Alex, yes always need to watch surveys - questions can be loaded and as you say the sample can be skewed or have a bias in it.
A friend works in academia and his department often gets corporate requests to survey products. Most of the advertisements claiming superior washing powder for example are based upon survey results and he has told me that the sample is tiny and they will repeat survey until they get the results their paying clients want to see. then thy claim "University results" or "independent Laboratory" results show this or that.
Its unethical but because the results are not required for peer review publication, they can do whatever they want.
So that washing powder cleaning brightness isnt based upon some fancy laboratory analytical instrument that measures brightness, but rather asking a group of 10 or 20 people what they think of the bed sheets and towels.
So 270 scientists from various disciplines and professional levels is probably a big sample compared to these "independent" academic surveys of corporate products and services.
(Political surveys are the worse for this sort of bias survey because they are deliberately used in the media as spin. 78% of Australians think we should be tough on refugees for example. You can participants "do you think we should help genuine refugees when they seek asylum in Australia?" and get one result or you can ask "Do you think boat people should be treated like all other asylum seekers and go through the normal channels in a fair and orderly manner" and get another result. Its PR spin industry)
