Tag Archives: signature

Sign First; Lie Less

A recent paper filed with the Proceedings of the National Academy of Sciences (PNAS) shows that we are more likely to be honest if we sign a form before, rather than after, completing it. So, over the coming years look out for Uncle Sam to revise the ubiquitous IRS 1040 form by adding a signature line at the top rather than the bottom of the last page.

[div class=attrib]From Ars Technica:[end-div]

What’s the purpose of signing a form? On the simplest level, a signature is simply a way to make someone legally responsible for the content of the form. But in addition to the legal aspect, the signature is an appeal to personal integrity, forcing people to consider whether they’re comfortable attaching their identity to something that may not be completely true.

Based on some figures in a new PNAS paper, the signatures on most forms are miserable failures, at least from the latter perspective. The IRS estimates that it misses out on about $175 billion because people misrepresent their income or deductions. And the insurance industry calculates that it loses about $80 billion annually due to fraudulent claims. But the same paper suggests a fix that is as simple as tweaking the form. Forcing people to sign before they complete the form greatly increases their honesty.

It shouldn’t be a surprise that signing at the end of a form does not promote accurate reporting, given what we know about human psychology. “Immediately after lying,” the paper’s authors write, “individuals quickly engage in various mental justifications, reinterpretations, and other ‘tricks’ such as suppressing thoughts about their moral standards that allow them to maintain a positive self-image despite having lied.” By the time they get to the actual request for a signature, they’ve already made their peace with lying: “When signing comes after reporting, the morality train has already left the station.”

The problem isn’t with the signature itself. Lots of studies have shown that focusing the attention on one’s self, which a signature does successfully, can cause people to behave more ethically. The problem comes from its placement after the lying has already happened. So, the authors posited a quick fix: stick the signature at the start. Their hypothesis was that “signing one’s name before reporting information (rather than at the end) makes morality accessible right before it is most needed, which will consequently promote honest reporting.”

To test this proposal, they designed a series of forms that required self reporting of personal information, either involving performance on a math quiz where higher scores meant higher rewards, or the reimbursable travel expenses involved in getting to the study’s location. The only difference among the forms? Some did not ask for a signature, some put the signature on top, and some placed it in its traditional location, at the end.

In the case of the math quiz, the researchers actually tracked how well the participants had performed. With the signature at the end, a full 79 percent of the participants cheated. Somewhat fewer cheated when no signature was required, though the difference was not statistically significant. But when the signature was required on top, only 37 percent cheated—less than half the rate seen in the signature-at-bottom group. A similar pattern was seen when the authors analyzed the extent of the cheating involved.

Although they didn’t have complete information on travel expenses, the same pattern prevailed: people who were given the signature-on-top form reported fewer expenses than either of the other two groups.

The authors then repeated this experiment, but added a word completion task, where participants were given a series of blanks, some filled in with letters, and asked to complete the word. These completion tasks were set up so that they could be answered with neutral words or with those associated with personal ethics, like “virtue.” They got the same results as in the earlier tests of cheating, and the word completion task showed that the people who had signed on top were more likely to fill in the blanks to form ethics-focused words. This supported the contention that the early signature put people in an ethical state of mind prior to completion of the form.

But the really impressive part of the study came from its real-world demonstration of this effect. The authors got an unnamed auto insurance company to send out two versions of its annual renewal forms to over 13,000 policy holders, identical except for the location of the signature. One part of this form included a request for odometer readings, which the insurance companies use to calculate typical miles travelled, which are proportional to accident risk. These are used to calculate insurance cost—the more you drive, the more expensive it is.

Those who signed at the top reported nearly 2,500 miles more than the ones who signed at the end.

[div class=attrib]Read the entire article after the jump, or follow the article at PNAS, here.[end-div]

[div class=attrib]Image courtesy of University of Illinois at Urbana-Champaign.[end-div]