Non-science, a cautionary tale

Scientific theories cannot be confirmed, an eminent philosopher of science has argued; they can only be falsified. But facts can be confirmed.

And here’s a fact about a famous (now infamous) social scientist: He’s a fraud.

It’s been confirmed.

The fascinating case of social psychologist Diederik A. Stapel serves as a cautionary tale. The caution is to those of us who write about science, who ruminate on what we read about science in the news, or anyone just curious about the world.

It turns out that not everyone who purports to have discovered something has discovered much of anything. Sometimes they’re simply lying.

Stapel is a liar. He’s been caught. He’s even confessed. In a rare instance of institutional integrity, the administration of Tilburg University has given him the heave-ho.

Stapel’s chosen social science is the weak, runt sister in the pantheon. It doesn’t have the mathematical and logical (and practical) honor that economics trumpets. Sociology encompasses a wider range of methods and targeted research programs. Cultural anthropology has nestled into a niche that seems impregnable: the study of backwater and Third World societies. And history has so august and firm a place in the study of man that no trendy influx of science or scientism into its narrative and archival traditions will lodge it from its high estate.

What social psychology has is charisma.

Its contributors are eminently quotable. They often seem to strike near the heart of what it means to be “human.” By focusing on the intersection between the individual and his/her mindset in the social context, and seeing how changes in such context can change opinion, feelings, and behavior, social psychologists often intrigue those outside the profession. Journalists adore and cover them.

According to Andrew Ferguson, they get credulous coverage.

Well, it wouldn’t be the first time. Credulity is a besetting sin of journalism. Journalists, who pride themselves on being so worldly wise and skeptical, regularly demonstrate otherwise, usually regarding politics. You know the reason; none of us are exempt. It’s easy to believe something that supports one’s own views.

Stapel was an extremely productive social psychologist. He published a lot, and was often quoted in mainstream media.

There’s a simple strategy to his productivity. Economist James Buchanan once confessed his secret: “Don’t get it right, get it written.” Quality will emerge from a wealth of attempts. (As it has done, in Buchanan’s case.)

Stapel’s method was a — call it “extreme” — version of that. Think of an idea, devise a test, and then just pretend you’ve done the test.That is, “the experiments . . . never took place.”

Ferguson takes credulity as the main lesson from this, and he has a great title: “The Chump Effect,” getting to the point up front:

Entire journalistic enterprises, whole books from cover to cover, would simply collapse into dust if even a smidgen of skepticism were summoned whenever we read that “scientists say” or “a new study finds” or “research shows” or “data suggest.” Most such claims of social science, we would soon find, fall into one of three categories: the trivial, the dubious, or the flatly untrue.

Stapel’s experiments and “findings” prove easy to mock. Ferguson does a great job — so good that I feel I can forgo the task.

Instead, I want to note the ideological slant to both Stapel’s agenda and to his excuse. He comes off as a self-parody of the modern doctrinaire lefty ideologue.

In the studies that Ferguson makes fun of, Stapel reveals himself as typically tendentious about sexual orientation, racism, and ideology. Some of what Stapel posited (“faked”) may even be true, for all I know. Do heterosexual men use racism and homophobia to “protect” themselves? Maybe some do. But other explanations come to mind for both attitudes.

In any case, the study published as finished by Stapel, and reported as complete, had not made the transit from potentiality to actuality. His lack of doing the work makes a mockery of his own isms and phobias.

He himself provided the best epitaph: Stapel explained his grievous faults with the same elaborate evasions that he “explained” heteronormativity, racism, and the sin of right-wingedness:

I did not withstand the pressure to score, to publish, the pressure to get better in time. . . . I wanted too much, too fast. In a system where there are few checks and balances, where people work alone, I took the wrong turn. I want to emphasize that the mistakes that I made were not born out of selfish ends.

Amazing. In the same breath that he confesses how he “wanted too much, too fast,” he asserts that he did not have “selfish ends” — as if proving one’s own ideology and getting famous at the same time is not selfish.

More precious, he insinuates that the “system” is somehow to blame for not having enough “checks and balances.” But wait. Social psychology is supposed to be a science. Science is the practice of public testing. (I’m pretty sure I heard that from more than one scientist, and not just Sir Karl Popper, referred to in my first sentence.) This makes science a “check and balance” system.

If social psychology lacks checks and balances, then there’s something unscientific about the way it is practiced today.

The problem appears to be this: Stapel’s faked tests were not performed by others. I read of no attempts to duplicate results. Thus, no opportunity for falsification — for actual public testing and, well, science. The faked experiments and results were, instead, merely reported and added to the “established findings” of social psychology.

Perhaps it’s worth taking the trouble to run some of Stapel’s un-run “experiments.” And then look at each, try to find ways to untangle causation from correlation and actually test the fraudster’s trendy hypotheses.

Sounds like a career path for some young soc-psych up-and-comer.

But would running someone else’s experiments — risking unwanted findings — get any press?

December 4, 2011

This column originally appeared on Townhall.com, on the first Sunday of December, 2011.