This is my thesis and I don't have quantitative data to support it. I'm going more on experience than anything else, so feel free to disagree with me. The thesis is that the dominance of phone surveys through at least the last two decades of the 20th century has changed market researchers' thought patterns in ways we are not fully aware of. This has resulted in sub-optimal behavior when it comes to designing research studies. Let me explain a bit more.
The telephone is an aural medium which naturally has restrictions that a visual medium does not have. When you have to ask someone about the two Presidential candidates, aural media work fine. When you ask to rate their satisfaction with a product again it works fine. But what happens when you have to ask about the importance of various features in a new product? Since it is an aural medium, the number of ways of asking the question is very limited. The easiest way to do it is to provide one feature at a time and ask for a rating. But this does not allow any comparison between the features and certainly does not allow trade-off methods to be used.
Telephone based methods were so dominant for a period of time that Importance Scales became the default method for measuring preference. An entire generation of researchers grew up believing that that was the natural order.
With the advent of online surveys we migrated to a visual medium, but even now the aural methods have a strong pull on us. Importance scales are still used too often when we would be better off with comparative methods like ranking, constant sum, max-diff etc. I think the generation that grew up on phones (and there are still a few of us around) needs to get out of the phone frame of mind to truly be able to use the media that are available. After all we are at a point where online surveys are being seen as old school, so why stick with outmoded phone conventions?
Scales versus comparative techniques are not the only places where the phone legacy holds sway. If you look carefully at online questionnaires you can still see the phone impact in the wording used, the set-up of questions and other places. Understanding that an interactive web experience is fundamentally different from a phone survey, and can provide a qualitatively different experience for a respondent, can lead to engaging surveys with interesting ways of asking questions.
While online surveys have been around for many years it really is past time to think of them and their capabilities in their own right rather than as modified phone surveys.