A promise is not just a promise
A large international study led by Aarhus BSS researchers uncovers which honesty oaths work best.
In a revealing experiment, researchers tracked how 21,506 people behaved when given the chance to cheat on their taxes in an online game with real prizes.
"Honesty oaths have been used as a mechanism to deter cheating and dishonesty throughout history. Recently, however, doubts have been raised as to whether honesty oaths have the intended effect. Our research not only validates the effectiveness of honesty oaths but reveals that their wording and implementation are crucial for maximizing their impact. When carefully crafted, such oaths can have a significant impact on honest behaviour," says assistant professor Janis Zickfeld from the Department of Management at Aarhus BSS, Aarhus University.
Janis Zickfeld led the MegaOath megastudy, a collaboration between 42 researchers from the fields of linguistics, psychology, economics and management, including Karolina Ścigała, Christian Elbæk, Mathilde Tønnesen, Laila Nockur, Daniele Nosenzo, Yngwie Asjbørn Nielsen, Stefan Pfattheicher and Panagiotis Mitkidis from Aarhus BSS, Aarhus University.
To investigate the effect of honesty oaths, the researchers designed a tax evasion game. Over 20,000 people from the US and the UK took part in the online game, which included honesty oaths to see if this would reduce their propensity to cheat on their tax returns.
A total of 21 honesty oaths were tested in the online game and the results were clear: Honesty oaths increased the likelihood that taxes were correctly reported by 0.5 to 8.5 percentage points.
Top 10
The Aarhus BSS researchers conclude that honesty oaths could potentially be a good tool for reducing cheating and fraud. However, how effective they are depends on how the oaths are worded.
Based on the participants’ behaviour, the researchers determined the 10 most effective honesty oaths (increasing the likelihood of taxes being correctly reported by 4.5 to 8.5 percentage points). These oaths highlighted the importance of being honest when reporting income, they highlighted the harm done by not being truthful, such as the harm to people in need, or they highlighted social norms. They also emphasised that misreporting is forbidden, they appealed to individual responsibility, or they defined what dishonesty means or explained that honesty is an all-or-nothing concept: either you're honest or you're not.
An AI-generated honesty oath that emphasised several of these aspects also scored well in the study.
What the researchers recommend
The researchers recommend that honesty oaths use clear wording. Avoid ambiguous wording and instead be specific about what the honesty oath entails, for example: “I hereby declare that I will provide honest information when reporting my final income from the sorting task." This honesty oath reduced lost taxes by close to 50% (from 21.9% without the honesty oath to 11.6% with the honesty oath), making it the most effective oath in the MegaOath study.
Honesty oaths that focused on self-consequences or on relationships with others, such as general trust, empathy and the sense of community, had little effect in the study. However, focusing on trust can have a greater effect in situations where group bonds are stronger.
An effective honesty oath should also define what dishonesty actually means in the specific context. For example, in the online game, taxes went to the Red Cross, so dishonesty was here defined as less money for the Red Cross for one honesty oath.
The best time to present an honesty oath is just before the person has to perform a specific action, such as declaring their income. If there’s enough time, it’s a good idea to actively involve the person by asking them to sign or copy the honesty oath rather than just asking them to tick a box. However, a checkbox is better than no honesty oath at all.
Finally, the researchers suggest that it is especially effective to use an honesty oath when dealing with young men who exhibit low levels of honesty and humility.
How can we use this knowledge?
According to Janis Zickfeld, the study’s new results can be used in several ways:
"This knowledge can be applied across various contexts where dishonesty is a concern. It could be relevant for situations like tax reporting, insurance claims, or organisational contexts where workplace deviance and dishonesty are issues, such as employee theft, lying during job interviews and questionable absences. Introducing honesty oaths in these situations could potentially serve as an important tool for reducing incidences of fraud, both large and small. Future studies are needed to demonstrate this effect," says Janis Zickfeld.
Further information
More about the Aarhus BSS researchers behind the study:
Janis Zickfeld, Karolina Ścigała, Christian Elbæk, Mathilde Tønnesen, Laila Nockur, Daniele Nosenzo, Yngwie Asjbørn Nielsen, Stefan Pfattheicher, Panagiotis Mitkidis
We strive to comply with Universities Denmark’s principles for good research communication. For this reason, we provide the following information as a supplement to this article: :
| Type of study | Experiment |
| External collaboration partners | No |
| External funding | AUFF Nova, ERC, Israeli Science Foundation Award, National Science Center Poland |
| Conflict of interests | No |
| Other | No |
| Link to scientific article | www.nature.com/articles/s41562-024-02009-0 |
| Contact information | Janis Zickfeld |