Exploratory visualization – Data visualizations that are used by the designer for self-informative purposes to discover patterns, trends, or sub-problems in a dataset. Exploratory visualizations typically don’t have an already-known story. Iliinsky & Steele, 2011
There is a certain power from letting people explore a dataset, many eyes having the potentially to provide many insights (not ignoring the danger of many misconceptions). That is not to say that data visualisations designed to ‘explain’ data are not without value and often in sites like the Guardian Datastore they use both explanatory and exploratory techniques in concert.
When presented with over 4,000 survey responses the challenge was how to let people explore the data set. When presented with this challenge the first thought invariably is what is the shape of the data. In this case the survey responses were collected in Survey Monkey. After considering options like consuming the data into the OER Impact Map via the Survey Monkey API, the overhead in terms of developing user interfaces and squeezing into a WordPress data structure resulted in the exploration of other options. The approach that looked like it would squeeze the most functionality out of little development time was to use Google Fusion Tables to host the data and put a visualisation layer over the top. The reason for choosing Fusion Tables is it allows a Guardian Datastore model of letting the user easily access and reuse the source data either in Fusion Tables itself or exporting into another tool.
If you’ve been browsing the OER Impact Map site and wondering why you haven’t seen the ‘exploratorium’ it because it’s currently a bit of an ‘Easter Egg’. This is because I never quite got it working the way the OU team would like. That also gives me a bit of comfort labeling it ‘a bit beta’. Here’s the magic link to try out the OER Survey Exploratoratorium followed by a short video highlighting the main features.
If you would like to peek at the data behind this there are two tables: one with the survey questions and another with the survey data. If you like a hint as how it was made it’s mostly Google Charts with a dusting of jQuery UI (and if you are really keen here’s the current source code). At this point it’s over to you. We’ll be developing how the survey data is interfaced and your input is welcome.
Iliinsky, N., & Steele, J. (2011). Designing Data Visualizations: Representing Informational Relationships. O’Reilly Media, Inc..