For example, half were asked if they agreed:

"People shouldn't drive in highly populated areas where other people have to breathe the car fumes"

and half got:

"People shouldn't smoke in highly populated areas where other people have to breathe the cigarette fumes"

4/14

(We originally considered specific v general questions, e.g.,

"People driving cars in public places should be liable for any harm..."

"People operating machinery in public places should be liable..."

but decided that changing the context was neater and less subjective) 5/14

Here's the full set of answers. As you can see, responses could change dramatically when driving was mentioned. All except Question 2 were hugely statistically different.

This doesn't make sense! The principle is the same in both forms of each question; only context changes 6/14

Is this self-interest? Cognitive dissonance? Most people drive, so it might make sense they'd make excuses.

But no. We separated out the subset of people who didn't themselves drive and they basically responded the same as the drivers, also making special pleading for cars 8/