The Week in Vis: Week 12 (Nov 18-22) Evaluation
The topic for this week is evaluation: how do we know that our visualizations (or visualization systems, or visualization research) is any good. This is a really important topic, which deserves a lot of attention - but since we have the design exercise going on, we’ll need to limit it…
- The Week in Vis: Week 12 (Nov 18-22) Evaluation
- Class: Evaluation (Mon, 11/18)
- Readings 12: Evaluation - How do we know it's good?
- Design Discussion 4 (due Mon, Nov 18) (CLO)
- Class Survey 12 (due Mon, Nov 25)
- Content Survey 12 (due Fri, Nov 22)
- Class: Experiments (Wed, 11/20)
- Discussion 12 (due Wed, Nov 20)
- Seek and Find 12 (due Fri, Nov 22)
- Design Exercise 12 (due Fri)
Remainders from last Friday (DATE): Discussion 11, Seek and Find 11, DE11&DE12: 4 Questions, 5 Visualizations
The big thing this week is Design Exercise 12 (the main handin for the whole series of assignments). It’s described at DE11&DE12: 4 Questions, 5 Visualizations (a page with DE11 and 12), and will be handed in at Design Exercise 12: Final Designs (due Fri, Nov 22). While the official due date is Friday, the “unofficial” due date is the following Monday (you can turn things in on Monday without penalty).
If you’re part of the CLO, there is Design Discussion 4: DE12 Drafts and Critiques (due Mon, Nov 18).
For everyone… once you’ve completed DE12, please do the survey Class Survey 12: Design Challenge 2 Anonymous Post-Mortem (due Mon, Nov 25) - the survey is set to be due after the assignment.
Even though there is a Design Exercise due, there is also a big topic to discuss: Evaluation. There are some very important readings. The Chapter from Munzner’s book introduced the “Nested Model” - which I think is one of the most important ideas in the field. Be sure to read about it. We’ll also discuss it in class.
One odd thing with the Nested Model: I always remembered it wrong, to the point where I now have my own version - that is divergent from the real thing. My 4 layers of Task, Resources (Data), Design, Details is an “adjustment” to Munzner’s model (I’ll explain why in class). You definitely need to understand hers. Seeing how mine connects will help you apply hers more broadly.
The other required readings are still important, but quicker. The article by North is a good, brief introduction to the challenges of visualization evaluation. Cairo gives his usual practical perspectives on what it means for a visualization to be good (with 5 qualities). And I include a chapter from Tufte where he complains about various mistakes in examples.
The optional readings are optional. Many good things (and there are so many more examples). For this week, I suspect that you’ll want to put your energy into the Design Exercise.