9 out of 10 cats . . .

As I mentioned in earlier post, we’re starting our second year of TELIC with an examination of learning spaces and what that means for the learner.  By way of introduction we’re analysing a paper – ‘Rethinking the Virtual’ by Nicholas C. Burbules, following an introduction from @GuyMerchant.  People approach this type of exercise in different ways, but we wondered whether some of the visualisation tools might offer a different perspective.  Each of the following accepts free text, then performs some black magic in which some element of visual importance is generated as a result of the frequency of occurrence of a word or phrase.

The popularity of Wordle continues to grow, so that seemed like a reasonable place to start

No surprises, given the title of the paper, that ‘virtual’ features prominently, but we can also see other patterns beginning to emerge. ‘Space’ is clearly of major significance here, with ‘experience(s),’ ‘sense,’ ‘people,’ ‘time’ and ‘learning’ all clearly important too.  Given that we’re studying learning spaces, this paper clearly has something to offer then and perhaps the other terms imply that the human dimension cannot be ignored.

Many Eyes is an online tool which enables visualisation of both numerical and textual data.  In addition to Wordle, Many Eyes provides three additional visualisation techniques:

The Tag Cloud is similar to Wordle in that frequent words from the text feature more prominently in the cloud.  So the words mentioned above are the same ones which stand out again, however because the words are arranged in alphabetical order, plurals for example (experience/experiences) are more readily seen.  So the word ‘experience,’ occurring more frequently through its plural may now be considered more significant.  Additional features that this visualisation offers are through its interactivity – hovering over a word produces a pop-up which provides some examples of phrases within which that word can be found i.e. some measure of context.

We can also dig down for more detail by making use of the Search facility – from an examination of the main cloud, we can see more words starting with ‘i’ than might usually be anticipated, so we can focus on that area for further analysis.

We can then go one step further and make use of the ‘2 word’ function which produces a cloud based on occurrences of pairs of words:

And at once see the emerging significance of ‘interest, involvement and imagination’

The Phrase Net produces visualisations based on words linked by a conjunction; some presets are offered, but there is also the facility to provide your own custom phrase.

Having a space as the conjunction between two words produces quite a rich net which shows the words with which ‘virtual’ is closely linked – space(s), environment(s) and learning and how they in turn are linked with other words.  Interesting that the significant words (the ‘i’s) which emerged from the Tag cloud don’t carry the same weight here.

The Word Tree allows us to explore the beyond simple word and phrases, whilst still drawing significance from frequent words.

Clicking on branches within the tree narrows down the focus and allows to analyse the context within which important phrases can be found.  From the main tree, ‘virtual space and time’ clearly plays an important role, so we can investigate why this might be by exploring the sentences which both commence with and terminate in that phrase.

So what has all this told me about ‘Rethinking the Virtual?’  Well it’s provided some targets I’d want to explore further: the relativistic link with space-time sounds intriguing and the ‘i’ words are clearly important.  The question is though, have I got more from this than simply reading the paper?  Well no, certainly not, but this is a 9000+ word paper which takes some reading. What these tools might be able to do then is to allow significant aspects to emerge more quickly.  A more experienced user would doubtless be able to pull greater detail and richer information than from my tentative exploration.  But textual analysis in this way is, for me an infant discipline.  As such I guess it’s no worse than the often rudimentary way numerical data is presented – 9 out of 10 cats . . . !