Grannies Killed by College Exams

It’s true. College exams are deadly for students’ grandmothers. A study determined that granny death rates spike tenfold before a midterm, and nineteen times before a final exam. One theory is that grannies’ health is undermined by anxiety and stress when their grandchildren face exams. Indeed, the study found that failing students are fifty times likelier to lose a grandmother in the run-up to an exam, compared to non-failing students.

This is reported in Dan Ariely’s book, The (Honest) Truth About Dishonesty. Ariely is a professor of psychology and behavioral economics at Duke.

But seriously, what’s really going on is that students commonly make up grandmother deaths as a pretext for requesting exam postponements. Shocking.

The book’s main theme is that we all lie and cheat. But that doesn’t make us sociopaths. In fact, we tend to lie and cheat only so much that we can still look in the mirror and see an honest ethical person. We sometimes lie to ourselves.

Ariely invokes numerous laboratory experiments. In a typical case, test subjects are asked to solve a set of puzzles within a time limit, earning a payment for each one solved. But on an honor system: they self-report their performance. Most fudge it upward, but only by a little.
I found much of this suspiciously artificial and unlike real life. In another example, people were asked to gauge whether more dots appeared to the right or left of a line. Sometimes it was obvious, sometimes not. But when told they’d be paid substantially more for saying “right” than “left,” the answers skewed rightward. This Ariely called dishonesty. I disagree. If told I’d be paid more simply for saying “right” rather than “left,” I’d shrug and say “right” every time. That’s just a rational response to the rules.

Perhaps I’m quibbling. But most of Ariely’s lab tests entailed honesty along a gradient, falling in shades of gray. Whereas in everyday life ethical questions are often either-or. For instance, in my coin business, I normally send out orders before payment. Perhaps if, Ariely lab style, customers calculated their own bills, there might be some fudging. But when it’s just paying versus not paying, over 99% pay. Some even correct errors made in their favor.

This bespeaks honesty of a high order. Maybe my customers are not a representative cross-section, but I don’t think collectively they’re that unusual. Nor is my business. Most of the world’s commerce proceeds on a basis of mutual trust between trading partners; it’s our default assumption.

I once got an e-mail from a stranger in Africa selling coins. I gave him a substantial order. He didn’t know me, but assumed that an American businessperson would likely pay. And I did pay him after receiving the package. That’s how it works.

This basic level of trust is a fundamental underpinning of civilization. Of course we know we must watch out for violators; we lock our doors. Yet still you assume the average person whose paths you cross won’t bash your head in and grab your stuff. Or that a store won’t sell you defective goods. And so forth. Otherwise civilization could not function.

A recent poll found a significant decline in the percentage agreeing that most people are trustworthy. There’s no evidence we’ve actually become less trustworthy – only that we think people have.

Ariely seems to, pointing to scandals like Enron. But were businesses more ethical in bygone times? I doubt it; indeed, it’s harder to get away with scams in today’s interconnected media world of constant scrutiny and exposure. Yet that parade of exposures – Volkswagen is a recent example – does make people believe misfeasance has become rampant, compared to a romanticized past. I also suspect that decreased face-to-face personal interactions undermines our acculturation to the idea that people are generally trustworthy. But if that makes us less trusting, the decline in perceived trustworthiness can become a self-fulfilling prophecy.