One Person’s Strategy is Another’s Barrier

Accessibility is very personal – what works for one person may not work for another.  The “one size fits all” approach has been tried and although admirable in its intentions has often proved difficult to implement.

The W3C WCAG (World Wide Web Consortium Web Accessibility Guidelines) v 1.0 tried a technical approach to accessibility by setting out a number of accessibility guidelines, which could be automatically tested by online validators such as Bobby. Whilst this automatic validation can validate the HTML code, many of guidelines require human input and common sense.  For example, whilst an automated accessibility validator can check that an image has an alt text tag, it cannot check that the tag actually makes sense.  People who don’t use images, such as those using mobile technologies or visually impaired people still need to know whether an image is important to the content or not.  An image with an alt tag of “image01.jpg” gives no information to website users, whilst an alt tag of “Photograph of Winston Churchill” would not only aid navigation through a web page, but it would also provide information that the image does not provide additional information to the text, so the user knows they are not missing out on anything.  As well as this, users could hover a mouse over the image to see the alt text.  This could be useful where an image doesn’t have a text caption underneath.

Difficulties with adhering to such guidelines and standards can cause barriers for people because content developers may then try to produce content to the lowest common denominator, i.e. text only.  Although text can be easily accessed by people using screen readers, it can be difficult for people with dyslexia to read and is visually unappealing.  So in this case, whilst the content is accessible for people using screen readers, it is less accessible for people with dyslexia.

Despite the drawbacks, this standardisation (“one size fits all”) approach is important.  Without a set of guidelines, developers may not know where to begin with accessibility and may not approach the basics in the same way, thereby reducing interoperability with assistive and other technologies.

One way to complement the standards approach is to produce alternative but equivalent versions of content.  For example, transcripts can be provided for podcasts, text heavy content can be offered as with animations or images or in simple language for people with learning disabilities or language learners.  This holistic approach has been proposed by Kelly, Phipps, and Howell in Implementing a Holistic Approach to e-Learning Accessibility.  This approach also takes student learning styles and pedagogy into account as well as technical and usablity issues.

Standards and guidelines are important but they need to be used with common sense and in combination with other approaches. Standards and guidelines can help with the physical presentation of the content, whilst holistic and other approaches can help the user to interact and use that content in the format best suited to their needs.

Personalisation – Many Things to Many People?

I finally got around to reading Designing for Learning: The Proceedings of Theme 1 of the JISC Online Conference – Innovating e-Learning 2006 (PDF Format, 788Kb) after several aborted attempts.  The paper I found most interesting was Diana Laurillard’s keynote, which got me thinking about personalisation of e-learning systems and resources. 

Laurillard talks about several different levels of personalisation:

* “…a pre-test to determine the level at which a learner might begin a learning design, or the chance to select the vocabulary set with which a language learner would like to work, or the opportunity to choose the order in which topics are confronted…”

* “…a negotiated learning contract that specifies the content topics, the prior learning and intended acheivement levels…”

* “…[an] adaptive system vision, in which opportunities are personalised for the learning, based on a diagnosis of their [learners’] needs.” (From Laurillard, D. Keynote: Learning Design Futures – What are our Ambitions? in Minsull, G. & Mole J (eds) (2007), Designing for Learning, The Proceedings of Theme 1 of the JISC Online Conference: Innovating e-Learning 2006, p10. JISC. Accessed 12/09/07).

There are no doubt other levels, but what struck me was that the idea that personalisation can mean different things to different people, depending on their requirements and viewpoint.  So, when two people talk about the personalisation of e-learning resources and systems, they may be envisaging completely different processes and interactions from each other.

To me, personalisation means accessible e-learning systems and resources that adapt themselves based on the learner’s learning needs and preferences.  So, for example, a visually impaired learner is offered alternative learning resources that have little or no visual element.  Or the e-learning system automatically changes the font colour and background colour based on a the preferences already set up by a dyslexic learner. However, these preferences and requirements for alternative resources are not just beneficial for learners with disabilities.  They are beneficial for all learners who may have learning, technology, or environmental requirements which differ from the norm (if indeed such a thing exists).  For example, a learner without access to an mobile audio device, such as an MP3 player, may prefer to print off the transcript of a podcast in order to read it on the bus.

The course designer may well have a completely different view of personalisation, whereby the e-learning system automatically presents the apropriate starting point on a course based on the learner’s level of competency and prior knowledge.  This could be established by online (or offline) pre-tests, tutor-entered proof of competency, such as certificated evidence of experience or skills, or other means of verification.  A learner who exceeds the initial competency requirements could be started a higher level of the course with options to view and/or take part in previous lower levels.

Both of these views of personalisation relate to Laurillard’s “adaptive system” approach, whereby the e-learning system “pushes” out resources or automatically places the learner at a particular level of a course, based on the learner’s needs and preferences.

However, there is another less formal approach to personalisation, which I’ve termed the “active approach” (although I’m sure there must be an official term for this out there), whereby the learner chooses what tools they want to use and/or the level at which they want to start the course.  In this approach, information and learning is “pulled” from the content managing system using the tools and approach that the learner prefers.  For example, a student may prefer to input all her assignment dates into a mobile device which she carries with her at all times, rather than input them onto the institution’s approved calendaring system, which she only accesses when she is on site. 

Although there does not seem to be much difference between the “adaptive” and “active” approaches, the details are quite subtle.  For example, in an adaptive approach, an institution’s system could offer a specific text-to-speech reader for all students to choose, should they wish, which is supported by institution’s IT (Information Technology) department.  However, in the active approach, the learner uses the text-to-speech reader they prefer to use.  Also, the active approach allows a learner to choose where they want to start learning based on their interests or prior learning. For example, a biology student with an interest in or prior knowledge of plants may want to start with a module on plant biology before moving to animal biology in order to orientate himself and gain confidence.  Laurillard’s idea of a “negotiated learning contract that specifies the content topics, the prior learning and intended acheivement levels” will help the learner to identify where they want to start learning.

Although there is a need for an adaptive approach to personalisation, there is also a complementary need for an active approach, which can empower learners, help hone their learning, and help them to gain confidence by consolidating any prior knowledge.

Personalisation does mean different things to different people – from a system which adapts itself to present content in the way the learner requires to learners actively choosing where they want to begin their learning.  Perhaps personalisation is all these things at the same time and it’s only that actual viewpoint that makes the difference.