Jim, even your simple poll could be a potential source of bias. All of the questions led with the same political party ( or its candidate), in this case the Rupublican. Being first on a list often confers an advantage whether it's ballot position, a list of web site links, phone book listings (AA Plumbing Service)etc. It could also generate a false coat tail effect with those who choose a named presidential candidate tending to choose the same party for the other offices even though there may be a very popular member of the opposite party. Becasue the names are not listed, that conncetion may not be made. There was an interesting article on the web magazine Slate today about polling. In the past few years, more and more people are refusing to answer polling questions (partly because there are more and more surveys being conducted). As a result the pollsters have to call many more people to get their 500 or 1000 sample size. If people of one political persuasion have a less favorable view of the media and pollsters, they may disproportionately refuse to participate, and skew the results.

As for upstate New York consider: 1) It could reflect reality; 2) It could be the Republicans in that area choose to spend their money on yard signs and the Democrats spend it on something else such as fliers or newspaper adds; 3) It could be the Republicans have so much money that they don't worry about wasting it on yard signs.

rich r