# How do you raise the stakes in an experiment without actually paying people money?

I am trying to run a psychology experiment for one of my courses and it involves people trying to guard some "treasure" of theirs and trying not to lose it.

My problem is that I can't ethically (or legally) force them to offer up their own money and then actually penalize them for not doing well. At the same time, many of the participants that I've run so far have figured out that it's all a ruse and it's fake and they're not going to actually lose any money so the data may not accurately show how people react in those situations when there is something on the line. Or I could give them a hypothetical scenario in which their job is tied to how well they do but again, it's not real so people often just don't try.

Ideally, I would have some sort of funding for this where I can offer them $10 if they do well and slowly start taking away money as they fail to give them some incentive. However, I don't have any of that (it's a course project) and I'm trying to figure out how to incentivize them. Any help or ideas would be greatly appreciated. Thank you! • Can't you just pay them a fixed amount (you have to pay them anyway for a psychological experiment, no?) and deduct from that? – Captain Emacs Dec 2 '16 at 14:51 • No, I'm not paying them anything. These are fellow students and it is a course project. There is no funding for a course project. – noblerare Dec 2 '16 at 15:04 • Are you sure that fake coins don't work just as well? People can be quite competitive over trivial games... – T. Verron Dec 2 '16 at 15:46 • Brownie points may just do fine. – Captain Emacs Dec 2 '16 at 16:05 ## 2 Answers Make them create something with their own hands, something hard to make and use that as currency. I first became aware of this type of experiments from a ted talk by Dan Ariely which I recommend you watch, specially the last part: What Pushes Us To Work Hard — Even When We Don't Have To? The gist from these behavioral economics experiments, is that we value things we make substantially higher than an impartial evaluator,even more if making them was difficult,challenging or rewarding; so if you don't have money but have willing experimenters, you could possibly make them create something, let them assign a value and then use it as currency. You experiment would be as follow: • Create something - Origami,painting, sculpture of themselves • Assign Value - Ask them how much would they sell it for • Substitute for currency - Take their thing away from them, now you have something you can reward them with or even fraction. I have run a number of studies with incentive paradigms. There are several ways to do this without spending any/very much money. One of my favorite methods that works quite well is called a "hall of fame" manipulation. The paradigm is set up such that participants earn arbitrary points during the task. After I explain the rules of the game, I explain that they will have a chance to make our hall of fame, which is simply a high scores list. It's made up, but it comes up on the screen for them to view, and they are asked to enter in a user name in case they have a high enough score to make the list. Of course, for anonymity I don't actually record their user name or update the list based on participants' performance. At my school, undergraduates seem highly motivated by this because they are competitive. Another thing we do in my lab is associate arbitrary points with earning candy at the end of the experiment. I like this manipulation less because many undergraduates are trying to watch what they eat, and are not very motivated by this. However, in combination with the hall of fame list, it seems to work quite well. If you are able to get a little bit of money, what you can do is buy one Amazon gift card for say, 50 or 100$ and tell participants they'll win raffle tickets which you will draw from to select one winner after you've run your experiment. This has obvious pros and cons, namely, it's good that you don't have to pay each participant, and it's bad because some participants will decide that they have no chance of winning the raffle and they won't be motivated by it at all.

Finally, another option that is good for if you want to have several levels of incentives is to use symbols during the experiment such as: \$for low incentive, \$\$for medium, and \$ for high. In reality these amounts can correspond to pennies, allowing a participant to earn, say, \$10 max over the study, but they don't have to see on each trial how minuscule the money is. The symbols will motivate them more so than the amount of money would.

Given that you are doing this for a course, I recommend using points and a high scores list.

One thing that is annoying about doing an incentive manipulation is that if you fail to find an effect, it may be that there is no effect of incentive, or it may be that your incentive wasn't strong enough; you'll have no way of knowing. As with any study, a good idea is to have some other aspect of the data that you're interested in studying so even if your incentive manipulation isn't successful, you can still address some other question.

Hope this helps! Good luck.