Thursday, November 15, 2012

Dashboard Design for Education

I've posted several versions of reporting tools over the young life of this blog...but I haven't shared any dashboards. I've been holding onto my ideas for other purposes, but that doesn't mean I haven't spent time perusing what others are doing.

I have yet to find a commercial product that I would recommend. Every implementation of Cognos I've seen so far has been awful enough to make me beg for eyebleach.

3D Stacked Cylinders. Yes, really.

I spoke with an Edmin rep for awhile this summer. Very enthusiastic about their product, but admitted that they'd never consulted with any designers about the interface. It shows. Engrade suffers from the same issues. Here in Washington state, something called Homeroom has rolled out to most districts. When I talked to these reps nearly a year ago, they bragged about how their design was based on what teachers wanted. In fact, all these companies say that---and that's great. Knowing what data your audience needs to see is critical. But most educators are not designers and the incredibly poor output for all of these companies reflects this. I am disheartened by all the rich meaning that is hidden or lost because none of these companies can be bothered to consult with someone about line, color, and other basics.

But what happens when we build a design and don't have audience input? You get a very pretty dashboard...but is it useful?

Stephen Few recently hosted his annual dashboard design competition. (For a full discussion of the entries and selection process, visit his blog.) And this year, the challenge was to develop a student performance dashboard. Here is the winning design, by Jason Lockwood:


It's very pretty---and very Few'ish: the colors, the style, the fonts, and so on. Once you know what Stephen likes, all of the dashboards look the same. (I'm not sure that's a good thing, but we'll save that discussion for a different post.) The second place design, by Shamik Sharma, is below:


I think the challenge with educational data is simply that it's hard to "snapshot." We need to see every student---not just the Top 5. And there is a lot of stuff to consider---not all of it fits neatly into little quantitative variables. As I mentioned to Mr. Few at the workshop I attended, schools aren't making widgets: we're about people.

Those who entered this contest used a dataset that was provided to them, so I can't blame them for the volume of information contained here. But as I look at these as a teacher, I see information that isn't necessary for a dashboard (e.g., the standardized assessment results) and a lot of very poor grading practices represented (e.g., averages and letter grades). The designs are completely disconnected from the real-world audience. It is a limited audience, at that. These dashboards might be adapted for use in a secondary (or higher ed) core subject classroom, but not for elementary (multiple subjects) or performance-based classes (like PE). In short, the designs have a beautiful form and almost no function.

Can't we all just get along? How do we get those with expertise in the classroom connected with those who have expertise in data design? What would you like to see included with a dashboard?

No comments:

Post a Comment