Hacker News Reader

Comment by pg_bot | Hacker News Reader

Instead of buying more/better equipment, teachers, or technology why hasn't anyone tried to buy a better student? I'm fairly certain that if you paid elementary students cash for better grades you would see a huge jump in achievement for the average student. A small amount of money to us feels like a huge amount to a kid, and there are plenty of studies showing that rewards increase performance for basic and mechanical skills.

View on HN

On mobile and wouldn't be sure where to find it anyway, but this has been done! It turned out that the proximity of the reward was critical. A cash prize for a end of unit grade had nearly no effect, while an immediate cash prize tied to, say, a daily quiz showed some decent effect on the particular quiz/assignment. The take away from the study was that shorter feedback loops are important. Questions were also raised on intrinsic vs extrinsic motivators.

There's advice not to tie your children's chores to their allowance, because at some point they may decide that they have enough money and rationalize that they don't need to do chores.

Chores are about responsibility and being a contributing part of the family. Not about drawing a salary. I suspect those people would give the same advice for education.

The question then becomes is that the type of student you want to create? Any elementary school teacher who frequently rewards students with candy can tell you horror stories of what this sort of system leads to. Every task becomes a bargain of 'what will I get for doing this?' as opposed to building an innate desire to learn.

Education is broken, but going this route won't fix much. Fixing standardized tests, or scrapping them, along with eliminating things like the SAT would go a long ways is creating better learners, instead of better test takers.

You just need to bootstrap the learning far enough for them to see the big picture where the learning provides concrete value to the student!

... you'd also need that last bit to be true.

Punished by Rewards [0] examines some of the downsides with this approach.

[0] https://www.goodreads.com/book/show/541132.Punished_by_Rewar...

It has been tried.

https://www.thecrimson.com/article/2010/4/13/students-result...

Those who don't click the link may assume that the result was unsuccessful, but FTA:

"Although critics have long argued that paying students for results can lead to a decreased interest in learning, Fryer’s new results show no evidence to justify or support these claims."

That said, the researcher found that while paying students for results didn't cause harm, they did find that what worked even better (that is, worked at all) was "when they were paid to perform specific actions, such as reading books or wearing their uniforms."

“When you are paid to do something you may not know exactly how to accomplish, it’s not clear how you respond. But if you are being offered money to do something very concrete...then that might be easier to respond to.”

Not a silver bullet as the researcher notes:

“I’m interested in what are the four or five investments that, when put together, have the potential to close the achievement gap,” Fryer said. “Incentives could be one of [them], I don’t know yet. This is the first kind of large scale experiment in urban schools on incentives, and I hope it’s not the last.”

But also I hope people don't just dismiss the idea of monetarily incentivizing children out of hand.

I think the key thing they discovered was that it was more effective to pay struggling kids for the activities that lead to better grades than to pay them for the grades themselves.

Might not be limited to just struggling kids. Might not be limited to just kids.

It's interesting to think about what implications this might have for our own, grown up adult incentive mechanisms. Is it such a great idea to gate pretty much everything on results, when a) this is at least partially subject to chance and b) workers might not quite know what to do to achieve those results.

How is this going to work for something that isn't completely trivial, for example a mathematical proof? 5 kids do the proof, 15 kids copy, everyone gets a dollar. How can you do math in that environment?

How do you give grades in that environment?

If not with real currency at the least some type of "social currency"?

rewards increase performance for basic and mechanical skills.

But for complex skills, e.g. learning, they decrease intrinsic motivation, and thus performance.

Reply on news.ycombinator.com