Just about every compliance certification has the employee sign at the bottom. We have been signing letters and contracts at the end for millenia.
But maybe there is a way to increase ethical performance by moving that signature to the top.
Lisa L. Shu, Nina Mazar, Francesca Gino, Dan Ariely, and Max H. Bazerman recently published a paper that found differences in compliance/ethical performance depending on whether the participant signed first or at the end.
In one experiment, the subjects took a test and scored it themselves. They would be paid based on their performance and reimbursed for their expenses incurred in attending the test. After self-scoring the test they went into another room to self report their income on a tax form. There were three forms:
- One with a certification at the beginning that all information is true
- A second with the same certification, but at the end
- A third with no certification
The test and reporting was set up to be very easy to cheat, with a simple and immediate cash reward for cheating. You should not be surprised that cheating was rampant.
With the third form, with no certification, cheating occurred 64% of the time. With the certification at the bottom, the cheating actually rose to 79%. The winner, with the certification at the beginning, only had a 37% cheat rate.
Moving the certification to the beginning had a dramatic, positive effect on reducing cheating.
The paper includes several other similar experiments with the same results. A slightly different test involved word puzzles. Those that signed an honesty pledge before engaging in the cheating experiment ended up solving more of the ethics-related words than the others.
The authors theorize that the certification at the top pre-sets the person to start thinking more ethically. If they don’t hit the certification until the end, they have already supplied the information with whatever ethical slant they may have.
I’m going to re-think how I design my certification. At the top will be a certification that all of the information is true and correct, before they start filling in the information.
Sources:
- Signing at the Top: The Key to Preventing Tax Fraud? by Carmen Nobel
- When to Sign on the Dotted Line? Signing First Makes Ethics Salient and Decreases Dishonest Self-Reports by Lisa L. Shu, Nina Mazar, Francesca Gino, Dan Ariely, and Max H. Bazerman
Very interesting that this small change could make such a difference.
Doug: I’m not sure this research has any bearing on legal documents. A test is a task you perform, so it’s not surprising that a reminder regarding accuracy would be more effective if you’re reminded before performing that task, rather than after.
By contrast, a certificate isn’t a task you perform. Instead, it’s a statement. So to be most effective, you’d want the reminder to come after the signatory has digested the statement but before they’ve signed it. That would suggest that it’s best to place any assertion of accuracy at the end of the document, before the signature.
Similarly, that’s why I think it’s best to place the signatures at the end of a contract.
Ken
Ken –
I agree that the study did not focus on contracts and the appearance of a signature at the beginning or end is largely meaningless. In many commercial contracts, the signature is generally somewhere in the middle, between the text of the contract and the schedules/exhibits to the contract. (I think the test of a contract is whether you read it thoroughly before you sign it, which rarely happens.)
I see a variety of certifications, from one that is a simple statement to one that requires many questions to be answered and information to be provided. As the certification moves more to being a task, like filling out a tax return, the benefits of moving the signature to the beginning are more likely to benefit. The study showed some remarkable results, especially when you compare the simple change to the form shown in the appendix to the paper.
You’re correct that if the certification is merely approving a statement, then it may be better to have the signature after the statement. (I would be interested to see some empirical evidence.) But once you need to start filling in blanks, the accuracy statement at the beginning appears to have positive effect on accuracy, at least according to this paper.