Results From another Area of Reasearch: Education Incentives
Not too long ago I worked myself into research about monetary incentives in education. Various economists and local initiatives experiment with rewarding students in school or university for specific performances.
The most renown researcher in the field is Roland Fryer from Harvard University. Together with Bradley M. Allan he published a paper, where I have the following success rules for incentives from. They sum up all the field-study findings quite well as I think.
I left away the ones that seemed irrelevant for gamification, you can find them on pages 17-22, Link:
1. DO PROVIDE INCENTIVES FOR INPUTS, NOT OUTPUTS, ESPECIALLY FOR YOUNGER CHILDREN.
- Inputs means, you have to do a task and when you did it you will receive a reward. For example the SCN mission to read and bookmark a specific document.
- Output means that the way of fulfilling the task determines your reward. For example the SCN mission where you get a batch if someone marks your comment as useful.
Inputs are fully known tasks whereas Outputs contain both uncertainty and work. People go after low hanging fruits, but only after the higher ones, when there is an occasion.
2. DO THINK CAREFULLY ABOUT WHAT TO INCENTIVIZE.
A banal result is that people do exactly what they are paid for. There are clearly spill-overs to non-incentified areas but you have to carefully check what you reward with what. You cannot rule out negative effects, neutral effects or positive effects. It always depends on the specific case.
For example, when you reach the SCN mission where you have to bookmark the rules of engagement but it’s not possible to access the document due to a flaw in the system it might make you disappointed about the whole system.
3. DO ALIGN INCENTIVES.
You will get better results if you grow the interest in a performance outcome to others, who are involved.
For example, when SCN moderators have an interest in helping newbies in the network through their first missions. (Not sure if that is the case;p)
4. DON’T THINK THE EFFECTS GO AWAY IMMEDIATELY AFTER THE INCENTIVES ARE REMOVED.
A task may be unimportant when it is part of the curriculum, but it could become important very much later in a very different setting.
For example, when you read about scientists who give students money for good grades and then you end up writing about it for people who professionalized in a completely different field, because you think it might help them.
7. DO IMPLEMENT WHAT WORKS.
Not everything that looks like it should work will work. You cannot translate the positive results from one incentive to another. Checking the specific results is important or you might end up wasting money and effort. On the other hand incentive structures that work in one area quite likely will also work in an other area.
For example there are more people on SCN wrote a status update than people who wrote a blog. Both tasks are about writing something on the network, but the one thing is a short brain **** whereas the other one is (supposed to be) an elaborate thought with actual content. You don’t want to screw up that, especially the first one.
11. INCENTIVES NEED AN OBJECTIVE BASE.
This last one isn’t part of the list, but I found it elsewhere and I think it is of relevance. Incentives can have a profoundly negative effect if they are based on subjective measurement. Usually people at least unknowingly give their own opinion a higher value when they see it somewhere else, even if it might be wrong. We are subjective beings and so is our judgement of reality.
Incentives and rewarded tasks therefore always have to be quantifiable with numbers. Otherwise you could create a lemon market that punishes good performances.
There are of course limits to possible transfers of this knowledge to gamification:
- It’s about grown-ups, not children or students
- There is a difference between monetary and non-monetary incentives
- Objectives are less individual but more collaborative
But as I think there might be some transfers possible:
- Inputs are better than outputs
- Always consider possible spill-overs
- You can multiply effects by aligning different interests
- Not everything works that looks like it could
- Every game will get you what you have designed, so watch out for what you have designed
- Illogic tasks, missions, rewards and structures can cause significant problems and should be avoided
oh, there is censorship here... the 4 stars were meant to be a "fort" with an a instead of the o 😉