Incentives Work

I've always thought there was something a little funny about No Child Left Behind's efforts to use standards and accountability to get teachers and schools to perform better. Why not just go right to the source and given students direct financial incentives to do well in school? Decent people find the idea abhorrent, I know, but there are clear theoretical reasons to think it would work and the empirical evidence suggests that it works.

Now naturally every individual actually has strong incentives to do well in school anyway. But children tend to exhibit a very high rate of pure time preference. Short-term financial incentives (or, indeed, non-financial incentives like the gold stars my elementary school teachers used to hand out) help align the short- and long-run incentive picture. Meanwhile, educational attainment has positive externalities, so it's worth spending money on.

Comments

But children tend to exhibit a very high rate of pure time preference

Hell, so do I, and I'm 45 years old and have a master's degree.

Posted by guyx | February 16, 2008 10:55 AM

Incentives need not be monetary with kids. Many teachers employ simple substitutes like a jar of marbles in the classroom. When the students perform or do as told, marbles are deposited. When it overflows they're rewarded with a party or something like that.

Posted by Jake | February 16, 2008 11:01 AM

Individual incentives still do nothing to fix the underlying problem of huge disparities across socioeconomic status. Just like middle and upper class suburban school districts are already rewarded for their higher quality education with scholarships and college acceptance, so to would it be with individualized incentives. The schools with poor infrastructure and lacking resources will fall further and further behind their more well-endowed counterparts, whose already better off students will get payed for doing well in school. Individual incentives are only one part of the solution to the disaster of NCLB. We need to be spending A LOT more on public education in this country.

Posted by Alex | February 16, 2008 11:27 AM

Incentives are tricky. It seems to have calmed down lately -- though I don't know the final consensus -- but there was a big fight in the psych literature about whether extrinsic rewards undermine intrinsic motivation, e.g., whether paying somebody to read made them read less when they weren't getting paid. here's an example abstract:

http://content.apa.org/journals/bul/125/6/627

Posted by stephen | February 16, 2008 11:41 AM

This is exactly the kind of incentive that doesn't work (i.e. incentives on qualitative activities).

"Punished By Rewards: The Trouble with Gold Stars, Incentive Plans, A's, Praise, and Other Bribes"
http://www.amazon.com/Punished-Rewards-Trouble-Incentive-Praise/dp/0618001816/ref=pd_bbs_2?ie=UTF8&s=books&qid=1203179687&sr=1-2

Harder, theory-oriented:
http://www.amazon.com/Not-Just-Money-Economic-Motivation/dp/1858988454

Not directly related to education, but there's the example from Freakonomics about the day care center experiments - a general lesson on blindly introducing monetary incentives:
http://www.steve-lacey.com/blogarchives/2006/02/a_freakanomics.shtml

These are all in contrast to something like a carbon tax - that type of incentive that doesn't demand adjustments in moral judgments or qualitative activities.

Posted by Steve C | February 16, 2008 11:42 AM

Frankly, this is an unbelievably *terrible* idea.

Once you set into place the idea that lots of students need to be paid or whatever to attend school or study, then you'll convince all the ones already doing so anyway that they're total suckers unless they get paid also.

And once things soon settle down, you'll need to start raising the pay/rewards endlessly just to keep getting those better results.

This is exactly the sort of monetarization-of-everything that has been a totally disastrous force in so many sectors of our society, from CEO salaries down to those hordes of bloodthirsty but incompetent mercenaries in Iraq.

As far as I know, virtually no country in the entire history of the world has ever had to pay/reward students to go to school and learn, yet they always have, including throughout America's entire previous history. In most Third World countries including China, schools charge fees, yet parents scrimp and save to get their children an education.

I'm afraid all the most totally moronic "neo-liberal" craziness in DC is not merely on the Republican side of the isle...

Posted by RKU | February 16, 2008 11:48 AM

I teach 8th grade American History at a Title 1 qualifying school in Central FLorida. I absolutely agree. Many of my students would respond to money as an incentive. Offering money would make up for the lack of encouragement and sometimes active discouragement they receive at home from their families regarding education.

Posted by Charles Coffee | February 16, 2008 11:50 AM

"Meanwhile, educational attainment has positive externalities, so it's worth spending money on."

Actually we don't even have to base it on something as fuzzy as "externalities." Since schools are financed primarily by property taxes, we can look at the average amount of property taxes paid by those who 1) drop out of high school and 2) graduate from high school. Then we discount their probable lifetime property tax payments to the present. In this way, we can see exactly how much a marginal increase in payments now (as "stay in school" incentives) is worth, assuming we can calculate how many more kids will graduate based on such an incentive program.

Posted by RWB | February 16, 2008 11:53 AM

"Why not just go right to the source and given students direct financial incentives to do well in school?"

Not a bad idea, certainly worth trying. But of course the educrats will go bonkers at ANY proposal to give education $$$$ to anyone but themselves, so this isn't going to happen.

Posted by kafka | February 16, 2008 12:14 PM

Well, again, repeating what was said above...rewards need not be monetary.

For example, George W Bush, and most republicans for that matter, use quite a few nonmonetary rewards (because it's cheaper than good governance as well), so as to leave more cash on hand for his cronies. You know, the sort of people who crave status so badly they'll screw there own children for a chance to meet the big man and bask in his glory of Big Bossness.

Posted by shah8 | February 16, 2008 12:21 PM

I'm afraid I have an all-too-boring discussion of this at http://www.shermandorn.com/mt/archives/001152.html -- but there are some moderate fireworks on this at Ken DeRosa's blog:

http://d-edreckoning.blogspot.com/2008/02/dorn-takes-umbrage.html

http://d-edreckoning.blogspot.com/2008/02/dorn-takes-umbrage-ii.html

Well, to be honest, compared to a bunch of political blogs, those are probably tame, too, but you'll get a range of views across the three entries.

Posted by Sherman Dorn | February 16, 2008 12:58 PM

The great unstated assumption that derails so much discussion of education is the notion that (Educator Input)=(Student Performance Output). It's such a bizarre thing to think, that the performance of any individual students or group of students is purely or even largely the product of what the educator does. We were all in high school once; was that true for anyone? I did poorly, comparatively, in some courses because I was too tired and distracted from swim season; because I was too busy cracking jokes to perform in the class; or because I was just too damn lazy. And yet when we talk about education, over and over again, you see the assumption that the student's output is more or less dependent on the teacher's input. NCLB is founded on that very notion. But it's just plainly wrong, and anyone who's ever been a student should know it.

Posted by Freddie | February 16, 2008 1:00 PM

And the dumb get poorer...

Posted by Brian | February 16, 2008 1:05 PM

Paying for "performance" would be an open invitation to vicious pressure on teachers to inflate marks. That's bad enough as it is, thank you. My friends in the school system tell me that they already have their hands full with parents demanding their little dears get high marks for what the parents, in unguarded moments, will concur is less than stellar work.

Posted by sunsin | February 16, 2008 3:18 PM

In most Third World countries including China, schools charge fees, yet parents scrimp and save to get their children an education.

You are talking intact families able to instill drive and a sense of honor, duty, obligation in kids to succeed. The same sort of people that come to America and find the teaching and educational oppportunities in American inner city schools far better than in their homelands and who excel.

The problem is native children born of lazy parasites or children of immigrants affected by native slum kid's values.

Without any home-based motivation and discipline from parents, and all the "superteacher", "exciting school breakfast programs", "motivating sports", better school infrastructure, "higher teacher pay" making a dent - a straight reward program makes a lot of sense to overcome cultural and parental poisoning.

Particularly in elementary school where high performance may net a white or Asian student high rewards and status at home, but do nothing for a kid in the underclass. The time horizon is just too long to motivate them. Hard to listen to people saying that if they perservere unrewarded and unacknowledged for another 15 or 16 years - their best behavior and efforts will finally be rewarded with a good job and income.

And there is the problem of kids of the underclass who perform better than peers in grades and behavior being ostracized and punished with beatings by fellow classmates in school or out of school.

A reward system of money for doing well on mastery tests and good behavior and grades in elementary and junior high makes sense. That is the point you need to instill good habits to hopefully overcome cultural and parental deficits.

HIgh school is close enough to work or college that the motivation is likely obvious - continue doing well in HS and there will be scholarships, work study programs with cash in hand, other opportunities.

*************
Another problem is insistance by black politicians that ALL kids are capable of excelling in school. They are not. Programs may be more successful if the promising students are segregated from the hopeless. You may only be able to get 1/3rd or so of the underclass kids out onto a better future. But keeping the hopeless cases, congenitally lazy and stupid mixed in with the ones with the drive and potential to move up in life may exert the "crab bucket effect" where the promising ones are dragged back down by the knuckleheads and low potential females and you only get 1/6th escaping.

***************************
And yet when we talk about education, over and over again, you see the assumption that the student's output is more or less dependent on the teacher's input. NCLB is founded on that very notion. But it's just plainly wrong, and anyone who's ever been a student should know it.
Posted by Freddie

It was basically a teachers/school services employees union ploy for more jobs and more pay. You know the drill. Students ONLY failed to learn because they (1)didn't have prestigious teachers who would uplift them(2)teachers weren't paid enough and needed more Masters degrees in dysfunctional pathologies(3)More white kids were needed because blacks were incapable of learning without white kids in the same class(4)students didn't learn because course content wasn't "afrocentric" enough - at least half the heroes taught should be black, at least half of civilizations advances by "1st Negroes"(5)Students were deterred from learning by being in older schools that lacked a good baseball field and ski club...(6)Students were "too hungry" to learn needing free breakfast and lunch to enable them to think, and also were distracted by the obesity epidemic....

Posted by chris ford | February 16, 2008 3:34 PM

what is 'pure time preference'?

Posted by M | February 16, 2008 3:35 PM

For some years now, I have been advocating my own version of the "pay for performance" approach to motivating students. If I could persuade some philanthropic but quirky billionaire to fund it, I would open a nationwide chain of "Quiz Parlors", akin to the pinball arcades of old.

Kids could come in to the Quiz Parlors and take tests -- for money. No pass/fail criteria, just so many dollars per correct answer. That's important, because the point is not to judge, but merely to reward, competence.

We would start with poor urban neighborhoods. That would either: limit the number of rich suburban kids flooding in to win all the money; or result in a healthy increase of interaction between the two kinds of kids.

We would NOT do this in cooperation or consultation with the local school system -- that's why we need a billionaire philanthropist. The point is not to change what the schools teach or how they teach it.

I would pitch the idea to our hypothetical billionaire philanthropist as a sort of "retail scholarship program". Instead of offering students the far-off incentive of paying their college tuition, someday, we would offer them the chance to earn it, one correct answer at a time.

-- TP

Posted by Tony P. | February 16, 2008 4:20 PM

Re: As far as I know, virtually no country in the entire history of the world has ever had to pay/reward students to go to school and learn, yet they always have, including throughout America's entire previous history.

RKU,

I dislike the commodification and monetarization of everything in our society more than most people. But whatever we've tried in education for the last few decades has not been working. I frankly have no idea what we ought to do.

I don't know of any societies that _pay_ children to go to school, but there are a number of developing countries where the free lunch/breakfast programs serve as a kind of material incentive to parents to send their kids to school- this is widely considered a good idea in international development these days. There may even be a few countries that pay families to send their kids to school, I'm not sure. Also, there are many developing countries that don't have school fees.

Posted by Hector | February 16, 2008 4:56 PM

Freddie has it exactly right. You can lead the horse to water...

I can only relate to my own twelve years of Catholic education. As amazing as it may sound, I didn't actually know what school was for until about a year before graduation. Some things piqued my interest and I did well in those subjects. Some things I had no interest in. And I never cared about grades. I actively tried to sabotage teachers I didn't like, just for the fun of it. I did have a few teachers who were authoritarians, but they didn't make me actually perform better.

The incentive of money would have totally changed my outlook. Sure, the dumb would be poorer. Best lesson of all. Welcome to life.

Posted by shecky | February 16, 2008 5:26 PM

If the reward were big enough, like $500 for a 4.0 GPA, $100 for a 3.0 or better, and nothing for less than 3, I bet the poor kids would work harder than rich ones, and poor parents would push their kids the hardest.

Posted by Aatos | February 16, 2008 7:16 PM

If there were an understandable incentive for kids to do well in American schools, I'd be teaching in America and not Korea. Here I usually know mommy and daddy will ensure that their proxy in competition with their friends works hard.

Posted by aleks | February 17, 2008 12:02 AM

Through reading Sherman Dorn's (great) links above, I was shown to this article which I like very much:

http://www.aft.org/pubs-reports/american_educator/issues/winter07_08/scientist.htm

The takeaway is that targeted, well-designed reward systems can be helpful, but this dude is skeptical that blanket reward systems can be very successful.

Posted by mk | February 17, 2008 1:10 AM

To take one fascinating excerpt from the link above:

But for the children who were promised a reward, the reason is less clear. A student might not remember that he drew because he wanted to draw, but rather he remembered really wanting the certificate. So when the markers were available again but no certificate was promised, the student may well have thought "I drew because I wanted that certificate; why should I draw now for nothing?"

In other words (and going back to a different example): monetary rewards might confuse a student's sense of why they read.

This is wild, because I might have expected the money motivation to work the other way, namely: "I already liked reading... I like money... reading = money... therefore I really like reading!"

But this research suggests that different sources of motivation can compete in something like a zero-sum game. You can "wash out" a kid's intrinsic motivation by giving too much reward.

I find this absolutely wild. I wonder if the same is true of adults.

I also wonder if the trick to building motivation is to give kids external motivators that are hard to concretely remember after the fact. If "reasoning about my past motivation level" is really a strong driver of my future motivation level, maybe we should try to "trick" this reasoning to heighten their perception of their own past motivation level.

How might you do this? I dunno. Promise the kid a pizza if he does well on a test, and then when he does well, give him an anchovy pizza and say "that's all we have"?

I doubt it would work, but maybe something would.

Posted by mk | February 17, 2008 1:35 AM

Extrinsic motivators wash out intrinsic motivators?

Oh dear. I think we may need to take this concept a bit farther. We tell folks they should work toward getting into a career they'll love... guess we should stop offering them salaries or wages. We might be destroying their love for the job by offering them money for their work!

---

Offering money for learning is rather impossible. But offering money for work... including work that requires one to demonstrate learning... that makes sense to me. Only because we *require* students to attend school throughout their childhood, and that mandatory characteristic tends to wear at intrinsic motivation regardless of other factors.

Posted by salient | February 17, 2008 3:43 PM

From my perspective as a teacher in public school whose population has one of the largest percentages of kids qualifying for free lunch in the state:

Most of these comments seem to be written from the perspective of folks who attended good public schools or private schools - where kids entered reading at or above grade level and consequently the role of the teacher was to assign work, which students mastered fairly independently. In that situation, student effort = student success, and teacher effort was not a causal factor.

The key point you all seem to be missing is that in inner city schools, students don't enter with a high literacy rate. True, in these situations students often present as unmotivated... however this is often the RESULT of low basic skills - in addition to the factors you cite, such as a lack of parent support, etc. Furthermore, the rigor of the curriculum in these schools pales in comparison to the rigor of curriculum in middle/upper class public or private schools. Here's where teacher effort can (and is often the only thing that can) make a difference - it doesn't matter how motivated a student is if he/she does not KNOW how to be successful in school: how to read at or above grade level, write fluently, or think critically.

Consequently:
1) $100 might motivate a child to try to study for a test, but it doesn't address the underlying problem facing our schools: if a child can't read well, it will take hours of instruction to teach him/her how to do so at a level even approaching grade level.
2) Assuming the $100 is a sufficient motivator for students to pass the watered down curriculum at failing public schools, it does nothing to address the fact that a diploma from this school does not prepare the child to be successful at an institution of higher learning... nor does it provide the child with the necessary social/cultural background knowledge to compete with his/her private school counterparts at said institution.

Therefore the $100 invested in a child at this point is simply a band-aid,and i have serious doubts about the long-term success of a program like this. It neither addresses the underlying cause of low test scores NOR does it address the long-term goals of our public education system: to produce students capable of making valuable contributions to society.

Sure, if you taught kids how to be successful in school, financial incentives might motivate them to apply these skills. But, the success of many charter organizations and rockstar public school teachers indicates that once you teach low-income kids basic skills and show them HOW to be successful, you don't need to pay them to do well. Every year, KIPP academy schools send 8th graders (who entered in 5th grade way below grade level) to Deerfield - not because of financial incentives but simply because of good teaching that started with basic phonemic awareness and ended with shakespeare.

There are certainly better uses of funds than paying kids to do well. How about performance incentives for high-functioning teaching teams? Get the best people in the most underperforming classrooms and provide financial incentives for them to teach students how to read, think, write and solve problems at high levels - beyond the paltry expectations set by state standards.

Posted by kb | February 17, 2008 7:50 PM