So an interesting study showed up on the internet in the past few days (Daily Mail). Three grad students at Princeton University and Indiana University did an experiment to test the effect a font had on reading comprehension.
What they did was have a control group read a list of facts in a normal black arial font while the test group read the same list in a light gray comic sans font. The test group scored higher in the memory recall test 15 minutes later, which goes against everything I know about readability.
But that was just the lab experiment, with only 28 subjects. I'm sure you're thinking that a lab proof doesn't apply to the real world. They beat you to it and ran a second experiment with 220 students in 12 classrooms at a Chesterfield, Ohio high school. The results were very similar to the first experiment:
If you want I could give you a lesson in statistics, but basically you can look at the numbers as being above and below average the average grade. That's not quite true but it's close enough. Note that when the funky font was used a class scored higher than when the control class.
I'm surprised. I went into this thinking they had to be wrong. If this had been left as a lab experiment, I could have pointed out all the possible flaws. (I had this whole clever post about web usability that I can't use now.)
But I suppose we shouldn't be surprised. Learning is hard. It's supposed to be hard. And apparently making it incrementally harder results in an improvement in test scores.