A recent Clarus poll was apparently designed to promote the candidacies of moderate Mitt Romney and a liberal mayor, Republican Michael Bloomberg, and diss Mike Huckabee and Sarah Palin.
Just below, has to be one of the worst polling questions in the history of presidential politics.
If you had to pick between the two––would you prefer the next Republican presidential candidate be a moderate conservative who has a good chance of beating Barack Obama …or… someone who is an outspoken conservative who has only a fair chance of beating Barack Obama?
It is unclear whether folks taking the poll were also forced to choose between the following two descriptions:
1. A moderate, pragmatic conservative 61%
2. A staunch, outspoken conservative 29%
How do I loathe thee? Let me count the ways.
1. The initial premise is flawed. Whenever a poll question forces you to choose, (“had to pick”) it is quite certain you are being had.
2. The question pairs ideas in a manner deliberately created to push one kind of candidate over another. Notice that the moderate candidate is presented as having a good chance to beat Obama while the conservative is only given a fair chance. If a respondent to the poll believed that the conservative had a better chance they would have no place in the poll to make the case.
The poll also asked voters to “vote” in a three-way race between President Barack Obama, a Republican (Palin or Romney) and Michael Bloomberg. Romney did better than Palin in comparative races, for obvious reasons. Palin is unquestionably considered more “divisive” than Romney at this time, but it is quite doubtful anybody outside of New York even knows who Bloomberg is. This question at best shows a general disdain with party politics, and could not possibly indicate a positive attitude towards Bloomberg.
All of this shows not only the futility of putting too much stock in polls generally, but also the need to be aware of the shenanigans going on behind the scenes. Polls like these are not only unscientific, they are downright deceitful.