Problems with this survey: "Wealth" versus "free time?" Why not a dollar value versus a time value?
Also, once you reduce it to that, the answer is very simple: look at how much people work versus how much they make per hour. Since when do we need to ask people how they would behave when we already have evidence of how they do behave?
The article did mention that, for the poor, those values were inverted. They valued money over free time.
So really it's just a supply/demand type curve, and most Americans are at a point where the nebulous 'free time' has a better value than an equally hazy 'more money'.
All polls are like that. The industry pretends that they are dealing with random samples, but they are never, ever doing so. They select for people who, unlike me, don't hang up the instant they recognize a cold call. I've got it down to 2, 3 seconds top...click.
One sure sign: A slight pause between my "hello" and the interlocutor's "Is so-and-so there?"--time in which the auto-dialer is connecting queued operators. I could probably hang up before I hear a human voice and still have a 95 percent success rate at identifying them.
I think this means a lot of people have jobs that require a fixed number of hours (say, 40 hours a week and 50 weeks a year) in exchange for a fixed amount of money. Americans may feel the number of hours is too high, but their wage per hour is fine. Basically, Americans work too many hours, and therefore would like to buy hours back.
E.g. Someone makes $100,000 on 5,000 hours per year. They can't work less than 5,000 hours, or they'll get fired. Their wage is $20 per hour. However, working 5,000 hours a year is painful and the dude doesn't need to spend $100,000 a year. He would happily buy back time at $20/hour. If a genie offered him free money or an equivalent amount of free time (with a $20/hour exchange rate), he'd take the free time.
Scott Adams recommended that loyal and productive staff be rewarded with additional holiday; that staff can purchase additional holiday and that everyone leaves work at 5PM. He called it the OA5 business process ( http://news.ycombinator.com/item?id=140712 ).
Also, once you reduce it to that, the answer is very simple: look at how much people work versus how much they make per hour. Since when do we need to ask people how they would behave when we already have evidence of how they do behave?