Generally the programmers that visit these kinds of websites, let alone participate in a survey, are the enthusiast programmers who are much more likely to be interested in exploring a new language in the first place.

There's a considerable potential for a selection bias here. Not that this disproves the survey, but generally these kinds of surveys tend to be a little bit ahead of the curve, so to speak.