Anyone involved in survey research today should keep abreast of the lessons of behavioral economics. However, it can be challenging at best to keep up on the latest academic research.
Luckily for us, Dan Ariely, one of the luminaries of behavioral economics, has written several books that are accessible and highly entertaining guides to the lessons of contemporary experimental psychology.
I just finished listening to the audio version of “Honest Truth,” read with a British accent by Simon Jones – which, as Ariely jokes at the end of the recording, is itself somewhat dishonest since Ariely himself has a prominent Israeli accent.
This book is an important and enjoyable read for anyone who cares about social science. There are many lessons taught about the experimental and real-life conditions under which people’s level of honesty varies.
Let me point out a few of the lessons that jumped out at me as having particular relevance in the world of research. In much of the research presented in the book, the idea of dishonesty is operationalized as cheating on laboratory experimental tasks for which money is being paid – a reasonable proxy for completing a survey online for an incentive.
- Cheating does not increase with a larger amount of money to be gained. This is useful information as we think about the effect of incentive levels on the propensity for respondents to speed and to become “professional respondents.”
- Increased likelihood of being caught does not have a negative effect on cheating. This counterintuitive but powerful finding can save us from perhaps ineffective efforts to communicate anti-cheating measures to respondents.
- People who are depleted physically, mentally or both are more likely to cheat. Perhaps we should give more thought to the conditions under which respondents take our surveys.
- People are more likely to cheat if they see others cheating. Here is further indication that the social proof from certain websites touting the benefits of quitting your job to take surveys full-time can have a deleterious effect on honesty in surveys.
- People raised in cultures which have more examples of socially acceptable dishonesty are more likely to cheat. Perhaps researchers should be open to the possibility that there is increased cheating in surveys conducted in some locations relative to others.
- Supervision decreases cheating. Is there a way to duplicate this “supervisor” effect for online surveys?
- Honesty pledges decrease cheating. This effect was only shown short term, so perhaps we should consider asking respondents to agree to complete the survey information accurately at the beginning of each survey.
Have you read “Honest Truth?” What did you think about the book? What lessons do you think the book holds for survey researchers?