Blog Post

Discussing data in Delhi

Last week I joined the DevInfo Field Reference Group at a series of meetings in New Delhi and Jaipur.  (DevInfo is the larger initiative that includes DIGW, but their core mission is to develop tools to help national and local governments, along with UN agencies and NGOs, compile and organize data related to human development goals.)  It was useful to see the new tools being developed by the DevInfo team and talk about possibilities for making simpler, "entry level" versions of some of the tools for learners, but probably the most interesting conversations revolved around potential ways to get users involved with human development data, with DIGW being just one example.  (One of my favorite ideas is for a system that pinpoints community "hot spots" -- say, a park or a streetcorner -- and then sends mini-surveys to volunteers' smartphones as needed when they pass by that spot.) Many of the conversations came down to what one means by "data" -- particularly the question of whether it is "structured data" (data that has predetermined types and attributes -- think spreadsheets) or "unstructured data" (e.g., natural language, images, video, etc.). Structured data has special status for many people -- and with good reason: it can be manipulated and compared relatively easily, and it's especially handy for measuring change over time.  However, most of the data in the world is unstructured, and structured data has a tendency to be, well, a little dull.  We like stories!

DevInfo leaders want to go from their current base of 20,000 users to 20 million users, and that has to mean broadening the data sets to include informal, user-generated (i.e., unstructured) data. From my point of view, of course, there is no other way if one wants to capture the interest of young people -- kids just assume that any website worth their time is going to welcome input from them and their peers.  And that doesn't mean we need to give up on information literacy.  On the contrary, the larger question really is, "how do we organize, display, and analyze information in a way that helps us make good decisions?"  And that, of course, is a terribly important question for learning in all sorts of domains, using all kinds of data.



No comments