After many regression analyses and much hierarchical linear modeling, the professors discovered that their conclusion matched their hypothesis: The “powerful” students—that is, the students who said on the questionnaire that they were feeling powerful that morning—showed less dramatic reactions to the stories than other students. Or, as the professors put it: “Our data suggest that social power attenuates emotional reactions to those who suffer.”
I told you it was boring. It was also preposterous, at least as an experiment designed to test a hypothesis. The questionable assumptions fairly cry out from where they’re buried. Just for starters, can a questionnaire asking a college sophomore how powerful he feels tell us whether he’s powerful? Researchers never measured the elements that made an “emotionally evocative story”; the stories were rated by grad-student coders whose own feelings of powerfulness were unrecorded. And underlying the endeavor was the silliest buried assumption of them all, that the way a college kid reacts in a psych lab while he’s wired to a machine and jabbered at by a stranger has some—any—relation to how “rich and powerful” people (Edsall’s phrase) live their lives.
If such a study claimed to prove a different conclusion, and presumed to tell us that rich and powerful people were more compassionate than those with less wealth and lower social standing, we could expect our psychopundits to approach it with more of the skepticism that journalists are so famous for. But skepticism would put a psychopundit out of a job, and so the violations of logic and common sense simply ramify. Among the studies that constitute the recent “academic critique of the right,” one used participants—more than 65 percent of them female—solicited over Craigslist; another recruited participants through Amazon’s Mechanical Turk website. Neither sample could possibly represent any group other than itself.