“About 7 in 10 U.S. adults between 18 and 29 say ‘a great deal’ or ‘a moderate amount’ of responsibility falls on profits made by health insurance companies, denials for health care coverage by health insurance companies or the person who committed the killing.”
“Young people are also the least likely age group to say ‘a great deal’ of responsibility falls on the person who committed the killing. Only about 4 in 10 say that…”
my take is that polls such as this one aren't very informative. they assume, afaik, that the survey respondents are passive describers of their opinions. i dont think people necessarily see polls this way. its one reason that issue polling and political outcomes are broadly inconsistent, in that people don't vote for the issues they say they support.
lets suppose you are a 25 year old guy who has had really bad experience with a health care insurance company. young people, on net, subsidize others' care, paying forward what they will consume later. in rare instances, young people need a lot of care -- and if the insurance company makes that hard, it is especially frustrating. like, "im not getting dialysis or hearing aids or cancer treatment. i was in a car accident that fucked up my back. i am not asking for that much. just cover my fucking physical therapy."
all right. so now a pollster calls you and asks what you think of the Luigi killing. if you are like most americans, you have almost no power to change anything. at most, you can complain to your employer, who might or might not care. you're probably a lower-level employee, easily replaced and lacking any pull within the company. you can complain to the insurance company, but nobody there gives a shit. you've watched elections in which health care has curiously played almost no role and was rarely even discussed.
so now you have someone specifically asking your thoughts about the matter. wouldn't you take this opportunity to vocalize your dissent? its a shitty medium, to be sure, but at least it's a medium. so do people say "the insurance industry is mostly to blame" because they legitimately think that? or because they are wanting to express a frustration that finds little outlet anywhere else. saying "the killer is to blame" is punting away this opportunity that might not come around again.
i don't know exactly why researchers are so intent on taking polls at face value. i suspect it's the same reason that statisticians so frequently assume gaussian distributions. its easier. the math gets much, much messier for other types of distributions (especially empirically determined ones), and its much harder to get a finding. so too with surveys. if you find yourself asking how many of the responses were strategic, its going to be hard to get a good paper. ignoring the problem gives you a publication.