Network News

X My Profile
View More Activity

Money matters -- even for teachers

A new study out of Vanderbilt tested whether giving teachers big bonuses if their students showed improvements would improve outcomes over three years. The answer? Not really. Linda Perlstein (in a very good post) isn't surprised. "Presuming that merit pay alone would elevate student achievement makes sense if you assume teachers have a hidden trove of skills and effort they are not unloosing on their students only because they lack the proper incentives to do so," she writes.

I'm more skeptical of merit pay than a lot of the people I know, but not for this reason. Consider the sentence "presuming that bonuses would elevate investment banker achievement only makes sense if you assume investment bankers have a hidden trove of skills and effort they are not unloosing only because they lack the proper incentives to do so." It works about as well. But actually, bonuses and high salaries appear to work in many professions. They do that in part by changing the incentives for people in the system now, and partly by bringing new people into the system.

There's a tendency to think of teachers as being somewhat more altruistic than people doing other jobs, but there's no real reason to believe that. Consider that doctors, who also have a fairly saintly reputation, are perfectly clear about their desire to be at the top of the income bracket, and make perfectly clear that if we start paying them less money, they will end up doing a worse job.

I agree with Perlstein that the evidence on merit pay is much more mixed than its supporters suggest, but at the same time, I think we do ourselves damage by assuming that teachers are operating at some sort of maximum efficiency level driven by care for their students. If we instituted a policy assuring every teacher whose students were in the top half of their school a cool million every year, I have no doubt test scores would improve over time. That's unaffordable, of course.

But it's not saying anything bad about teachers to suggest that the possibility of wild riches might make them, like everyone else, somehow increase their effectiveness, or might attract more effective people to the field. I do wonder, however, if we focus much too much on pay incentives (a bonus for helping your students) as opposed to pay levels (doubling teacher salaries).

Update: Incidentally, this observation goes both ways. A lot of the people who think money is important for teachers also hate teacher's unions. Those unions, of course, make teaching a more attractive job with better wages and conditions, and that should increase the quality of the talent. Now, it so happens that teaching isn't incredibly lucrative, but that's a question that more about the pay level than about whether unions impose too much pay equality.

Also, I made some slight edits to the post above to make my point about salary levels a bit clearer. An e-mailer then pointed out that Perlstein's comment was quite narrowly tailored to the issue of merit pay, which is also correct. So consider this more of a general post on the utility of thinking about money as a useful motivator for teachers rather than as an argument with Perlstein.

By Ezra Klein  | September 22, 2010; 3:29 PM ET
 
Save & Share:  Send E-mail   Facebook   Twitter   Digg   Yahoo Buzz   Del.icio.us   StumbleUpon   Technorati   Google Buzz   Previous: Did Summers really leave to return to Harvard?
Next: Looks like we're getting an Abramoff movie

Comments

It's important to remember, too, that this sort of incentive is designed to operate not only on current teachers who might be withholding their talents in lieu of higher salaries but also as enticements for non-teachers, especially college students and other young proto-professionals, who would otherwise take their talents to the more lucrative fields of law, finance, medicine, and the like, to consider jobs in teaching. If smart, ambitious college students start to understand that their superior skills will be met with additional remuneration in the teaching profession, they may well be willing to give that profession a second look. Have there been multi-year studies tracking the effect of performance incentives on recruiting teachers of high quality, and not merely enhancing the output of those already established within the profession?

Posted by: alevitov | September 22, 2010 4:15 PM | Report abuse

For existing teachers, the bonuses are a windfall; they already signed up as teachers knowing the pay wasn't great. The real measure will be the new people who enter the profession after the bonuses are in effect.

My mother was a teacher and always discouraged me from becoming one - not because she didn't love her profession, but because the male teachers she worked with painted houses in the summer.

There's no reason for anyone to assume I would have made a great teacher, but the disincentive is there.

Posted by: Jason45 | September 22, 2010 4:17 PM | Report abuse

You might want to read what Dan Ariely has written about "social norms" versus "market norms."

I can't say it is true for all schools, but as the husband of a public school 7th grade teacher (and brother or brother-in-law to four other public school teachers), I can tell you that "social norms" (i.e., a culture of giving your best) produce amazing results. By contrast, what Dan Ariely has pointed out is that when you attempt to inject a "market norm" you may actually erode the existing social norms.

Here is an example: ask a friend to help you move some furniture this weekend. Acting in accordance with social norms ("friends help friends"), your friend will probably say "yes." Now, ask another friend if he or she will help you move furniture this weekend for $2.00. By offering money, you have moved the transaction from "social norm" to "market norm." You have offered substantially below minimum wage. Your friend will almost certainly say "no." In other words, what a friend is willing to do for free, they will not do for low pay.

The teachers I know work 50-60 hours a week during the school year. They could work fewer hours and collect full pay. They are acting on social norms that are based on wanting to do their best.

If you are going to use "market norms" to boost teacher (and student) performance, you can't "do it on the cheap." Assume a teacher can "get by" working 40 hours a week. Assume a really good teacher (one who draws forth top student achievement) puts in an extra 10 hours a week. Assume 40 weeks of school per year. The "good teacher" in this example works 400 hours per year more than the teacher who is "just getting by." If your bonus is going to be, say, $2,000, you are paying $5/hour of additional time. That is well below minimum wage. The message you may be conveying to the "good teacher" by relying on a market norm is that the extra time they are putting in is not highly valued. You may very well get the opposite of the result you want.

Posted by: bearclaw1 | September 22, 2010 4:18 PM | Report abuse

Two points:

1) Your hypothetical looks to be the dynamic in Wall Street finance. Is that REALLY a great argument on behalf of merit pay, to make the parallel with the screwed-up incentives that led to a financial crisis?

2) There's a substantial literature on incentives in industrial-organizational psychology, and it's not consistent with the picture you're describing of loads of industries improving productivity with bonuses. Bonuses exist for a variety of reasons, and they may be good or not depending on the structure, but that doesn't mean that they improve productivity. (It's somewhat disturbing that as an historian, I know the relevant literature more than self-identified wonks. That shouldn't be the case.)

Posted by: ShermanDorn | September 22, 2010 4:19 PM | Report abuse

Ezra, I agree that giving individuals incentives (money or otherwise) to act in certain way would affect their behavior, but I'm not sure it makes them "better" in any meaningful way. Giving Wall Street traders millions of dollars to meet certain targets caused them to become better at--meeting those targets. If that meant leveraging the firm to absurd levels, moving risk to off-balance vehicles, sinking the entire economy, it didn't matter. They didn't become "better" bankers, they just anything they could to meet those targets.

I'm afraid, in the near term, that's what will happen if teacher salaries drastically increase. Teachers won't become better teachers, they'll just find new ways of teaching to the test, or whatever other metric is used to measure their performance. Kids may become better at memorization or mnemonic devices, but general learning and creativity will be absent. In the long run, however, better pay will attract better people to the job; I just hope their incentivized to actually teach

Posted by: JonM316 | September 22, 2010 4:23 PM | Report abuse

My son in high school has seven classes every day. We are keeping track, and so far every day this year (four weeks) at least one of his teachers has been absent. Maybe some teachers work 50 - 60 hrs per week, but not the ones at my son's school. Bonuses might get them to show up more often.

Posted by: cummije5 | September 22, 2010 4:26 PM | Report abuse

The issue is that teachers are often paid below what others with similar amounts of education are making. If you paid them more you would get a larger pool of applicants including more qualified ones who would raise test scores. Similarly if doctors were paid less a smaller pool of qualified applicants would be willing to subject themselves to 6 years of medical school to go into the field and the quality would go down. The problem is you need positive incentives to get better potential teachers to become teachers and you need the ability to fire bad teachers. When reformers talk about 'merit pay' they need to have proposals that make clear that it is dependent on the teacher, not the financial situation of the school district and it will stay high enough to remain competitive with the private sector. The people you need to convince are qualified people in their 20s and 30s who could take teacher certification courses or an MAT and replace the bottom say 10% of teachers in the country over the next generation.

Posted by: tmorgan2 | September 22, 2010 4:31 PM | Report abuse

Interesting study. A few comments:

Why is it always standardized test scores? The best teachers I had in High School taught Physics, Pre-Calculus and European History. You wouldn't be able to tell on my proficiency tests whether or not I knew much of anything about Physics or European History. For that matter, you can get a lot out of a class without necessarily knowing all the answers to some test developed by a random bureaucrat. Some students are just lousy test takers, and no amount of effective teaching is going to register on a standardized test.

The idea that "increasing teacher pay via merit bonus has no effect" might imply another conclusion - that reducing teacher compensation wouldn't harm student achievement. In fact, I'd hazard a guess that for many teachers it would make no difference (though a sharp cut might lead to a spite-based drop off in performance). As you say, we probably aren't at some magic efficiency level for teachers - cutting the starting base compensation for new teachers will probably make no difference in quality.

Here is another paper that might be interesting:

http://www.montana.edu/ttt/pdf_files/Pulled_Away_Pushed_Out.pdf

It discusses whether high aptitude teachers were either pulled away (by higher salaries outside of teaching) or pushed out (by low salaries due to wage compression).

The proportion of high aptitude teachers has declined from 5% to 1% since 1960. A full 80% of this decline can be explained by wage compression (basically meaning that union wage schedules meant that high aptitude potential teachers fell further and further behind what their high aptitude peers were earning in other fields).

If the authors are correct that teachers with high academic aptitude are typically better teachers, then perhaps the Vanderbilt study missed the important factor.

Peformance based pay could attract more high aptitude college students into education majors, though it would take many years before this effect became noticable in terms of more teachers with high academic aptitudes.

Of course, we've been trying to figure this thing out for decades. We've funneled ever more money into the education system and tested all sorts of theories and we have little to show in terms of results. Maybe there is no silver bullet. Maybe there aren't any top down solutions that will fix things, and we should let schools manage themselves consistent with the goals of the parents who pay for their kids to go there.

Posted by: justin84 | September 22, 2010 4:32 PM | Report abuse

"I think we do ourselves damage by assuming that teachers are operating at some sort of maximum efficiency level driven by care for their students."

But who the hell makes that argument? No one I know. It seems like when it comes to education, you are opining in an evidence-free zone. What we do know is that social norms in teaching are extremely important (as bearclaw pointed out above), and that good pay, good working conditions, and strong teacher networks where experimentation and sharing of expertise are supported and encouraged lead to better outcomes for students. These things are challenging to institute and not sexy. By contrast, "merit pay" as the solution du jour makes intuitive sense to people with b-school backgrounds and gives the impression of a quick fix.

Posted by: tediouspedant | September 22, 2010 4:33 PM | Report abuse

*The best teachers I had in High School taught Physics, Pre-Calculus and European History. You wouldn't be able to tell on my proficiency tests whether or not I knew much of anything about Physics or European History.*

What? Don't be ridiculous. You could take the Physics SAT II test or the AP European History exam, and we'd have a pretty good idea about how much information you absorbed from those classes.

The problem is that poor school systems have good teachers desperately to get out and into better school systems and we have smart, ambitious young students who will avoid going into teaching because of the low compensation and structure which makes the work less than intellectually rewarding.

Posted by: constans | September 22, 2010 4:48 PM | Report abuse

The idea of merit pay isn't wrong because it mistakenly assumes that teachers have a hidden trove of skills they would draw on if only they were paid more. It's wrong because it implies that teachers would work harder and be more dedicated if you gave them more money. As most 'market-based' solutions do, it assumes the target institution is really just a poorly-run McDonalds, and proceeds from there. There are many inter-dependent factors at play when a student fails to learn, and pumping one of them full of money probably won't solve the problem.

And I think if you're going to continue these posts that equate higher test scores with increased student learning, you should start including some evidence that it's true. In my experience, higher test scores are fine, but they are absolutely not indicative of a student learning more in the way we want kids to learn when they come to school. Maybe I'm wrong, and maybe all my students whose math scores went up last year but who still can't solve a basic story problem are in fact learning more, but I'd like to see some objective proof of that from other sources before I pat myself on the back and ask for more money.

Posted by: andrewbaron78 | September 22, 2010 4:49 PM | Report abuse

cummije5,

I'm sorry your experience with your son's high school has been disappointing. Want to see what a dedicated high school science teacher can do? Spend some time at www.sciencegeek.net -- my twin brother's website, which he started just for his high school science classes, but is now used by teachers around country.

Posted by: bearclaw1 | September 22, 2010 5:23 PM | Report abuse

Some of my friends just raised about $900 to give to our friend, a teacher. In spite of having a ton of school debt, she was plunking down a lot of her own money to buy stuff for her classroom. Not toys and trifles: books published after 1970, pencils, paper, etc. Obviously, if she had more money, she would be doing more.

This is very common practice among teachers, so I disagree with Ms. Perlstein's read on what proponents of merit pay believe about teachers. It's not that they are necessarily hoarding these skills out of meanness or laziness, but that teachers do a lot of things for their kids with their own money, so giving them more money might be reasonably expected to affect outcomes. Of course no one can say that out loud because we don't want to talk about how teachers spend their own money on their classrooms because schools are underfunded. :)

My personal feeling is that there's a lot more we can do for our kids in terms of improving efficiency of teachers (anyone who has listened to the This American Life "Rubber Room" episode can think of some stark examples). On this side of the equation, I really consider the teacher's unions to be more of a blight than a benefit. It's not that teachers don't need unions, or that unions are all bad, it's that I think they've gone about protecting their constituents in the wrong way, and for too long.

The other side of this though is that our school years are too short. And we don't invest enough in early childhood education and parent programs. Improving teacher performance compared to these things is, in my view, the area where there is the least amount of efficiency to be made up. Teachers need more support from the state in regards to early education and after-school programs, funds for longer school years, and much as no one wants to hear it: parents.

Posted by: roquelaure_79 | September 22, 2010 5:58 PM | Report abuse

This finding shouldn't be surprising, given what we know about money as motivation for cognitive tasks. You linked to this video a few months ago:
http://comment.rsablogs.org.uk/2010/04/08/rsa-animate-drive/

Basically, the optimal amount of money seems to be "enough so that it's not an issue". Beyond that, fostering a sense of ownership/involvement/meaningfulness is the best motivator.

Posted by: adamguetz | September 22, 2010 6:08 PM | Report abuse

"What? Don't be ridiculous. You could take the Physics SAT II test or the AP European History exam, and we'd have a pretty good idea about how much information you absorbed from those classes."

But would you have a good idea of how much information I absorbed from the teacher, or how much I absorbed from the book?

All I was saying here is that I didn't take things like the SAT II for Physics. I took a state proficiency test and the regular SAT. You'd have no idea how effective my Physics teacher was based on those tests.

But let's suppose I did take the Physics SAT. Would a good score have been because I had a great Physics teacher, or because the school had great Math teachers and I could succeed despite the mediocre instruction in Physics? Would a bad score have meant that I had a lousy Physics teacher, or was I just bad at math? Should we subject students to a standardized test for every single class that they take? Should that be a top down decision for the entire country?

AP courses select for strong students, and it's not as if there is a series of AP tests in a subject, there is only one. If most of a teacher's AP kids pass a test, is it because of effective teaching, or because the school was effective at weeding out kids unlikely to succeed?

Say someone takes an AP Macroeconomics test and gets a five without taking the class. He/she might have absorbed a whole lot, but they would have gotten a five whether the class was taught by a rock or Austan Goolsbee. Others might get a one despite the efforts of a great teacher. Clearly, all students involved are probably better off sitting in a class with Austan Goolsbee than in a room with a rock, despite the fact that for many their particular AP scores won't be affected. Goolsbee might impart some wisdom the test doesn't happen to capture.

Culture matters too. There are schools in which you are a dork (or worse) if you try to learn and get good grades, and there are schools in which you are a moron if you don't get straight As. The exact same teacher might look mediocre in the first example and poor in the latter. In a school where most kids think they need to do well, the extra effort by a teacher might not translate to much better test scores - those who are already capable and want to learn are doing so, though they might still benefit in other ways. Likewise, in the school with an anti-learning culture, more effort from the teacher is good for the several kids who are trying to learn despite it all, and those few kids continue to have strong standardized test scores, and the remainder continue to perform poorly.

So in a nutshell, the point is that standardized tests don't capture everything (either because a subject isn't directly tested, or because of spill over effects, or because it measures something internal to the students, etc), but despite that, people still have a good idea who the most engaging and effective teachers are.

Posted by: justin84 | September 22, 2010 6:17 PM | Report abuse

"Incidentally, this observation goes both ways. A lot of the people who think money is important for teachers also hate teacher's unions. Those unions, of course, make teaching a more attractive job with better wages and conditions, and that should increase the quality of the talent. Now, it so happens that teaching isn't incredibly lucrative, but that's a question that more about the pay level than about whether unions impose too much pay equality."

Let's clarify a bit.

Unions raise the average wage, but compress the spread. Wages are determined by how long a teacher has served and whether or not one has a master's degree, regardless if that degree has made one iota of difference. It's not just the pay level. Most probably don't need a whole lot (as long as it's enough, per an another poster) but really rewarding the standouts might get more standouts into the business.

And conditions? We aren't talking about coal mining here. In any case, the relative difficulty in disciplining kids these days probably does more to damage conditions than anything an particuarly evil school board might have up its sleeve.

So instead of a spread of $25,000 - $100,000 and an average of $45,000 perhaps the wage spread is $35,000 - $70,000 with an average of $50,000. More over, there is no way anyone at age 30 will be earning that $70,000, that won't come until age 55.

Anyone who is unwilling to wait for that $70,000 at the end of the road is probably not going to consider being an education major, even if the job itself sounds interesting. No amount of effort on their part would make a difference.

More troubling is that unions are pigheaded in tough times and often fight for raises when tax revenues are falling, the end result often being motivated young teachers getting fired. This is bad for students both because there are fewer teachers and because lots of the young ones find the work new and exciting, whereas a a decent portion of older teachers (not saying all/most, just a portion) are jaded, but hey why not stay on for another 10 years because the retirement bennies are so good, and why work hard because it's damn near impossible to be fired.

Or suppose the unions succeed in getting a levy through - now the struggling taxpayers, many without work and some at risk of losing their homes face higher property taxes.

Unions also stifle initiative. I know a teacher who wanted to start an after school science club, but the union didn't allow it because the school didn't have the money to pay for the extra time. They couldn't have teachers working for free for an hour or two each week - it would other teachers look bad, and we couldn't have that!

I could be wrong on all this, but don't worry, I don't want to impose anything on any school district at any level of government. Those in charge of the school should be free to run it as they see fit, with the goal of satsifying their customers (parents/kids).

Posted by: justin84 | September 22, 2010 6:55 PM | Report abuse

I'm a high school teacher and I have finally come to the conclusion that I'm not paid poorly. I teach outside Boston and have a MA. After 10 years of teaching both my wife and I make approximately $70,000 each, have an 8 week summer vacation, 2 weeks for Christmas, one week off in both February and April.

With my 10 years of service and the seniority system I have almost no fear of losing my job. I can't quantify the value of this lack of anxiety but it is worth a heck of a lot to me.

Lastly, I hope (if the State doesn't alter the deal) have a great pension where I will earn 80% of the average of my last three years of teaching.

Given all of that I don't really complain anymore about my salary, especially in this economy.

On top of that I love my job and spend it with highly educated, intellectually curious coworkers. And in the classroom I get to engage with young people about ideas concerning US government, economics and modern world history.

I hope more people start to see teaching as the great gig it is and we attract more of the best and brightest. It is hard work, but worth it.

Posted by: seospider | September 22, 2010 8:00 PM | Report abuse

I'd guess for the cool million we would get widespread fraud. At least you would get universal teaching to test.

Odd that you mention incentives for investment bankers in a post praising merit pay. Do you really think those incentives caused them to do a better job than they would have done if they worked for a flat salary ? The lesson I take from finance is that huge incentives combined with poor measures of performance lead to huge recessions.

You only very briefly mention the effect of bonuses on retention of able teachers. I suspect you are differentiating yourself from your friend Matt Yglesias.

Clearly if the issue is keeping able teachers from quitting, we should praise the teachers unions. They lead to higher pay, but also, most people say really like to be represented by a union (well actually most people say they really wish they had a union but that they don't believe in effective enforcement of the Wagner act or in the tooth fairy).

Posted by: rjw88 | September 22, 2010 8:26 PM | Report abuse

The thing that's tiring about the merit pay discussion is not just that it assumes that teachers are, on the whole, complacent about performance. Its that it also assumes that the really bright people with the answers are bypassing education altogether.

The argument seems to be that the answers to our education problems are out there, but the incentives to find them are lacking. Evidently people would look for a way to raise our children out of ignorance if the money were right. Yeah that's it. Either, current teachers are holding out or the straw-people with the answers are. There are people who are education experts who can't be bothered to share there ground-breaking understanding of pedagogy because the compensation isn't adequate.

I suspect that this is wishful thinking. The answers to our problems are probably not as simple as a standardized test score. Probably part of our problems are systemic, but there are a lot of people who have dedicated their lives and reputations to improving the system. We have a large gene pool of educators and educational idea people. That may not be out biggest problem.

I grew up with public schools in the 70's and work in public schools now. Compared to what I grew up with, students are presented with a buffet of education staffed by waiters who are pressing each and every dish. If you go away hungry in such a situation, maybe it's not the restaurant.

Posted by: Droffas | September 22, 2010 10:18 PM | Report abuse

cumije5,

i've noticed the same thing with each of my kids this year too. My son's first grade teacher has been out 3 times this year and they've been in for 3 weeks. My daughter's 4th grade teacher has been out 4 days. My daughter's 9th grade teachers haven't been out nearly as much. I don't know if there's any rhyme or reason to it or if it will continue but I'll see and I'll definitely be watching and paying attention.

I also know many teachers who are from two parent's working families and many of them go to all the school functions (at different schools) and take off for that. Sometimes I can take off for that sometimes I can't. Maybe its a jealousy thing on my part.

Posted by: visionbrkr | September 22, 2010 10:24 PM | Report abuse

Those unions, of course, make teaching a more attractive job with better wages and conditions, and that should increase the quality of the talent

I disagree, Ezra. Unions make teaching a more attractive job for poor or mediocre teachers, but not for good ones, who would be able to demand better conditions and pay because of their talents. Instead, they get paid the same mediocre salaries as everyone else, or are working in another profession.

Posted by: brickcha | September 22, 2010 10:50 PM | Report abuse

The bonuses studied were poorly designed because they were not longitudinally based. What you have to measure is how much students improve in a teachers' class over the course of a year, not how this year's students compare to last year's students.

If you ignore the baseline, you are measuring student quality, not teacher quality, so it is unlikely that *anything* a teacher could do, except to politic for better students, would affect his or her rating.

So, garbage bonus plan, garbage study.

Posted by: staticvars | September 22, 2010 10:54 PM | Report abuse

"The argument seems to be that the answers to our education problems are out there, but the incentives to find them are lacking. Evidently people would look for a way to raise our children out of ignorance if the money were right. Yeah that's it."

Droffas,

The point is we just don't know. In most industries, pay for performance seems to be beneficial.

Maybe it's not paying for higher standardized test scores (I'll agree teaching to the test is not something that should be encouraged). Maybe it should be subjective performance evaluations. Maybe performance awards should be team based. Maybe money doesn't make a bit of difference, but this in turn suggets we could reduce compensation over time and get the same quality for less money.

"There are people who are education experts who can't be bothered to share there ground-breaking understanding of pedagogy because the compensation isn't adequate."

It's not necessarily a new method (though maybe it is), but getting the most jaded teachers to try a little harder, and getting more bright energetic people into the classroom to begin with.

In any case, it should be up to the team running a particular school to structure compensation with an eye to what's best for both the kids and who is paying the bill, rather than a compressed union compensation structure complete with tenure and a first in last out employment structure. That *can't* be the best for kids.

"I grew up with public schools in the 70's and work in public schools now. Compared to what I grew up with, students are presented with a buffet of education staffed by waiters who are pressing each and every dish. If you go away hungry in such a situation, maybe it's not the restaurant."

Some public schools do provide great educations, no question. But many surely don't, or else this wouldn't be an issue. I don't buy the theory that the current generation of kids just happens to be particularly dense despite all around amazing schools.

Furthermore, there are plenty of examples of public schools which aren't "buffets of knowledge" but are best described as day care centers. Without competition and new ideas this is never going to change.

Posted by: justin84 | September 23, 2010 10:06 AM | Report abuse

Post a Comment

We encourage users to analyze, comment on and even challenge washingtonpost.com's articles, blogs, reviews and multimedia features.

User reviews and comments that include profanity or personal attacks or other inappropriate comments or material will be removed from the site. Additionally, entries that are unsigned or contain "signatures" by someone other than the actual author will be removed. Finally, we will take steps to block users who violate any of our posting standards, terms of use or privacy policies or any other policies governing this site. Please review the full rules governing commentaries and discussions.




characters remaining

 
 
RSS Feed
Subscribe to The Post

© 2010 The Washington Post Company