It’s easier than you think to get people to commit bad deeds

When asking someone to do something unethical, we underestimate what a difficult position we've put them in. New research suggests that to avoid social discomfort, many people will agree to perform a bad deed rather than say no.

Vanessa Bohns and her colleagues first asked 52 student participants (31 women) to estimate how many people they'd have to approach on campus in order to get three people to tell a white lie. The lie was to sign a form saying the participant had given them a verbal introduction to a new university course, when really he/she had done no such thing. After making the estimate, the participants went out on campus to test their persuasiveness.

On average, the participants thought they'd have to ask 8.47 people before 3 agreed; in reality they needed on average to ask just 4.39. In all, 91 per cent of the participants overestimated how many people they'd need to approach.

A second study was similar but this time 25 participants estimated how many people they'd need to ask before 3 agreed to vandalise a library book by writing the word "pickle" inside in pen (ostensibly as part of a prank the participant was involved in). The participants' average estimate was that they'd need to ask 10.73 people on campus; in fact they needed only to approach an average of 4.7 people before 3 agreed to this task. Eighty-seven per cent of participants underestimated how compliant people would be.

The final two studies involved hundreds of people recruited via Amazon's Mechanical Turk online survey website. The participants were to imagine they were either the "actor," the "instigator," or a neutral party in a range of hypothetical scenarios involving such things as buying beer for underage kids, illegally downloading a movie, claiming expenses on personal dinners and so on.

The key finding was that people playing the role of actor said they'd feel a lot more uncomfortable if a friend or colleague (the instigator) nudged them toward behaving unethically (e.g. by saying it's stupid to pay for a movie you can get for free), compared with advising them to behave ethically. By contrast, those participants playing the role of instigator, or a neutral party, did not anticipate that the actor would experience this difference in social discomfort depending on the nature of the advice they received.

This result fits the researchers' belief, that the reason we underestimate how willing other people will be to comply with our unethical requests is because of a failure to take their perspective. The "truly startling" finding from this work, the researchers said, is not how many people are willing to lie or vandalise, but rather "the lack of awareness people appear to have of this tendency when they are in a position to influence someone else's ethical behaviour."

Other possible reasons for the results, Bohns and her colleagues suggested, are that recipients of unethical requests reframe them as prosocial acts – after all, they're helping someone out – or maybe their compliance is simply a way to win popularity. Future research could examine these and other possible explanations.

This new research has echoes of Stanley Milgram's classic work. His students and colleagues dramatically underestimated how many participants would be willing to obey a scientist and administer a deadly electric shock. Thankfully there's also a positive twist to the phenomena documented here: similar past research has shown that we also underestimate how willing people will be to comply with our requests that they help in prosocial ways – such as lending their phone, or giving money to charity.

Further reading

Bohns VK, Roghanizad MM, and Xu AZ (2013). Underestimating Our Influence Over Others' Unethical Behavior and Decisions. Personality and social psychology bulletin PMID: 24327670