Antony Unwin writes:
I [Unwin] find it an interesting exercise for students to ask them to write headlines (and subheadlines) for graphics, both for ones they have drawn themselves and for published ones. The results are sometimes depressing, often thought-provoking and occasionally highly entertaining.
This seems like a great idea, both for teaching students how to read a graph and also for teaching how to make a graph. I’ve long said that when making a graph (or, for that matter, a table), you want to think about what message the reader will get out of it. “Displaying a bunch of numbers” doesn’t cut it.
Hmm, I bet the developers of MS Excel's graphing would not like this idea.
Unfortunately things are not necessarily better in R or some other statistics package. Although the generic plot() function in R could help us to prevent the worst mistakes – though there is no substitution to thinking about what you are trying to do …
Does Columbia offer a class in statistical graphics?
Raymond:
Yes, we do. Or, I should say, we have. I don't know if it's still in the course listings.
I often assign my PhD students to take a table in a paper they were assigned to read and convert it into an intuitive graph. I then show all graphs they came up with on the screen and we discuss them. I think it is a useful exercise.
Matt Shum at Caltech has his students write an entire paper based on a single table! http://www.hss.caltech.edu/~mshum/gradio/mystery….
Frank:
Do you have a link to Shum's table? Of course, I'd much prefer a graph, but I'll take what I can get.
From Shum's assignment ":IG&GR"
Well I tried to copy from the pdf, the "provide your interpretation, then an alternate and then argue for yours"
Similarly useful advice I got in MBA school was to fully argue pro, fully argue con, then pick one (even flip a coin maybe?) and argue for one and "undo" the arguments in favour of the other.
Assume there is similar training in law schools.
K?
For me, nothing substitutes actually making a chart. I agree the Unwin assignment is nice, and most charts need much better titles, titles that tell a story rather than coldly describes what the data sets are.
My assignment is for the students to create a Junk Charts style blog entry. Find a chart, critique it, make it better.
Andrew: No, I can't find the table. It looks like he only handed it out (and I wasn't in the course, nor at the school…).
I support strongly the suggestions of comparing tables and charts and it is worth emphasising that tables and graphics are not in competition, they complement one another.
While we are on the subject of teaching graphics, another key issue is to ask the students to thnk about what they expect to see before looking at a graphic. I'm sure this has been said many times before (sometimes things that have already been said many times cannot be said often enough). The ideas of asking for expectations before and asking for headlines afterwards are both aimed at getting students to think, to be active. Passive acceptance of graphics happens too often.