Full circle effect of a survey poll

I have this thing going on in me for sometime. It was kind of "you feel it but can't figure out". Now after seeing the presidential campaign 2016, I think I got it. It's the poll survey of popularity in the presidential race, to be specific, but in general, any kind of statistics based survey poll. I strongly feel that the survey poll has "Double Loop Effect" of the survey result. The survey result influences people's view and inclination. For example: when a survey comes up with a poll survey where Donald Trump leads the race, it influences the wide variety of population and due to the "peer pressure" and "social conformity" effect, the population reflects that popularity in the next poll or other polls. I hope I can do a social experiment sometime in future to prove this hypothesis