Blogger typology: using IBM’s Many Eyes to build matrix charts

Thanks to IBM’s Many Eyes service it’s relatively simple to create complicated visualizations that my current version of Excel can’t handle. For example, this “matrix chart” that I built using Excel’s bubble chart function is clearly unacceptable. I can’t easily link statements or values to the X and Y axes, and there’s lots of overlapping that seems (after many attempts) to be impossible to fix.

Matrix chart built using Excel - not very satisfactory!

Matrix chart built using Excel



I’d considered using something like Processing to draw these charts, then remembered Many Eyes. Sure enough, not only does the service help me build matrix charts, but it also lets me embed the visualizations into this blog post and lets you interact with the data (feel free to click on the image below and have a quick play.)

This chart depicts the statements that generated most “consensus” (either agreement or disagreement.) While this might be interesting to some people, I’m actually looking for areas where there’s least consensus — the whole point of the blogger typology is to identify those variables that will, so to speak, separate sheep from goats and apples from oranges. To continue with this spurious analogy, I’m looking less for statements like “I am a ruminant” or “I am a fruit” and more for statements like “I enjoy eating laundry” and “I am a citrus.”

So here’s a quick look at those statements:

You should immediately be able to see that these responses are much less polarised than in the previous chart. That’s the sort of thing that we’re looking for.

Caveat

Please bear in mind that this current sample is self-selecting, and sourced online via Twitter, this blog, LinkedIn and WOM referrals. What I’m working on here are the techniques I’m going to be using, rather than the actual analysis. I’m really quite into the prototype-and-fix approach to planning (more like “get it up and fix it later” — it means I spend less time beard-stroking and more time in front of a martini.) So the current results aren’t very robust, but they will help us focus our future research.

Where can I download the latest data?

Here, as a Google Docs spreadsheet. If you do anything interesting with the data, do please let me know, and I’ll link to it from here. Feel free to share this spreadsheet as http://icanhaz.com/blogger_data