I think this comic gets the media perfect

http://www.phdcomics.com/comics.php?f=1174

I've had it happen with one of my studies. Someone else noted and gave a label to the following media/user phenomenon: Whenever the major media reports on something with which you're really familiar, you realize that they inevitably have almost everything distorted or just plain wrong, yet when it comes to other subject areas you tend to ignore this and accept their reporting as it were reliable information.

The phdcomic has it right but it is actually worse because often the colleagues accept the research as being valuable. I knew of a person who did research on collaboration and groups interaction. BUT it only used students, which means that you COULD NOT generalize from the findings to the general population because students are not the general population. But researchers too it as gospel as if it was generalizable.

The premise fit the preconceived notion so run with it!