No, we get lots of data because lots of people answered. Since there is no information about how the people that answered the survey relate to the demographic of all GNOME users, the data interpretation will be hard, or probably even impossible.
Originally Posted by 3coma3
To make an analogy: If I make a survey about the morality of eating meat, and then post that survey at a PETA meeting, you would get lots of data, but it would hardly reflect the true response of the general public.
And yeah, of course you can "parse the survey data" to make it more understandable. But no amount of parsing can take care of a bad survey. It's the old "garbage in, garbage out" rule.
Looking at the design of this survey I'd say any data you get out of it is useless, and even trying to interpret it is a waste of time.