First off, let me put my little disclaimer out there: There are thousands, if not millions, of factors which influence approval ratings/poll numbers. I am not saying that this effect, if found to exist, is decisive. Additionally, this is not a political potshot or dig against anybody. Poll numbers are taken for all Presidents. Bearing that in mind, let me introduce what I hope to address here. There was a period a few months ago when I watched CNN daily for more than two weeks. It seemed to me that CNN reported President Bush's approval ratings every day during that time period. Even if the approval ratings were a week old, they were used as segways into segments; i.e. "with approval ratings at all-time lows, President Bush today defended his NSA wiretap program." After a time I began to wonder whether the downward trend of Bush's approval ratings was due at least in part to what I perceived as aggressive reporting of the same. It made sense to me: an avid newswatcher might be induced to erase his "slightly disapprove" and mark "strongly disapprove" if he thought to himself, "you know, two-thirds of this country disapproves of George Bush; 'slightly disapprove' isn't strong enough." Even if the poll numbers aren't directly in the conscience of the polling sample, they might contribute to an overall mood the sample has toward the President, similar to a dislike some of you may have had toward a certain person, but you can't remember why. So, in my opinion, poll numbers do influence subsequent polls. From what I observed a few months ago, there may even be a little bit of a snowball effect as each polling sample is exposed to the downward ratcheting of approval ratings. There would probably also be a positive snowballing if public approval for the President increased considerably. That is my two cents. Feel free to weigh in. Try not to be influenced by the thread's poll numbers.