and accessed, that were commonly available at the time of research.
It was not practical to survey the diverse local provision in terms of extent of implementation, or quality. Rather, it was decided to establish a generic list of electronic information services that were in current use in public libraries, or available to them. This would be the core subject of the survey questions.
The survey required a sample of people who were not just blind or partially sighted, but also had used the services being studied. The evidence was that the sample group were not likely to use a public library frequently, if at all. For this reason, it would be counter-productive to require that users experience had been in a public library. The key aim of the research was to gain users opinion of services that were appropriate for the role of public libraries.
Because use of services for the subject group depends on assistive technology, the survey would also need to include questions about this, as well as demographic questions about users circumstances, and level of visual ability.
To summarise, the above points suggested that a suitable methodology was to conduct a survey of blind and visually impaired people from the demographic base of the total UK population, about electronic information services relevant to UK public library services
This research is concerned with use of electronic information services suitable for use in UK public libraries, and is aimed at staff working within them, although its findings may be of use to other audiences. This research was undertaken towards a Masters degree in Information and Library Studies. The author felt strongly that the outcome be of direct practical use in a working environment, whilst acknowledging its limited scale.
The findings were based upon experienced users of these services, and should be treated as only directly representative of such users. However it is hoped that the benefits of their experience will be of benefit to designers, commissioning staff and future blind and visually impaired users of electronic information services.
This literature review is organised into eight parts that were felt to cover the main themes of the research problem in a logical order.
The consensus on their role and purpose, and services that public libraries should be providing.
The nature and availability of services available, to serve the specific service needs of the public library sector, as identified above.
In particular, its effect on peoples information seeking behaviour and approaches to accessing information in general. This includes factors that act as barriers to access for people with sight problems.
A review of what systems exist to assist by people with visual impairment when using electronic information services.
The published advice addressing the needs of visually impaired people that is available to library managers when implementing electronic information services.
How far public libraries practise is in line with the guidance, and the extent of implementation of available technological solutions.
What visually impaired people themselves say about using electronic information sources; their feelings about using assistive technologies for this purpose, and research into user behaviour and viewpoints.
A review of specific factors that affect research the needs of people with visual impairment.
To understand the information needs of their users, it is necessary to review the purpose and scope of public libraries in the present day. In other words what is the remit of public libraries, now and in the immediate future, why they are there, and what is the consensus of their role is.
The users of public libraries are the general public, in other words anyone and everyone. Within the user group, there is a vast range and level of information needs. Information and cultural stimulation may be required on any subject and at almost any level.
Importantly this broad scope applies not just to the information that is provided, but to the way in which it is delivered - the library services themselves. These services need to be usable by people with very different needs, and cater for the range of physical and learning ability of the whole population.
There are legal requirements to enforce this. These result from the Disability Discrimination Act1. Historically the DDA was introduced at just the time when graphic interfaces were established as the dominant means of using a computer, and that have in some respects made them less accessible for people with visual impairment.
The needs of the general population therefore define both whom needs to be provided for, and in what ways services need to be delivered. Electronic government services have been championed by the Labour Government elected in 1997, and they chose libraries unique national network of physical buildings to implement an infrastructure of public access to the Internet.
Tony Blair, in New Britain: My vision of a Young Country 2, stated just as books are available from public libraries, the benefits of the [Internet] must be there for everyone. This vision ultimately led to the Peoples Network3, which when completed in 2002, delivered almost universal broadband Internet access across UK libraries.
In 2001, the Government published the first ever set of library standards4. There were three standards for electronic services. Two of these, access to an online catalogue (PLS5) and the number of electronic workstations (PLS 6) were only a measure of amount of provision. The third referred to visits to a library website (PLS10), and at least measured usage, but none measured impact of services.
The Governments current position on what libraries service outputs should be achieving in terms of purpose and impact, were set out in Framework for the Future.5 Published in 2003, this declares the aim of setting out an agenda for libraries a ten year vision which states: the following areas of activity should be at the heart of libraries modern mission:
Framework reflected a welcome and positive continued Government interest in libraries. However about electronic services, it paints an optimistic picture of the future based upon very limited evidence without critically considering the potentially huge changes that are likely to occur within ten years. It provides mostly conjecture about the role of digital citizenship, and electronic content creation, and makes some rather unsubstantiated claims such as that creating community web pages increases democratic engagement.
There is similarly little specific detail about how to widening accessibility, beyond the odd reference to the need to be focussing particularly on the needs of people who do not currently use libraries.
Although published before Framework for the future, CILIPs report Start with the Child6 has crucial things to say about how services should be run. In particular that individuals nurtured in literacy and information literacy will be better adapted to succeed. Its findings show that children are positive about books and reading and that they identify strongly with technology, viewing ICT as a normal channel of information.
As well as all the general information that libraries provide, and help with accessing and making use of it, they also provide information about their services themselves. Electronic delivery of this, especially the Internet, using has completely changed the way in which this can be delivered.
There are now no local authorities in the UK without a website7, and the web is a primary source of information. These vary considerably, but there are some advanced experiments in web based library services to be found, such as Gateshead Libraries website8. The web as a means of access to general and service information, and direct remote access to electronic services is discussed in detail below.
There is also a considerable amount of on-screen service information offered on library computers, such as terms and conditions of use, and disclaimers. Similarly display screen equipment is used as electronic notice boards in libraries.
No research was found into the implementation of this equipment in libraries or consultation with customers about this use. However from personal observation at libraries and shops using display screen equipment, these frequently feature scrolling or changing visual images. These are both features likely to cause problems for people with sight problems.
To summarise, the type of information that libraries should be delivering is defined by the dreams and aspirations of the general public: What they want to do, and what they need in order to achieve this. This may be culture, learning in the widest sense, or access to information for all. This includes published information and literature in a wide range of formats.
As important are the associated services required to be able to effectively use them. Providing technological support for users of services provided is one of part of library staffs essential modern duties.
Similarly the role of expert assistance in navigating information is just as important as it ever has been, and requires a wider staff knowledge and pro-active attitude, not least because the amount produced is simply too vast to manually classify and so diverse in format.
All of these are traditional missions of libraries, but the steadily growing range of electronic information and services is changing the way they are delivered. As Moyo9 puts it there is:
a demand for a new breed of librarians who understand the entire electronic information scenario: procurement, organization, access and public services. All these come into play when making decisions and choices for meeting users' information needs.
The fact that hard copy printed media are no longer the dominant for in all areas has great potential to improve the accessibility of information for visually impaired people. There is great potential for efficient use of electronic information to widen accessibility, but this is dependent not just upon development of new services, but their commercial success, and availability. Libraries can only offer them if they are available.
The range of available electronic information services The preceding section considered what it is generally agreed that libraries are there for. The scope of this research is to examine those services that use electronic information and related ICT services that support its use. To do this requires a definition:
Throughout this research the term electronic information services is used to denote two things:
These definitions are based upon use by others. Brophy and Craven10 use the term information sources in its widest sense to include not only factual documents, but works of creative imagination...representations of museum or gallery objects and presentations. They go on to point out that these non-factual and non-textual resources present particular problems. Examples of the problems they refer to would be how to make available the information conveyed purely by the spatial arrangement of navigation on a page, or the information within a photographic image, to someone who cannot see.
For the purpose of this research, a problem arises here about the wording of the data collection instrument. If the term information is used without qualification, then the data collected may not be valid, as it is arguably too subjective.
This includes computer files, but excludes digitally stored recordings of audio and video files. They make the distinction that these cannot be converted to and from text for example to output devices such as voice synthesisers
This research largely follows these definitions, but it is not intended to attempt to tie this down too tightly, and indeed this is an elusive goal. While this in practise means the scope is predominantly convertible digital text files text, it is not quite as simple as that. Visual elements of a website give information about navigation, that is either unavailable or difficult to use for visually impaired people
As identified above public libraries provide culture, learning support and information. This includes:
They also require staff skills for:
And they have a community role:
To provide these things using electronic information or electronic methods of delivery requires that these to be available. At the present time, it is clear that they are not uniformly available for all aspects of libraries work. Content is not published in all fields to the same extent, and some fields are notably more advanced than others.
Garrod11 points out reference may be more suitable for use on a computer than fiction, but this may be that the current technology is not usable enough. Developments in e-book readers such as the Sony/Philips Libre12, may change this. More importantly for visually impaired people, research conducted by people such as Helen Petrie13 at City Universitys Centre for Human Interaction into e-book usability and open source ebook standards may have an early influence. It is fair to say however that the potential for e-services remains underdeveloped in many areas.
There can also be a lack of implementation of standards for delivery of electronic information services. This applies between different services that may use widely varying solutions each to a different standard. This isnt helped because although it is not legal under the Disability Discrimination Act1 to provide inaccessible services, there is no obligation for commercial suppliers to make their products accessible.
Delivery mechanisms vary hugely. There are some areas such as e-books where several competing proprietary solutions exist, none of which is established as the market leader. This is exacerbated by the lack of standardisation across platforms.
The ability to implement electronic services that are available is dependant on local conditions such as financial resources, management structure and ICT policy within a local authority. There are many active debates about these issues that can be found on public library discussion lists, such as LIS-PUB-LIS14, PEOPLE-NETWORK15 hosted by JISC.
Providing adequate security on public computers in libraries can lead to restrictions on their usability e.g. customising settings in Windows.
It is obviously essential to review the extent and practical implications of sight problems for people. This section does not attempt to describe eye conditions themselves in any detail. The literature on the subject is extensive, and there are many excellent sources. A good starting point is the RNIB16 website.
This section is an attempt to summarise the relevant findings and views of others about the issues that visually impaired people face when using electronic information services, related to how these either serve their needs or act as barriers.
Definitions of what is meant by blindness and visual impairment are as important as the above definitions of electronic information. Brophy and Craven17 summarise these in detail, and quote the National Assistance Acts definition of blindness as: so blind as to be unable to perform any work for which eyesight is essential, They state that there is no legal definition of partial sight.
Anyone registered blind or partially sighted has rights to services under the 1995 Disability Discrimination Act18 The RNIB website19 states: From 14 April 2003, if you have been certified as blind or partially sighted by a consultant ophthalmologist, or if you are registered as blind or partially sighted with a local authority, you will automatically be regarded as disabled for the purposes of the act, and further goes on to say that even if a persons eyesight is not considered to adversely affect enough to register, minor defects may contribute to being eligible to be considered disabled for the purposes of the law.
For the purposes of information seeking behaviour, the RNIBs advice goes on to interpret the term substantial adverse effect as an Inability to read ordinary newsprint. This is a simple and obviously relevant criterion
In this research, the issue is not about whether someone is technically eligible to be registered, although the legal rights are obviously important and to be welcomed. Rather the spirit is to aim to make services more widely useable by anyone with any eyesight problem, even if slight.
Simpkins20 distinguishes between a medical model and a social model of disability, and argues that much research into disability in the past has been flawed because it is wrongly focussed. She gives the example of a disabled person who is not able to use a bus. In the medical model the problem lies with the persons condition preventing them using the bus, whereas in the social model the problem is that the design of the bus is the reason why the person cannot access it.
In providing electronic services it is easy to find comparable situations where access to a service is difficult, not because a person is blind, but because the design of the service failed to include their needs.
The MLA disability portfolio21 makes the case that the concept of disability is changing. It is likely that most people will be disabled in their lives. A high proportion of people will experience sight problems at some point in their lives. For some people this is a lifelong factor in their lives. Their experience will help the many others of us, who may experience sight loss for a shorter part of our lives.
It is important to acknowledge that visual impairment is a range of ability, not a homogenous group. Westling et al22 in discussion of the relationship between Braille and languages identify two poles in this range, "severe visual impairment" and "low vision". This affects how services should be tailored. What suits a person with some vision may be of no use at all to someone who is totally blind.
The mental process of navigating an environment for a person who is totally blind is affected by the fact that they are unable to scan their surroundings visually. In the physical world, navigating from one place to another means following a set of sequential steps. The significant factor here is that the process is serial. Each step follows the last.
This applies equally to electronic interfaces. Craven and Brophy23 found the parallel design of websites to be a significant barrier to use by visually impaired people. For example they describe a website with over a hundred links organised on a single page, not displayed in any particular order, because there was an assumption that the user would scan it for the one they want.
This presumption of the ability to scan is likely to be because there is an assumption that the user can see, often because the designer puts aesthetic considerations first. There are other studies such as the TOWEL project24 investigate the methods visually impaired people use to navigate.
Studies such as these support the idea that usability is as important as accessibility. Because usability is specific to individual users needs, then it is essential to understand that the ways which people find usable, who are visually impaired and especially blind, are not the same as those of sighted people. This means that getting an interface designed for sighted people, and reading it out is simply not addressing these needs, as Yesiladas study25 of screen readers shows.
Oppenheim and Selby26 list some interesting facts about vision impaired people, such as 1.7 million people are unable to read standard print, 36% of blind people can read large print, only 2% (approximately 19,000) of blind people are fluent Braille readers, and 30% of blind and partially sighted people use audiotape for their information needs.
Bosher et al27 state: Computer-related aids and equipment for people with disabilitiesare not easy to learn or intuitive to use, the barriers facing those who need some extra equipment to access reading material or catalogues are considerable. Libraries besides providing equipment, may also therefore need to re-direct resources to lengthy induction for users to overcome the steeper initial learning curves, that may be faced by visually impaired people.
The ubiquitous web transformed the use of the Internet by providing a graphic interface. Ironically this visual element that has made it so popular is in itself likely to cause barriers for vision impaired people. The TOWEL24 Project at the University of Manchester, looking at the concept of travel says: Visually impaired users find mobility on the Web particularly difficult because of the reliance of hypermedia on visual layout . Where sighted people visualise a map, vision impaired people follow pointers sequentially as a set of instructions.
This something that is also universal in desktop applications like Windows and Mac OS. As far back as 1987 Gralla28 pointed out the potential for graphical interfaces as barriers to vision impaired people. Simple concepts such as icons and desktops are map-like, but if you cant see it the image of a bin is not a helpful shortcut to the deleted items folder
Designers of interfaces introduce barriers to use for blind and visually impaired people if they do not understand their needs. Kaufman-Scarborough29 makes the general point that: Advertising relies on colour to place the message in the right emotional ContextFaced with these colour choices designers get carried away. On the web, this is compounded by animation, pop-up windows and frames.
There are many technological aids that can blind and visually impaired people use. JISCs Techdis database30 claims to contain details of over 2,500 items of assistive technology alone. It is beyond the scope of this research to discuss these in detail, but as Bosher et al27 point out, lo-tech and no-tech solutions should be considered as well as expensive hi-tech solutions.
Craven10 classifies access technologies into three basic types, which are visually enhancing, audiobased, and touch-based. It can be easy to overlook the fact that input as well as output devices need to be accessible. For example having screen magnification for a visually impaired person is unlikely to usable if a keyboard does not have large print keys.
Apart from research such as discussed by Dendrinos32, input for blind people is an area with room for technological development with little evidence to suggest widespread implementation. The 2003 Netbase33 survey lists 15.8% of libraries with audio input technology compared to 88% with audio output. These figures, although not detailed to make firm conclusions, indicate an imbalance between input and output.
For technology to work efficiently requires standards to be in place. There are standards for web design such as the Web Accessibility Initiative (WAI)34, and for electronically enhanced audio such as DAISY35.
An issue for electronic formats is the control of unauthorised copying. The SEDEDOL project36 highlights the fact that whilst electronic editions of documents may be available, they are open to copyright misuse. Producers understandably tend to favour strict control, which can affect the access to information. For example the physical nature of hard copies in effect acts as a protection against copying. Ironically, whilst many sources of information, such as newspapers are commonly born digital, they are published in hard copy.
Screen readers can be used with more than one application. This makes them attractive, but their popularity may risk killing off dedicated audio browsers that read the html code itself. This would seem to be supported by cases such as the audio browser PW WebSpeak, which is no longer supported.
Something that is not unique to VIPs is that users have to learn how to use any solution. It may only be simple to use if there has been some investment by the user in doing. This is likely to be accepted for a frequently used service, but for occasional use this will act as a barrier.
This has been shown to be the case for the wider community for example in the uptake of e-government services, where people consistently prefer to have human contact rather than to take them up as self-service. In a survey about implementing e-government, Holdup37 noted that 90% cited the telephone as preferred means of informing the council over e-mail, even with high technical knowledge amongst the respondents. For example 61% agreeing with the statement I find it easy to use new technology
This section reviews what is available to help guide staff in libraries when managing commissioning, or designing services for visually impaired people, specifically related to electronic information services.
These vary in detail from general help such as MLAs Disability Portfolio21, aimed at public sector managers, to the Tiresias38 website which provides information for professional working in the field of visual disabilities
The best single source for libraries and technology is the NLB39. Their Visugate40 portal collects research on blindness, they have an access technology primer41, and their Manual of Best Practise for Libraries42 is an invaluable resource.
The latter collates a great deal of information and has breadth of coverage with individual chapters written by experts in their field. This includes other guidelines such as the Share the Vision/Library Association guidelines dating from 199643. As might be expected, these are dated in the specific area of electronic information services, with a few limited comments such as that information should be made available via computer link between home and library. Likewise the section about electronic stock formats is largely speculative about likely future developments, including REVEAL44, which is now available on the web, and not including other formats such as DAISY35.
However the chapter by Bosher et al27, about aids and assistive technology and cited above, is excellent, and contains a considerable amount of detailed information on approaches. This gives a thorough overview of the practicalities of using these technologies, from sizes of screens to different types of speech readers.
The chapter by Brophy and Craven45 about designing web pages for accessibility reflects the established standards and guidelines that have developed with the web. There are some good basic accessibility tips, and references to other resources such as the automated checking tool, which can check websites against the Web Content Accessibility Guidelines46.
Although a 2004 Disability Rights Commission report47 suggested that these guidelines were not as good as they might be in practical terms, the reality is that lack of compliance with them causes far more problems for visually impaired people than any limitations that they may contain. They are still crucial for accessibility, and form the basis for standards such as the accessibility of local government websites. The reports finding that only 9% of designers test their websites using disabled people is far more significant.
It is clear that technical solutions often only make things accessible rather than useable. As Craven and Brophy48 discuss, it is not enough to just magnify existing systems or use screen readers to read them out, the designers must understand users habits and needs. Similarly Oppenheim26 and Selby discuss options for making the web accessible such as being consistent, offer error prevention and simple error handling, permitting easy reversal of actions and reducing short-term memory load.
Start With The Child6 makes a significant that services are fragmented, and vary geographically. This is particularly worrying given that it implies that if you are visually impaired and dont happen to be born in the right place, then you will be disadvantaged by the failure of the infrastructure to respond to your needs.
While there is a good deal of guidance for providing services to visually impaired people, there is less information about the extent of implementation of these services. MLAs annual NETbase Survey49 started collecting data about access/assistive technologies in UK public libraries in 2003. Data from the NETbase sample of about half of UK libraries, shows just over three quarters of libraries have screen magnification, and nearly 90% with text to speech software33.
In 2001, Davies, Wisdom and Creaser50 found that of 581 blind or partially sighted people using libraries, 38% used speech software, 23% enlarged the text, 3% used a screen magnifier, 1% used a refreshable Braille display, 3% quoted other and 32% said they used more than one method, e.g. using both speech software and enlarging the display.
These two sources were about the only data the author found, and they are not directly comparable. It is likely that there is much local variation, a point made in the Audit Commission report Building Better Library Services51.
Ashcroft and McIver52 make the point that provision of equipment is often reliant on bidding for grants, and warns there are implications for social exclusion where authorities do not succeed in attracting funds in this way.
All this makes the case that there is much to be learned from further study. McCaskill and Goulding53 make the point that a common way of discriminating against disabled people is making assumptions about their needs without consulting them. The evidence for VIP users experiences of access to electronically delivered information This section looks at the testimonial evidence from visually impaired people themselves, as documented in the literature. The author found that it was easier to find discussion of theoretically guided good practise than evidence of the reality of users. It may be the case that the body of research is in a lag period following the introduction of services, and that many studies are in process, but not published, such as the work by Yesilada54 at the University of Manchester.
If this is the case, given the vast number of web based services already developed and implemented, it would imply that little research into use by visually impaired people has been conducted in the design process of many of these. In 2004 the Disability Rights Commission47 found that only 9% of web designers claimed any sort of expertise, and the same proportion had ever included disabled users in user testing.
There are a small number of excellent studies, notably Craven and Brophys NoVA55 project, already cited extensively here, and these are frequently cited in guidance such as the recent MLA accessibility portfolio21. A key point made here is that accessibility is only one aspect of good design and that usability is as important.
Craven and Brophy55 studied the actions of controlled study samples (both sighted and those with visual impairment), when using of a number of real websites. They recorded not just their comments but also the actual keystrokes taken. Despite only studying 4 websites, the of the recording of the actions is very detailed and they were able to place accompanying user comments in precise context.
The key findings of NoVA were about the critical nature of the design structure of the interface in determining usability of a resource. They found several recurring issues that caused problems for people with limited sight.
Amongst the most significant of these was the widespread extent of parallel display of information and navigation on websites. That is that the information might be displayed anywhere on the screen. This is a fundamental issue as it assumes the ability to scan for relevant information.
People who cannot see must listen serially, in order of occurrence to the information, and this order has a profound impact on usability. They cite examples of web pages where the hyperlinks are literally not in order at all, but randomly scattered. They further found that even if navigation is organised into logical sections, the number and the position of these could still render a page tediously long-winded to use.
They also point out that because serial reading is always more time consuming than visual scanning, blind people tend to use search boxes. The make the observation that simply placing the search box near the top of the page can greatly improve usability.
There also comment on the use of language, about the use of language that Davies et als study of visually impaired users50 is of significance because it targeted public library users, and compared their use with other major source of provision such as the RNIB16, NLB39, Calibre56 and TNLUK57. They found that the age of the respondent was significant.
The biggest finding of relevance here is that the sample was consistently found to be unaware of services provided by public libraries. Respondents tended to have set views about what was meant by library services. This reflected poor marketing and promotion, so that even if services were being improved, a key factor must also be to make an active effort to make people aware of this. For visually impaired people, this requires specifically targeting them as non-users. This point was supported by the fact that several services appeared to have had increases after campaigns.
Others are recognising marketing as a major factor. CILIP advises: "A lot of blind people think libraries are irrelevant to them so you must make sure that they know what is already available. In public libraries, for example, they may not know that music and spoken word are available for loan as well as books and that information can be found, and that staff are actively encouraged to help anyone who needs it. 58. The issue is that services themselves are not enough. The target audience needs to know they exist, and this seeking them out rather than assuming they will know or even bother to find you
From the specific perspective of electronic services, Davies et als study50 is limited. There are several references to electronic formats such as computer disks, but these are not clearly defined. It is probably the case that these services were just not established enough to be widely used. This study preceded this dissertation by only three years, and it is likely that even now usage may not be widespread.
Connells small but useful study59 of 9 users experience of using the well known screen reader JAWS to navigate web pages, found a strong consensus that the Internet had greatly increased the content available, and the speed of access compared to Braille. There was a similar agreement that speech readers were the best way to access the web, because they were generally cheaper, and that listening to speech was much easier and quicker than reading refreshable Braille output.
This was felt to justify the deliver of the survey by e-format only
This section looks at the issues that might impact upon the methodology.
Davies et al50 make a number of important points relevant here about surveying people who are blind or have a visual impairment. Not least of these is the moral issue of identifying people by their conditions and cold calling them.
The most obvious challenge for a study of people who have reduced vision is ensuring that the method used to survey them is accessible. That is can the information be conveyed to them in a way that they are physically able to receive it, for example, by magnification, speech output or Braille. Equally important is whether they are they able to convey their responses back.
Accessibility though obviously important is not sufficient. It is also crucial that is simple to use. Nielson provides many thought-provoking articles on usability relevant to this research problem, notably that the length of a survey will affect the returns and results60. Elsewhere he also discusses the problem of information pollution61 and that people feel to be drowning in too much information with too little prioritisation. This has implication for the chances of any electronic survey getting being able to get through to its intended audience.
Nielson also found that most email people who receive them do not read them thoroughly, but only scan them for items of interest60. Usability is dependent upon to users needs. Craven and Brophy55 comment upon the inability of blind people to visually scan, highlighting the need to provide other means of allowing no non-sighted people to quickly assess any survey.
The order of questions in surveys will have an effect upon the perceived importance of the questions to the user, and must be worded in language that demonstrates either understanding of the issues, or recognition that the organiser is willing to learn.
Connell59, and Davies et al50 place demographic detail at the end of the survey. This is important because if placed at the beginning of a survey it might imply that the surveyor viewed the problem as the persons sight, and not services inability to provide for them, a point made by Simpkins1.
For the reasons outlined in the introduction, the basic methodology chosen was to conduct a survey of blind and visually impaired people, to primarily gain qualitative data from end-users in the form of testimonial evidence about electronic information services, suitable for use in UK public libraries, but not necessarily used there. It was also decided to use the opportunity to consider the issues of consulting blind and visually impaired people.
The literature review identified various factors that guided the approach in terms of what data to collect, the most realistic way to collect sufficient data, and ensuring that the specific needs of the study population were addressed.
To gain a sufficient body of suitable data within the project timescale suggested a questionnaire as the most effective method. It had the benefit of targeting a wide range of people, across the UK, which meant that the data should not be based upon any one demographic area, or biased by a single local level of service provision.
Davis et al50 had used guided interviews in person, and by telephone. Being oral techniques, these were very well suited for blind people. However, project resources were restricted to one individuals time, and this was felt to an unrealistic approach to collecting enough data and introduced an unacceptable risk to the project succeeding.
However the credibility of the research depended critically upon addressing issues for the target population, and reports such as A Closed Book62 and Out of Sight, But Not Out of Mind50 made it clear that blind people would be rightly critical of any method that did not accommodate their requirements to use it.
To address this it was decided to use a questionnaire so long as it was designed and conducted within acknowledged limitations. Firstly it was clearly essential that is was not just accessible, but readily usable by blind and partially sighted people. Craven and Brophys usability findings in NoVA55 about the barriers created by parallel interfaces and the importance of the structure and language of documents were a major influence.
As well as design issues, the means of delivery was also important. A deliberate choice was made to target blind people who had access to electronic information services, by only producing the questionnaire in electronic formats, and similarly mostly by promoting the questionnaire using electronic communication, such as direct email and distribution lists. It was decided only to offer telephone assistance to those if asked.
This decision meant that the data would not represent the overall population of blind or partially sighted people. However, the decision was felt to be justified, because the data required was by definition feedback from real users of electronic services. McCaskill and Goulding53 cite lack of consultation as a form of discrimination. Data collected from people with a body of experience was therefore more likely to contain the required information about actual use, rather than general opinion about the subject from non-users.
Another deliberate design factor for conducting the survey using electronic formats was that they were examples of some of the services being surveyed. Comments about the questionnaire would be given as it was being used.
This added to the other descriptions of services based upon recollection. This was likely to provide data about very specific features, such as whether a text box was pre-filled with text or blank, to complement. If more than one user commented upon one aspect of the questionnaire, this method would allow direct comparison between users, and give clues about factors such as use of different technologies, or use with different visual ability.
Information to be gathered fell naturally into a number of areas. The primary reason for conducting the survey was to collect information about the use of electronic information services provided by public libraries.
It was also important for comparison and evaluation of the sample, to collect personal standard demographic information, and information about visual ability that might effect upon use of electronic services.
It was felt that it would be good practise to further include a short section about the survey itself, to give the respondents a chance to point out any weaknesses in the methodology as users, and help inform future consultation of blind and visually impaired people by others.
There are comparatively few user-studies in this area. NoVA55 was user based, although it used a more rigorous experimental framework involving observation. Although only a few years old, both Out of Sight50, and A Closed Book62 were felt not to cover the range of electronic services available in libraries.
It was felt to be very important to collect data that would be as comparable as possible with data from the limited evidence available.
The questions relating to personal details here were closely based on Out of Sight50 in which the questions were felt to be clear and covered most of what was required with some minor differences.
As the standard survey instrument for public libraries users, the Institute of Public Finances PLUS survey63 was also reviewed. The definitions for ethnicity from this were used.
From other PLUS questions standardised terminology was used with modifications, for example terms for frequency of use were the same, but referred to use of electronic services, rather that frequency of visit to a public library. It should also be noted that the methodology of PLUS is different being a staff assisted form rather than a self-service questionnaire
The 2003 Oxford Institutes Internet user survey OXIS64 was used as the basis for questions about why people use the web. This was because it contained simple broad categories that were easy to understand and was recent, being published during the early stages of the research. This was felt to make it more comparable. The web was the only service for which users were asked to say what they used it for, as it was judged to be the most general, and commonly used service.
The use of lengthy descriptions was kept to a minimum. However ensuring that users knew what was meant was more difficult for some services than others. For a commonly encountered service such as using the web, a simple title was almost sufficient as people are familiar with the terminology. For uncommon or unfamiliar service, such as virtual enquiry services, a more detailed description of the service was felt necessary.
Trying to create a comprehensive list representative of all electronic services that could be encountered, could only practically be achieved by producing a set of generic types of service to keep the number of questions manageable, to avoid an unworkably long questionnaire. It was therefore necessary to ask people questions about generic types of services such as email. This allowed a general comparison of patterns of usage across all libraries.
A draft list of electronic services was produced. This was comprised of known services that are in use public libraries in the UK. This was posted onto JISC electronic discussion lists14, 15 used by professional public library staff. Some amendments and additions arose from this, and the list was refined.
Even a list of generic types produced over twenty questions. It was decided to include as many questions as possible, but this required balancing usability against comprehensive coverage as several of the sources referred to above made the point that reading using speech readers takes longer than by eye.
The questions were designed to allow the type of scanning referred to in guidelines such as the RNIBs See It Right65 scheme and the WAI guidelines46. These suggest using consistent heading structures so that the user could listen to the heading to see if they needed to answer the particular question, and move on if they did not. This was justified because it was unlikely that each individual user would have used all of the services listed, and as long as the questions had clear headings, they could be skipped if irrelevant.
To further ease the job for users, for each service three questions were asked using a common recurring format. Users were asked if they had ever used the service in a public library, followed by a multiple choice question about how usable each service was, with regard to the individuals visual. This was complemented with a question asking if they used a similar service elsewhere that was better, and what it was that made it so. This was felt to be very important because of the evidence of low use of libraries such as Davis et al50.
It was important to ask about technologies used in libraries to access electronic services. It was again not considered to be practical to attempt to list all possible types. The TechDis30 website lists over 2500. The questions were based instead on Cravens three standard types of visual, audio, and tactile aids31, although this was expanded to differentiate between input and output methods. Respondents were also asked what technologies they used to complete the survey itself.
Apart from standard demographic information it was also clearly required to get detail about the particular visual ability, causes of reduced ability to see, and related data. A question about what users thought about electronic services was also included to gauge the effect of positive or negative attitudes.
It was felt to be essential to check that the very act of surveying the sample was a barrier in itself. Users were asked if the survey was easy to use and understand. They were also asked if the questionnaire caused their specific access technologies any problems.
The brief for the survey format design was to produce an electronic questionnaire that was usable and accessible.
The question format outlined above had been designed with blind people in mind, not to require any visual information to use it, beyond text, with a serial structure designed for listening to, rather than viewing. This made the job of designing the format fairly straightforward matter. It meant ensuring that it was clear for visually impaired people, many of whom can read, as pointed out in guidance such as the RNIBs See it Right65 scheme.
Also as the RNIB website16 points out, a wide range of visual abilities was likely to be encountered, so it needed to be readily adjustable by the user to suit. It also needed to be in a standard file format usable by any technology.
Initially it was decided to produce a web questionnaire in standard html. However, it was decided to offer more the survey in plain text email too, to give a choice of formats.
This was also for usability reasons. Most guidance suggests that blind people are unlikely to have up to date technology and for this using only a web format might exclude people who do not have easy access to the Internet for prolonged periods. Also a web form is required to be undertaken by the user in one sitting, and cannot be left and returned to. An email form could be done offline at leisure, and this was felt to be better for any users who found reading or concentrating difficult for more than short periods.
The simplest way to create a web questionnaire is to use a specialist web hosting survey service. These allow one to create ones own surveys, using standard templates. The advantage of these is that they require no coding knowledge and are cheap and easy top use. They also include built in automated data analysis tools.
After looking at a small number of these, it quickly became apparent that these services were not suitable. All the services looked at failed the WC3 accessibility guidelines46 in several significant ways such as producing non-standard coding, or using client-side scripts, and clearly were not designed with visually impaired peoples needs in mind.
These guidelines indicate that web pages are likely to be accessible to the widest audience if they use standard web coding with content separated from style. It was decided to create the web version by writing all the coding from scratch, using a specialist server-side script service was used to process the output from the forms.
It was felt that the questionnaire was required to be accessible to the highest standard to ensure it did not invalidate its own purpose by excluding any users. The web form was written in strict xhtml 1.0 with formatting in CSS2, written to WCAG AAA standard. Using this method and not including any graphics, also produces very small file sizes to reduce download time. Again this was important for people with lower specification dial-up Internet connections.
This was carried out using the following validation tools:
Bobby is much more comprehensive that WAVE, but requires a fairly considerable amount of interpretation. Bobby reports will indicate how to correct some errors, and includes example of html code for addressing individual points. However the author found these examples frequently gave code that was not strict html, and certainly not strict xhtml. There was some further investigation required to ensure xhtml compliance.
Unlike web standards, there was found to be very little comparable guidance for email. ASCII plain text was chosen as the widest standard, and does not depend on proprietary software to read it, but no was discovered.
The closest to a general accessible email standard was the TEN standard70 developed by Headstar, for their free e-Access Bulletin71, published with RNIB. Although this standard was designed specifically for newsletters rather than surveys, it was felt to have value as it gives guidance on using plain text to in defined ways, for example to identify headers and sections with characters.
Comments from the piloting of the survey form were fed back to Headstar, and were reflected by changes made in a revised version, such as changing the headings from all capitals, as these were not liked. Testing with access technologies. During design, both web and email survey forms were tested using Dolphin SuperNova in document reader mode, to see what they sounded like, and how long it took to listen to the form. This helped to reduce confusion caused by the speech readers interpretation of document text. Characters were sometimes not correct in the context they were intended. For example a dash () was read out as minus.
It also became clear that punctuation needed to be much stricter for use with a screen reader. For example failure to use a full stop at the end of a heading meant the heading and the following sentence was read as one passage without a pause.
PW WebSpeak, a speech browser, was also used to test the web form. Although now discontinued, it was still possibly in use by some people. It worked well with the form, and was felt to be easier than using a standard browser with and added screen reader.
The participants required to undertake this survey were required to fit the following criteria:
There are ethical and practical problems associated with trying to directly target people who are blind or have partial sight. As Davies et al50 mention, as a social group, blind people could be considered to be vulnerable. Furthermore the information that would allow selection is by its nature confidential, and therefore protected from public view.
Fortunately a major source of potential participants was established early on in the project, in the form of the RNIB research departments database of people who have volunteered to take part in research. In response to a request for people matching the research criteria, a list of 257 people was received. As part of the decision contact people by electronic means, the number with email addresses listed reduced this to 151.
As these were volunteers, a fairly high return was expected, likely to provide most if not all that would be required in terms of numbers of returns. A further advantage of this source was that these people was not associated with libraries as such, and likely to be more representative of the general population of blind and visually impaired people than if found by sampling existing library users. Again as discussed earlier, surveys such as A Closed Book62 made it clear library use was not high amongst the target population.
This list provided a degree of project security. However, selecting a sample from a single organisation could introduce bias towards the mission of that organisation, and exclude people who do not feel it represents them. It was therefore necessary to find other sources.
The second largest source of potential subjects was created by searching the web for organisations run by or helping blind and visually impaired people, including libraries that had support-groups. A further 80 or so email contacts were found using this method.
Other sources are listed in more detail in the results under the summary of response rates.
The two versions of the survey form were sent out to a number of people who had various degrees of visual impairment, until at least four responses had been received for each format. Overall most people seemed to find the pilot questionnaires easy to use and understand, although a number of points were identified that were fed back into the design of the survey forms.
One person commented that the email should be in larger print, although as this was possible to adjust this by the user, it was decided not to increase this. Another comment expressed concern that the web survey may not be usable if not using a mouse. This was from someone who occasionally used a mouse. However there was another positive comment from someone who had successfully used the web form without using a mouse. This indicated an individual issue, rather than the form not working.
One person assumed that the survey was not for them, as they dont use the library. To address this some wording was added to the final form, and to accompanying emails to indicate that use in a library was not essential
Interestingly, comments from users themselves seemed to be able to use it, but there were a few comments expressing concern that others might not be able to use it.
There was also some unintended formatting automatically added to the pilot plain text questionnaire, due to copying and pasting into web mail. This was addressed in the final version by careful checking and manual correction.
The pilot survey is contained in Appendix E.
The results listed in this section are listed according to responses to specific questions. Free comments were also invited, and these are listed in categories in appendix C.
Of 22 people who indicated answered this question. Of these 6 (27.3%) people said they had used an electronic catalogue whilst in a public library. 16 (72.7%) people said that they had not used them.
Of these, the ease of use varied considerably:
The following comments about how the service could be made better were given:
22 people answered this question. Of these, 8 people (36.3%) said they had used a catalogue remotely. 14 (63.6%) said that they had not used them. In contrast to use in a library, those who said they had used a catalogue remotely could do so with ease. This reflects the fact they would all be using their own equipment at home.
The only comment given about improving this service was:
There were two other comments by respondents that said they preferred web-based services.
24 people answered this question. Of these, 3 people (12.5%)said they had used this service. 21 (87.5%) said they had not used this service.
Of these:
There was only one comment about how to improve this service:
23 people answered this question. Of these, 7 (30.4%) said they had used this service, and 16 (69.6%) said they had not. Of those who had:
The following comments were made about improving this service:
24 people answered this question. Of these, one person (4.2%) said they had used the service. 23 people (95.8%) said they had not.
One person indicated that they required occasional help because it was not well suited to their needs.
The following comments were made about improving this service, both referring to remote access:
23 people answered this question. Of these, 3 people (13.0%) said they had used this service. 20 people (87.0%) said they had not. Of the 3 people who had used this service:
The following comments were made about improving this service:
As predicted, use of the web was the service that had the greatest number of respondents indicating they had used it in a public library.
23 people answered this question. Of these, 9 (39.1%) people said they had used or tried to use the service. 14 (60.9%) people said they had not.
People who had used the service gave the following responses to ease of use:
The following comments were made about improving this service:
This question was the only one where respondents were asked about how they used a service. It was included because the web was felt to be likely to be the only one that would have a high usage amongst most people. This proved to be the case as all people who responded in detail had used it either at home or at work.
The questions are based upon the 2003 OXIS survey.
23 people answered this question, of which all 23 (100%) said yes.
22 people answered this question. Of these, 14 people (63.6%) said yes, and 8 people (36.4%) said no.
22 people answered this question. Of these, 16 people (72.7%) said yes. 6 people (27.3%) said no.
22 people answered this question. Of these, 14 people (63.6%) said yes. 8 people (36.4%) said no.
20 people answered this question. Of these, 7 people (35.0%) said yes. 13 people (65.0%) said no.
The following two questions varied from the OXIS survey, which did not differentiate between study towards a qualification, and personal study.
22 people answered this question. Of these, 8 people (36.4%) said yes. 14 people (63.6%) said no.
22 people answered this question. Of these, 13 people (59.1%) said yes. 9 people (40.9%) said no.
22 people answered either one or other of the preceding two questions. Of these, 16 people (72.7%) said yes to EITHER question about studying. 6 people (27.3%) answered no to BOTH questions.
22 people answered this question. Of these, 3 people (13.6%) said yes. 19 people (86.4%) said no.
22 people answered this question. Of these, 11 people (50%) said yes. 11 people (50%) said no.
23 people answered this question. Of these, 19 (83.6%) said yes. 4 people (17.4%) said no.
22 people answered this question. Of these, 9 people (40.9%) said yes. 13 people (59.1%) said no.
22 people answered this question. Of these, 6 people (27.2%) said yes. 16 said no (63.6%).
The answers to this question were slightly complicated because 2 people who said no also said they had tried, but indicated that they found the websites inaccessible to use. In contrast, 4 of the people who said yes were totally blind. This may be down to different banks having different services.
21 people answered this question. Of these, 10 people (47.6%) said yes. 11 people (52.4%) said no.
22 People answered this question. Of these, 4 people (18.2%) said yes. 18 people (81.8%) said no.
23 people answered this question. Of these, 18 (78.3%) said yes. 5 people (21.7%) said no.
24 people answered this question. Of these, 8 people (33.3%) said yes. 16 people (66.7%) said no. Of people who had used this service:
The following comments were made about improving this service:
23 people answered this question. Of these, 4 people (17.4%) said yes. 19 people (82.6%) said no. Of people who had used this service:
The following comments were made about improving this service:
23 people answered this question. Of these, 4 people (17.4%) said yes. 19 people (82.6%) said no. All 4 people who had used the service said they had used one or other Microsoft Office product, including Word, Excel, Access. No other software was mentioned by name, but the generic terms Internet, word processing, and spreadsheets were used. Of people who had used this service:
The following comments were made about improving this service:
24 people answered this question. Of these, one person said yes (4.2%). 23 people (95.8%) said no. Of the people who said no:
22 people answered this question. Of these, 21 people (95.5%) said no, and one person said yes. Of the people who said no, one person indicated that this was because it was not available.
One person said that they required occasional help because it was not well suited for their needs
The following comment was made about improving this service:
23 people answered this question. Of these, all 23 people (100%) said they had not used this service.
The following comments were made about improving this service:
The lack of use of these services was expected for this question.
This was described as meaning scanning for any reason, and included scanning to input text for use with a screen reader, or Kurtweil reader.
23 people answered this question. Of these, 4 people (17.4%) said yes. 19 people (82.6%) said no.
Of the people who said no, one person indicated that this was because the service was not available. Of the people who had used this service:
The following comments were made about improving this service:
23 people answered this question. Of these, 6 people (26.1 %) said they had used or tried to use this service. 17 people (73.9%) said no. Of the people who had used this service:
The answers to this question were expected to say this service was easy to use by those who had tried, because it was most likely to be used from peoples own equipment at home. For the one person who could not use it, this was because the library provided a webform, rather than email.
The following comment were made about improving this service:
23 people answered this question. Of these, 7 people (30.4%) said yes. 16 people (69.6%) said no. Of the people who had used this service:
There were no comments made about improving this service.
22 people answered this question. Of these, 8 people (26.3%) said yes. 14 people (63.6%) said no. This question was slightly different in format to others, in that ease of use was not asked for. Instead respondents were asked how helpful staff assistance was. The following comments were made about staff helpfulness:
The following comments were made about improving staff helpfulness:
23 people answered this question. Of these, one person (4.4%) said yes. 22 people (95.6%) said no.
One comment was made about improving this service, which was that is was available elsewhere at the RNIB.
23 people answered this question. Of these, all 23 people (100%) said no.
No comments were made about improving this service.
23 people answered this question. Of these, one person (4.4%) said yes. 22 people (95.6%) said no.
The following comments were made about improving this service:
16 people answered this question. Of these, 2 people (12.5%) said yes. 14 people (87.5%) said no.
Of the people who had used other services:
No comments were made about improving these services.
15 people answered this question. Of these, 4 people (26.7%) said yes. 11 people (73.3%) said no. Of the people who had used this type of service:
2 People said they could use it unassisted, and could adjust it themselves to suit their needs. These services were described as:
One person said they could use it unassisted, once it had been set up to suit their needs. This service was described as:
One person said they required frequent help because it was not well suited for their needs. This service was described as:
The following comment was made about improving this service:
14 people answered this question. Of these, one person (7.2%) said yes. 13 people (92.8%) said no.
One person used this type of service, but did not indicate what system it was or how easy to use it was.
The following comments were made about improving this service:
20 people answered this question. Of these, 4 people (20%) said yes. 16 people (80%) said no. Of the people who had used this type of service:
2 people said they could use it unassisted, and could adjust it themselves to suit their needs.These services were:
One person did not comment on the ease of use of the service they use. This service was a Kurzweil scanner.
One person said they required frequent help because it was not well suited for their needs. This service was Dolphin Supernova.
The following comments were made about improving this service:
20 people answered this question. Of these, one person (5%) said yes. 19 people (95%) said no.
The one person who had used this type of service may have misunderstood the question, as they indicated they had used Dolphin SuperNova, and preferred Outspoken for Mac. Both of these are speech output devices.
The following comments were made about improving this service:
19 people answered this question. Of these, all 19 people (100%) said no.
No comments were made about improving this service.
18 people answered this question. Of these, all 18 people (100%) said no.
No comments were made about improving this service.
The survey was distributed in a number of staggered waves for reasons discussed in the methodology, mostly by email. Where contacts were known these were sent directly to individuals, and in these cases the responses can be measured directly.
Invitations were also sent to organisations, which it was hoped would distribute further. There were also postings to email distribution lists, blogs and a mention on national radio. In these cases, it was not possible to measure response rates specifically, because there was no way of telling for sure where respondents had learned of the survey.
The first invitations were sent to volunteers whose details were provided by the RNIB. Of 150 individual emails sent, only 102 were successfully delivered. 46 were returned as failed due to not being recognised as valid or current, and two were returned because the mailbox was full.
Of the 102 people who were successfully contacted, 18 people undertook the survey, a response rate of just under 18%. These 17 respondents made up 67% of all responses to the survey.
The fact that 30% of volunteers supplied email addresses that could not be contacted was not followed up, and may simply be down to the currency of the RNIB research database. It is also possible however that those volunteers who had email, did not use it frequently. While this cannot be said to represent the general population, further study of the use of email as a means of communication might be worthwhile when contacting visually blind and visually impaired people.
The next set of invitations was sent to 54 email addresses published on the web as involved in work with blind and visually impaired people. Some of these were to individuals, and some were to generic email addresses.
There were only 2 identifiable responses from these invitations
A further set of 21 invitations was made to top up the final numbers of responses. This yielded one definite response.
As well as targeting people individually by email, the survey was posted onto a number of electronic distribution lists, which were JISCs Peoples Network List14 and Peter Scotts Library Blog72 published by Xrefer used by public library staff, and BECTAs On the Nail newsletter, that accompanies the Help Is At Hand73 website disability magazine. The survey was also featured on Radio 4s popular In Touch74 programme.
7 responses could not be attributed to direct email. It is not possible to say if these were received as a result of these promotional methods, or whether they were achieved though word of mouth through others, but it is clear that direct contact by email was the most successful method.
The total number of number of surveys returned was 27. Of these 22 (81.5%) used either the web or the email form with just their usual access technology settings. Of these exactly half used each method. One person (3.7%) used the email version with manual changes they made themselves. To do this, they used the email form but first pasted into Word to enlarge the text to 16 point.
3 (11.1%) people requested further help from the author before they undertook the survey. Of these, one person required a large print version. This was supplied as a 16 point large print rich text format to use in Word. The other two people requested the survey be undertaken by telephone.
For this the author used the questions from the survey with as little prompting as possible, to try to ensure that the data was comparable to that returned by others.
Any comments that were raised by these two respondents that were not based directly on the questions were included as free comments. These were treated in the same way as comments arising from others from the further comments form, or from email communications.
There was one person who did not complete the survey, but who responded with specific comments about it. This was the email version in the TEN format, and they commented that they felt it was too confusing to use with the system of symbols used to mark up headings.
The data for what access technology was being used came from a direct question on the email version, although not all people gave this information. This question was not included on the webform version, although most supplied this information on request. Further information was also retrieved from users of the webform, which was set up to capture the details of the browser that had been used to access it.
There were 24 people who used the survey without direct assistance. Of these, 18 people gave details about the technology they were using when actually undertaking the survey. Those used ranged considerably. There were also comments about other systems that were used in some situations, even though they were not being used at the time.
The most commonly used system was JAWS, with 10 people using it. However of people using JAWS there were four different release versions being used, ranging from version 4.02 to 5.0. One person also said they were using JAWS in Form Mode. This was using the webform. Another said that they find using JAWS to fill in forms on the web quite hard! I much prefer email. Several JAWS users did not specify which version they were using, and one person said they also used HAL, a component of Supernova.
2 people were using Dolphin Supernova, one using version 4.5, and one using version 5.0 professional. One person said they had more or less lost the ability to use the magnification part.
One person was using Windows Eyes, version 4.1.
One was using LookOUT screen reader version 4.14.
One person was using TalkBack, described as: a screen reader for passages which have been highlighted.
Two people specifically mentioned that using the webform with its enlarged text size and contrast was sufficient. However one of these said that they were using Apple Macs CloseView to invert text to compose a follow up email.
Two people completed the email form by transferring to Word and enlarging the font size, and one other person mentioned that they use any combination of enlargement that they can
One person reported using Zoomtext, although they were not using it to do the survey.
One person was using Braillenote BT, described as: a sort of specially adapted Notebook PC which has a 32 cell Braille display and uses Keysoft 5.0 operating system.
Two people were using Focus Braille display. One was a Focus 44, the other not specified
One person was using VideoLight, a CCTV for use with a standard TV set, with a screen reader (TalkBack as above)
One person also reported that they use an oversized monitor.
One person reported that use a Kurtzweil reader, but that they were not using it to complete the survey.
There were a few comments about not making assumptions about what people would be using. These were summed up by one respondent:
Many users will always be out of date with computers and access software, so never assume when designing web sites, that they all have the latest software to be able to read content like flash or suchlike. Most do not use special browsers, especially if on library equipment, just using IE on a standard system with jaws or supernova. At home its the cost of being up to date that is the killer.
Most of the access technologies used were systems that work with other standard equipment and software, such as operating systems, browsers, and the like. As with the access technology, the standard technology also ranged widely.
Information about the operating system was available for 15 people. Of these 8 people were using Windows 98, 3 people were using Windows XP, 2 people person were using Windows 2000, one person Windows Me, and one person Mac OS9.
The most important point here is that most were using an old version of Windows (Windows 98), which was superceded by Windows 2000, 4 years before this survey was conducted. At the time of the survey the most recent version of Windows was Windows XP. Given that operating systems usually come installed onto computers when bought, this suggests a significant number of users may have used old equipment.
Information available about browsers was available for 12 people. By contrast to operating systems, browsers are available for free download over the web, and this is reflected by the data. At the time of the survey, the most recent version of Internet Explorer was version 6. Used by the majority of people for which this was known.
Of these, 8 were using Internet Explorer 6.0, 2 people were using Internet Explorer 5.5, and one person was using Internet Explorer 5.16 on a Mac.
While use of services provided by public libraries was low, and although not always in the answers to specific questions, the free comments (Appendix C) show that respondents were aware of recent advances in technologies.
There were examples of this from specialist organisations. These included the NLB, whose free access to e-ref such as Xreferplus and Know UK was mentioned, as was their specially adapted version of WhichBook, as was REVEALWEB. The RNIB was also mentioned as a source of information in various contexts, as was the TNAUK.
In general terms there were a number of comments about using the web, or trying to. Mention was also made of using banking services, satellite positioning systems, talking household appliances, and
21 people answered a specific question about whether they thought computers and electronic information services in general, make things easier or harder to use for people who have a visual impairment.
Most people gave a positive response:
11 people (52.4%) said that electronic information services have made things easier to use or more accessible.
6 people said that electronic information services tend to make things more difficult to use, or less accessible
2 people said that electronic information services tend to make things more difficult to use, or less accessible.
One person did choose an option, but made the following comment:
SOME THINGS EASIER - COMPUTERS EASY - for example trekker BPS [a satelitte navigation system], and colour reader. However [music] synthesisers use text menus - on screen with no way of getting this to an audio output.
Of 23 people who indicated whether they had a computer at home, all 23 (100%) said yes.
Of 23 people who indicated whether they had Internet access at home, 21 people (91.3%) said yes. 2 people (8.7%) said no.
Of 23 people who indicated whether they normally use a mouse, 10 people (39.1%) said yes. All of these people had some visual ability, although one person said they only used a mouse if they had to.
13 people (60.9%) said no. All of these people described themselves as totally blind or having very poor vision.
Of 23 people who indicated how frequently they used electronic information services 22 people (95.7%) indicated more than once a week. This was the highest category available, and 5 people (21.7%) commented further that they use them daily.
One person (4.3%) said they used electronic information services about once a fortnight.
Of 27 respondents, 18 were male, and 9 female. This gives a ratio of 67% men to 33% women.
4 people declined to give their age. The average age of the remaining 23 participants was just under 54 years old, with ages ranging from 31 to 80.
The spread was fairly even between these two, with 2 people in their thirties, 5 in their forties, 10 in their fifties, 4 in their sixties, one in their seventies, and one in their eighties.
23 people gave a geographic location, and these were dispersed widely across the UK. 19 were fairly evenly dispersed across England, with the North, Midlands, South and West as far as Bath being represented.
All places listed had only one representative except Reading, Brighton and Watford, who all had two. There were three people from Scotland, and one from Northern Ireland, but none from Wales.
23 people indicated their ethnicity. Of these, 22 people (95.6%) selected White British, except one who selected Asian or Asian British Indian.
The same 23 indicated a preferred language. Of these English was the choice of 21, people, with one person indicating English/Welsh, and one Indicating Gujerati/English. Of those who indicated English, one indicated Received English pronunciation, and one indicated English, excluding gobbledegoop and political correctness
23 people indicated whether they were in employment. Of these 13 people (56.5%) were not in work, and 10 people were (43.5%), with one of these being self-employed.
23 people indicated whether they were studying. Of these 4 were in formal education. Of these 3 were part-time and one full-time. Two of the part-time students were also working. One other person indicated that they were studying, but were not in formal education.
10 people out of 23 were neither working, nor studying. These people were all in the upper age range 54 to 80, apart from one person aged 36.
There were 3 questions about respondents visual ability. These asked for their own description of their ability to see, the cause of their reduced vision, and the length of time they had experienced reduced vision.
A further question asked if they were registered blind or partially sighted. This was not intended to give information about their ability to see, but rather to give a means by which the sample might be identified more closely with existing statistics.
23 people gave specific details of their ability to see. 18 respondents were permanently unable to see in any distinguishing way, with 5 people who could see enough to distinguish text and objects with difficulty.
Of these 11 people said they were totally blind with no light perception.
6 people indicated that their eyesight was severely limited to the extent that they were unable to distinguish between light and dark, or only able to make out large contrasting objects.
Two of these people also indicated that their sight varied so that they were not continuously able to see to this extent.
24 people indicated whether they were registered as blind or partially sighted. Of these 19 were registered blind, and 4 registered partially sighted.
23 people indicated the length of time they had experienced reduced vision. This varied from 3 years to 64 years. Most had experienced reduced vision for a significant length of time. For 18 people this was 20 years or more.
23 people gave the cause of their visual impairment. These included some of the common causes amongst the general population.
The other 15 gave a number of reasons listed here:
A number of questions were asked about conditions other than reduced vision that might require adaptations to services. This was to identify, as far as possible, whether it was peoples visual ability alone that was not matched by services, or whether it was a complex of needs that was not being met.
23 people indicated whether they had any mobility difficulties. Of these 7 people (30.4%) said they did.
Although this question was not intended to include mobility due to reduced vision, answers to other questions mention that just getting to a library can be difficult, so it is possible that this question was interpreted as meaning this.
3 people indicated a cause. One person gave the cause as multiple sclerosis. The other 2 people both indicated cases specifically related to vision. One person said that visual disturbances make it hard to walk straight.
Another person commented that:
Total blindness impacts effective mobility, through lack of information and use of equipment through unsuitable interfaces.
22 people indicated whether they had a physical difficulty that affect using equipment. Of these, 3 said yes and 18 said no. Of the 3 people who said yes, the causes given were amputation of an arm, Thoracic Inlet Syndrome, causing loss of use of an arm, and Multiple sclerosis, causing difficulty in using a hand.
One person interpreted this question as meaning that their physical mobility was made d