I've been doing a series of internet polls ...
that generally ask political/economic questions.
The most recent poll asked a series of questions regarding my confidence in the economy, e.g., "How confident that your household will be better off, about the same, or worse off, or Not Sure, in the coming 6 months?" and "How confident that your household will be better off, about the same, or worse or Not Sure off in the coming 5 years?"
At the end of the survey, they ask questions about the design of the survey, asking participants to rate (on a 10 point scale) whether the survey exposed a bias; whether it was interesting; and, How easy was it to complete. The survey also included a comment section.
In the comment section, I wrote: "Regarding the economic confidence question, this survey's choices completely ignored that any honest and/or political informed/engaged respondent (regardless of partisan/political affiliation) would have been forced to answer "Not Sure", because that answer would depend on the outcome of this coming Presidential and Congressional elections. For me, a win for President Obama would lead me to answer, "Better Off", since my household is better off than it was 4 years ago; but a romney win would have me answering "Worse Off" because his policies are a return to the pre-President Obama policies, that exploded the economy.
I have not yet seen the results of that survey, though they ussually post the results the following week. Maybe, they got my point!