Yes, I concede that could be a problem, but it’s a problem with pretty much any poll. For example, in politics, there are surely a lot of uninformed, poorly informed, or misinformed opinions. The purpose of polling is to get descriptive information about how a population feels about certain issues.
Personally, I think that allowing stats to be obtained with real money in such a direct way is dangerous. Because imc might rationalize: “Look at all these sales we got for magic scrolls (we both know that such an item would drive sales)! This must mean they are okay with the community!”
In fact though, the data thus far (as of 3/23/16 10:46 EST) shows that the community has a slight majority opinion that such scrolls are too powerful (15%) or that the scrolls should be completely removed (42%) [presumably because they are too powerful under any circumstance, is my interpretation]. So the view of the community indicates (in my view) that 57% of currently polled community find magic scrolls to provide an unfair advantage.
You are saying that at least some of these views are uninformed or misinformed, which could be true. But opinions in the public sphere can bias decisions. If someone is looking into this game and they have a ~6/10 chance (based on the population sample) of running into someone who thinks that there are cash options in this game that are too powerful, then they might decide: “screw ToS, I’m going to go play something else”. It might not be true, but again that’s imc’s problem not yours or mine.
I hope that you can agree though that I’ve tried my best to keep the polls that I did post unbiased while allowing for alternative options.