I always like surveys like this, but I think that this one isn't really that helpful because of it focuses on the devices and not on the number of books read.
I.e., if this survey spoke to a person who read one book last year (say "Alice in Wonderland") on his iPad, and a person who read 20 books last year on his Kindle, the survey would conclude that the market was split 50-50 between Kindle and iPad. This, I think, also explains why the computer ranks so high as an e-book reader.
This data is useful for some purposes, I'm sure - but what I think everyone would really like to know is, of all e-books sold, how many were read on the iPad, Kindle, Sony, etc. And we'd like to know this by format, including DRM. And we'd like to know it by unit sold *and* by amount sold. And we'd like to know how many were indies vs. tradpubs.
Also, we'd like a pony. :-)
But, more seriously, I think the interesting question is not, "if you read an e-book last year, what device did you read it on," but "*how many* e-books did you read last year, and on what device or devices.*