this headline is pretty funny. but it highlights the mistake of polling the cost and benefit separately.
if you polled most americans and said "do you want a new car?" i expect a clear majority would say "yes." but if you separately asked them "do you want to spend $40,000?" a big majority would probably say "no." who wants to give $40k away?
those two polls, taken together, don't tell us anything about the number of people who would be willing to spend $40,000 for a particular new vehicle. on that question the numbers might be completely different than the responses to those two separate polls. if we're talking about an "is it worth it" question, you have to ask a question that causes each respondent to weigh the costs against the benefits. otherwise, the poll isn't all that illuminating.
