Coursera Studies


Over the course of “A Beginner’s Guide to Irrational Behavior,” we attempted to demonstrate the scientific methods of behavioral economics by giving students the opportunity to participate in research. We sent out a weekly survey and shared the preliminary results with students, and will continue to update this site as we learn more from the data.

Pre-course (and Week 1) Survey

Most of what we learned from this survey is demographic information, which is shared in the About the Class section of this site. However, we also did something tricky with the pre-course survey when we asked how much time students planned to spend engaging in outside material related to the class. For half of the students, this question was accompanied by a recommendation from Dan to spend an extra 1-3 hrs/week. And if students saw Dan’s recommendation, they were much more likely to say that they’d spend more time with outside material.

We also asked what students planned to do so that we could compare their predictions with what they actually did.


Week 2 Survey (Psychology of Money)

The Week 2 survey looked at the pain of paying. We were curious about whether the pain of paying in one domain translated to the pain of paying in other domains. Perhaps surprisingly, we found that this is not necessarily the case, even when we control for factors such as your country of citizenship. There is still a lot to be done with this data, but nothing is popping out at us yet.

Week 3 Survey (Dishonesty)

The Week 3 survey looked at cheating behavior and whether students might fudge the number of times that a coin flip resulted in an outcome that was in their favor. We asked students to send the survey to a friend or family member, and that the money that students made from the coin flip would (potentially, based on a lottery) be given to themselves, the person they sent the survey to, or a random person who completed the survey. So far, it looks like our students were too honest and reported the outcome of the coin flip to be about 50% heads across all conditions.

Week 4 Survey (Labor and Motivation)

The Week 4 survey looked at gift giving and receiving. We’re still working on this one…

Week 5 Survey (Self-control)

The Week 5 survey was all about self-control. We were interested in whether self-control behavior, willpower, and strategies translate over different domains: eating, exercising, spending and working. We haven’t yet found any interesting individual differences in these areas (we looked at gender, age, religiosity and country), but we did notice that students’ precommitment plans for this class in the pre-course survey (by blocking off time, rewarding themselves for positive behavior, or setting consequences for negative behavior) predicted whether they were more likely to report using the environment for self-control — far more than greater willpower. (This wasn’t true for social plans, such as joining study groups and telling others that they were taking the course.)



Week 6 Survey (Emotion)

In the Week 6 survey, students were matched with a fellow participant and asked how they would split $10. However, instead of being connected with a fellow classmate, students were actually paired with a video of an actor. In this study, we were interested in how people perceive deliberation. To do that, we set out three different sorts of offers: either $1 out of $10 (an unfair offer), $3 out of $10 (a moderately fair offer), or $5 out of $10 (a perfectly fair offer), and then we either had the actor make the offer immediately or after a period of deliberation. We needed to use a video of an actor so we could precisely calibrate the length of the deliberation.

With all students in the role of Player B, granted the right to either accept the proposed offer or leave both players with nothing, we were curious about how the amount of hesitation would affect 1) willingness to accept the offer and 2) perception of Player A.

Not surprisingly, fairer offers are more likely to be accepted in the Ultimatum Game. It turns out that acceptance rates are unaffected by how much Player A deliberates before making an offer. (Interesting? Maybe..) However, there is a teeny tiny effect of hesitation where greater hesitation leads to greater likability of Player A. We see this effect the most in the $3 condition where the fairness of the offer is slightly ambiguous. We wanted to share this initial glimpse at the data with you, but also say that there is much more to be done — we’ve only scratched the surface!

Post-course Survey

time-spent rate-the-class regarding-achievement your-participation-1 your-participation-2