« Sexy Feminism | Main | Food Blogs »
October 17, 2007
Pay For Performance
I'm always amused by well-paid journalists and pundits complaining that teacher's compensation isn't closely enough linked to performance. Is Megan hauled into James Bennet's office once a week, presented with updated traffic numbers where traffic boosts and drops are disaggregated from intra-Atlantic links and general noise, and then paid less or more depending on her performance? Of course not. Andrew Sullivan doesn't even trust his traffic numbers. Are pundits, once a year, asked in for a sit-down with their editors, during which they are presented with a comprehensive list of their predictions and opinions, and then offered a bonus based on accuracy? Does The Atlantic pick through reader mail and commission surveys to evaluate how much interest is generated by Clive Crook's work, and then reconsider his contract based off those metrics? Of course not.
Indeed, not only are these practices not in place, but journalists would sob and shriek and scream were they ever to be implemented. Our work isn't supposed to generate traffic, it's here to inform; our work isn't merely to inform, it's here to provoke and spark interest; our work isn't merely here to provoke and spark interest, it's here to add to the paper's reputation and cover important topics; blah blah blah. There's no merit pay in journalism, and no agreed upon metrics measuring quality. There's no way to track improvement, and no agreement on what improvement would mean, or how to disentangle individual effort from luck, chance, personal relationships, and so forth. But no end of journalists are happy, when talking about teachers, to demand merit pay, and performance tracking, and all other sorts of incentives they're neither exposed to nor would support.
By the way: I support merit pay (and for journalists too: How much the better if we could boost our salaries by doing graduate work in our subjects!), but if you've never thought through the difficulties of implementing such a program, you might want to read through these comments.
October 17, 2007 | Permalink
Comments
Actually I think the Gawker sites do pay based on traffic, so it's not unheard of; and frankly, shouldn't McMegan and Sullivan and others hired essentially to draw traffic get paid (or some portion) for performance? That seems a good deal for them, never mind the company that hires them. And it provides incentive to actually pay attention to what interest readers (something Sullivan, at least, ought to heed)... which is to say, your disdain is principled... but I'm not sure it's the best argument you got against pay for perf.
Posted by: weboy | Oct 17, 2007 12:47:00 PM
But here's the thing: I actually am in favor of pay for performance. The problem is, I think it's very complicated to implement. For instance: Do it just on traffic and you trivialize the content (if my every post were long, wordy lashings of conservatives, a la my hit on Malkin, I'd get 25,000 visitors a day). So how do you account for covering necessary subjects which may not be as widely desired? And what about accuracy? Who's checking? Does it matter? And does a traffic boost because people are linking to something stupid you said count differently from one that comes from a brilliant insight? Or a harsh takedown?
Etc. I think some bonuses for traffic increase are fair, as for frequency increases (my salary does not take into account my output). I think increases in pay for developing expertise -- through further schooling, say, or fellowships -- would also be good. But it's tough to add these incentives without creating real costs. Doesn't mean it shouldn't be done, but it shouldn't be spoken about so blithely.
Posted by: Ezra | Oct 17, 2007 12:57:52 PM
I just love the double standards ...
teachers and schools perform a mission critical function in our society, therefore their pay and funding should be dependent on performance and benchmarks
soldiers and the military perform a mission critical function in our war on terror by being in Iraq, but don't you dare criticize any soldier (even a high ranking general shilling for the admin) or request there be benchmarks in that war or you get tarred and feathered as un-patriotic.
Posted by: DAS | Oct 17, 2007 1:35:25 PM
Merit pay for teachers is one of two things:
-From conservatives, it's just another attempt to hinder public schools from functioning efficiently. When are we going to fully understand that a movement dedicated to the idea that "government doesn't work" is going to actively impede and/or dismantle every governmental program they can in order to make their slogans true?
-From liberals, merit pay is a way to be contrarian - as Ezra has ably demonstrated - and/or a completely theoretical debate held by people who don't understand the real-world implications.
Teachers are supervised, you know, and there are several ways for parents and administrators to give feedback on a teacher's performance. Tenure can be a problem, but of course it's not nearly the major issue that certain elements in our society would like everyone to believe. Teachers that get additional certifications and/or degrees are paid more.
Yes, people in technical fields generally have to take pay cuts when they become teachers. But there are programs all over the country that will help those mathmeticians and scientists who are interested in teaching pay for whatever schooling they lack and will often allow them to start teaching while they attend school for their certification. That gives them the ability to switch careers without having to work student jobs or raise money to be unemployed for a year or so before teaching.
As for the difficulties in implementing merit pay, what would be the criteria? Standardized testing? Those tests are ruining our schools, creating an entire generation of people who can memorize and fill in little circles but who can't think. Never mind that some people naturally do well on those tests, and some don't. Making standardized testing the only way to assess how our schools are doing is counterproductive to say the least.
Now that NCLB has been around, we're seeing more and more special education kids being forced to take the same flippin' tests as everyone else - kids that would not have been mainstreamed in that way before George Bush's perfidious influence. For example, my daughter's elementary school has ITBS scores that are way lower than the district average. I was able to speak to the testing coordinator - yes, that is a full-time position schools have to fund now - and found out that if the school's special ed class scores are excluded, the average is right where the other schools are. It's just that my daughter's elementary school gets kids from all over the district because they have the facilities and personnel needed for that type of program. But in a merit pay situation, teachers at that school would be penalized, because there simply isn't a provision in the ITBS or other standardized tests that will allow for students' individual abilities.
Posted by: Stephen | Oct 17, 2007 1:40:38 PM
Ezra, I love ya buddy, but come on.
There absolutely *is* merit pay in journalism. Not every reporter makes the same salary. You think every journalist at a metro newspaper gets the same paycheck? You think Anderson Cooper makes the same coin as Rick Sanchez? You think magazines don't notice that some writer's pieces get huge responses and others get none? Or that some are brilliant and some write junk?
Once a year -- not once a week, which is a strawman idea no one is proposing for teachers -- we sit down with our bosses, and they go over the year we had. Then they decide whether we deserve a 2% raise, a 15% raise, or whatever. That's pay for performance.
As opposed to the teaching world, where in most schools the most brilliant AP Physics teacher gets paid the same amount as the most slackerly, disinterested PE teacher. Pay is based almost purely on credentialing and years of experience in most school districts -- both of which are poor metrics of quality.
Now, if the debate is over whether teachers should get raises based purely on their students' test scores, I'm with you -- that's a dumb system that'd be easy to game and that would misreward a lot of teachers. But it's silly, silly, silly to think that teachers should not be judged holistically and rewarded for good performance. That's what happens to journalists, and to just about every other profession in the private sector.
Posted by: Josh | Oct 17, 2007 2:09:34 PM
What Josh said.
Posted by: wisewon | Oct 17, 2007 2:19:00 PM
Josh, I'd be all for judging teachers "holistically," if my Dad, a wonderful elementary school principal beloved by both teachers and students, were doing the judging. But becoming a principal in most systems is largely a matter of getting another degree, not of having good judgment. I've seen principals who were simply petty tyrants and others who wouldn't know good teaching if it hit them in the head; in fact, they might be opposed to it. Moreover, really judging teaching fairly would require spending a fair amount of time in each teacher's class. Nobody's going to pay for that. So in principle, I'm with you, but in the practice of most school systems, I don't see it working.
Posted by: Dcounsel | Oct 17, 2007 2:23:44 PM
I don't see a lot of difference between "I've seen some bad principals, ergo we can't judge teacher quality" and "I've seen some bad bosses, ergo we can't judge employee quality."
There are stupid principals, stupid editors, stupid middle managers -- stupid people who rise to a degree of power in every industry. Does that mean we throw up our hands in every field of work and say, "Okay, I guess everyone gets the same salary?" Yet for some reason we do that in public schools.
Posted by: Josh | Oct 17, 2007 2:30:52 PM
The problem with "pay for performance" as evidenced by the State of COlorado is as follows. In 2000 Colorado decided the dropthe "step increase" (5% raise every year)and go with "pay for performance". We were promised the same amount of money would continue to be in the pot and that raises would be based on performance evaluations. We immediately went into a "wage and hiring freeze", so with our new merit pay system we got more work and no raises at all for three years. In 2005, we got a 2% raise. All this time health insurance premiums were going up, match on 401K was taken away, and our portion of PERA contributions increased. Basically we have lost pay due to PFP. And I have had a "peak performer" rating every year for 10 years.
This year I got the best raise I have gotten since PFP went into effect: 4% and a 2% bonus for "peak performer". Before paty for performance we got 5% step and about 4.5% catchup which was an effort to bring us to parity with the private sector.
All that money we were promised would still be in our raises has never once been there since PFP was instituted. That is what is wrong with merit pay, it is an excuse to eliminate raises for hard working public serveants.
Posted by: APISHAPA | Oct 17, 2007 2:41:15 PM
Josh, you make a reasonable point. I think the difference is that the stupidity of managers in many systems is tempered by reliable, concrete measures of performance, and indeed, the quality of management itself is subject to such measures. That's not true in education, especially since so much of teaching goes on beyond the eyes of supervisors, and the supervisory time necessary to assess adequately what goes on in the classroom is costly.
Posted by: Dcounsel | Oct 17, 2007 2:53:42 PM
I’ve made this point before – it’s been a while since this discussion’s come up – but implementing merit pay is not difficult if one takes the proper systematic approach. Define success, figure out how you measure it, and then go from there. This is something that’s done in Corporate America all the time and there’s no reason to think it can’t be implemented within out school system. Granted, this isn’t like a sales job where performance is easily quantified with bookings and deal margin; to me, it more like my field, consulting, where you need to measure on softer metrics because you have the unruly customer affecting your performance (customer being analogous to student in this case).
Aside from the fact that I think any organization should institute merit pay, it’s reasonable to argue that merit pay alone is not going to help improve the school system’s overall performance because I think it’s too dependant on the students, their families and their approach to education.
Posted by: DM | Oct 17, 2007 3:13:29 PM
I agree with Josh. Merit pay simply isn't what Ezra thinks it is. Even when managers try to use empirical methods it still mostly boils down to their ability to understand what their employees are doing and how valuable they are. And yes, office politics kick in.
The problem with merit pay in teaching is principals don't get to observe teachers in the classroom very often. But still, a good principal should know who the best and most valuable teachers are.
Posted by: Mark | Oct 17, 2007 3:28:49 PM
What's really hilarious is that the right-wing pundits who love teacher merit pay would not have a job, let alone merit pay, if they were judged by their performances during the Bush administration.
If, that is, truth were considered a meritable item.
Posted by: Barry | Oct 17, 2007 4:21:03 PM
Obviously, again, as my wife has repeatedly pointed out, this is just another example of teaching being one of those particular fields where everyone with an asshole seems to think that they are informed enough to spout off about it. Simply put, merit pay based on the performance of students is a bad idea, a bad policy, and an incentive towards grade meddling.
As a bad idea, it would be similar to paying dentists based on how many cavities their patients had. Students come to school more or less prepared for the material based on a complex series of factors including their socio-economic background, their family's cultural attitudes toward education, their learning disability rate, the stability of their home environment etc. Then once they get in the building, environmental factors there can likewise influence their performance.
For instance, my wife teaches in a district comprised of mostly affluent-ish white suburbanites with minority representation trending mostly toward Pacific Rim and Southeast Asians. Yet, because of poor money management issues foisted on the district by a crappy school board including a refusal to put up levies when necessary (and a general wholesale antagonism toward levies by the community), their per pupil spending is the second lowest in this area of the state. It's lower than the mostly poor, mostly minority urban school district. Honors classes exceed 30 students in almost every single case and regular classes exceed 35.
Now, as a matter of experience, I'm doubting most commenters here have figured out the appropriate way to not only provide adequate classroom management for 35 fifteen year olds, but also to adequately grade each and every students' work when you're teaching five classes of the same numbers. What kind of effective grammar instruction and composition advice do you think you could provide when you have 175 six page papers to read, as well as essay tests, short answer tests, vocab essays, etc.?
Now how should we arrange our merit pay metric? Test students at the beginning of the year, then give them the same test at the end to see how much they've learned? Do we measure the gap between early grade and later grade? If so, then won't teachers in more affluent districts (where kids routinely test better) merit smaller pay performance bonuses, since the kid who scored a 95 at the beginning of the year can only improve by 5 points? Whereas the kid from less affluent district who scored, say 65, made it up to 75, and thus we say that that is indicative of significantly greater improvement?
And how do we then measure actual improvement when Teacher A teaches toward the test and improves scores more than Teacher B whose students actually learn to think better but aren't better able to regurgitate the right string of facts to fill in the correct bubbles?
Or would you suggest a hodgepodge rubric of merit metrics wherein we say, well, this district measures it this way, that district measures it that way, so we're going to pay accordingly? And since teacher salaries are a complicated calculation based on length of school year (different for each district), in-service days, vacation days and sick days; percentage of retirement package paid into by district (ditto); level of benefits bargained for by that particular district's union (ditto); property tax rates for district if funded in that fashion (sigh, ditto), by what standard of equivalence to you mean to implement this?
And, will there be staggered levels of merit based performance pay dependent on level of education and field in which education is received? Will those with more background in educational rubrics (adding to their classroom management techniques which can improve performance on particular tests) receive a greater percentage than those whose additional education is in their field of specificity (advanced science degrees might also improve performance on tests)? And how are we to say that students in art class improved to such a degree that we might give all the teachers in that department bonuses, while the music department did not improve of a sufficient nature? Will we weigh our kids and provide bonuses to PE teachers who succeed in trimming off some pounds of their class while increasing their speed on the one mile? And what sort of weighting will we give to newer teachers who are likely to have adjustment difficulties with classroom skills and will have to spend more hours creating lesson plans than older, more experienced teachers who are not only deferred to more in the classroom typically because of their age as opposed to fresh college graduates?
All of which is to say that merit pay sounds like a good idea because no one of you actually spends any time in a classroom doing this work. It is a complicated business, it is not a "product" that is being generated, and it isn't something that for-profit business models can be easily or even correctly applied to.
Posted by: The Critic | Oct 17, 2007 4:26:05 PM
Hear hear Critic!! What a great post.
"This is something that’s done in Corporate America all the time and there’s no reason to think it can’t be implemented within out school system."
Yes, corporate America should be our model. Did Robert Nardelli "earn" his merit pay? Frankly, this post misunderstands the compensation system in corporate America; instead of "merit pay," most corporations engage in "tournament pay," don't they? Importing tournament compensation to public education would be a travesty...
Posted by: Michigander | Oct 17, 2007 4:56:38 PM
Merit pay for teachers is a smokescreen. It will not fix problems in education. It will not help get "better" teachers to enter the field of education. It will not help retain teachers. The bonuses are just not big enough to make people not change careers. Look, think of the personality type of people who go into education. By and large, they are people who want to make the world a better place not folks who want to get rich. They want a modest lifestyle with security and dignity.
Every dollar earned by teachers is "merit pay". We deserve our salaries. We do not work for tips. You want to offer us bonuses for exemplary performance? Knock yourself out. But at the end of the day, we are not going to let our salary be determined by something outside of our control . . . how hard the students work. Teachers can't "fire" their students.
To the comment about "stupid principals" . . . look, if you think teachers aren't held accountable, its even *worse* for principals. And school superintendents. And School Board members. And don't get me started about state and federal level education people. In the real world, a stupid middle manager gets reviewed just like their employees.
But moreover, a principal can *totally* screw up a teacher's chances of actually teaching by the students that are assigned to that teacher's classroom. If one teacher is assigned 8 students with special needs while another teacher has none, and both are teaching the same class, who will have better test scores? If teachers are being paid individually on merit pay, there will be *countless* arguments about class assignments. Teachers will avoid classes where they are at risk of losing their merit pay. Teachers will avoid sharing their "secrets". They will be looking out for number one instead of working together as a team. It is a bad idea.
The way to fix teacher recruitment and retention is not via merit pay, it is through things like scholarships, good benefits, and making sure that there is excellent leadership in our principals. If anyone thinks all we need are better teachers in our classrooms and the other stuff isn't important, they are a fool. Merit pay does not address the real issues.
Posted by: William | Oct 17, 2007 5:01:33 PM
As a bad idea, it would be similar to paying dentists based on how many cavities their patients had.
Actually, The Critic, this is precisely how most dentists are paid, though this actually kind of helps make your point. Dentists get paid for every cavity they find and fill in your mouth. So dentists have an incentive to find and fill cavities and very little incentive not to find and fill cavities. This, in a nutshell, is why I don't trust dentists. It seems that every time I go to the dentists, no matter how long it's been between visits, I have 3 cavities. Go figure.
When you reward some thing you get more of it. And people find a way to game the system. This is management 101.
There are many problems with merit pay for teachers. I can actually speak from experience of having been a student in a school when a form of merit pay was in effect. Back in the 1980s (yes, this debate has been going on that long or longer), when I was in high school, Seminole County, in Florida, implemented a form of merit pay. I remember quite well sitting in class on the days when our teachers were being observed by the merit pay analyst. It was quite absurd. The teachers would be more smartly dressed than usual and they would take us through a teaching extravaganza the likes of which we never saw again the rest of the year.
There was no merit to this particular merit pay system, and I can't think of a better one either.
The issue of whether principals are qualified to judge the merits of their teachers is interesting. Some might be. But teaching is unique in that there are no reasonable metrics to use. The metrics that exist (test scores, number of passing grades, or even individual student improvement in either area) are obviously no good for judging merit. In the case of test scores, individual teacher's influence on the outcome is impossible to measure. In the case of grades, the teacher controls the grade and so can push grades up or down to achieve the desired result. And teachers work alone, in private, so it's difficult for principals to gauge teachers' skills in areas other than paperwork.
And unlike in the private sector, there is no financial incentive pushing down on the principals to force them to make the correct decisions in doling out merit pay. This might be an argument for some form of school privatization were there not so many other arguments against that dubious concept.
Posted by: Rob Mac | Oct 17, 2007 5:09:37 PM
All of which is to say that merit pay sounds like a good idea because no one of you actually spends any time in a classroom doing this work. It is a complicated business, it is not a "product" that is being generated, and it isn't something that for-profit business models can be easily or even correctly applied to.
Exactly. I also have a hard time with the characterization of corporate America as some place where pay increases are always only given to the worthy, where everyone always understands exactly what's expected of them, etc. Seriously, Dilbert wouldn't be all that funny if it wasn't so accurate.
Posted by: Stephen | Oct 17, 2007 5:11:08 PM
Actually, The Critic, this is precisely how most dentists are paid, though this actually kind of helps make your point. Dentists get paid for every cavity they find and fill in your mouth.
Yeah, I think the better analogy would be to say that merit pay for teachers is like charging dentists for every cavity their patients get after being seen the first time. If the dentist was doing his job, then there wouldn't be more cavities, right?
Or maybe some things can't be measured like that.
Posted by: Stephen | Oct 17, 2007 5:18:09 PM
The Critic’s comment is akin to how my customers act and talk when they’re trying to fight change. First, you talk about how people don’t understand the nuances of the profession; then you ask a lot of questions, rapidly, as if the answers are supposed to be easily accessible; and they you point out some complexities, that while far from insurmountable, at least sound difficult to solve for. But this is just noise; a distraction if you will.
The reality is, any of these issues can be worked through towards the implementation of a merit pay system; teaching is not so difficult, or so misunderstood, or too ambiguous that such a compensation model can’t be applied. It’s like other customer facing professions in the sense that success isn’t always completely in the control of the employee and that bad customers/students are what really drive outcomes.
I agree that merit pay is not the panacea that some on the right may think. It’s a small step forward, but education success largely resided in the hands of the students and their families.
Posted by: DM | Oct 17, 2007 5:29:08 PM
The Critic’s comment is akin to how my customers act and talk when they’re trying to fight change. First, you talk about how people don’t understand the nuances of the profession; then you ask a lot of questions, rapidly, as if the answers are supposed to be easily accessible; and they you point out some complexities, that while far from insurmountable, at least sound difficult to solve for. But this is just noise; a distraction if you will.
The above could be seen as typical for someone whose livelihood depends upon finding problems and then coincidentally having the solutions for those very problems. Not that I see it that way, mind you, but someone else might.
The real problem is that for some reason "merit pay" has been accepted by a lot of people as a glaringly obvious solution to many of our schools' problems. Why is that? Can anyone here really spell out what merit pay will solve?
I suggest that merit pay isn't intended to solve anything, but is yet another in a long line of conservative red herrings. If merit pay is the answer, then obviously the problem is with the teachers. Teachers being the problem fits quite nicely with the conservative caricature of the lazy, incompetent government worker. Merit pay means that schools aren't underfunded, we just have bad teachers. And remember, if some teachers are getting raises based upon performance, that means other teachers won't be, which will keep our taxes down.
If merit pay isn't going to give people with in-demand technical skills the ability to make just as much as teachers as they can in the business world - and it won't - then there will still be a shortage of people with those skills. Nor will it necessarily mean that the slacker teacher is going to quit.
In fact, the mania with standardized testing means that it's easier now than it ever has been to mentally check out from teaching, because poor teachers know exactly how they will be evaluated and can spend their time just teaching to the test. If we use test scores to award merit bonuses and raises, it means that the people who make the most money will be the ones that can most effectively run a test-prep shop, not necessarily the best teachers.
Posted by: Stephen | Oct 17, 2007 9:54:08 PM
"And unlike in the private sector, there is no financial incentive pushing down on the principals to force them to make the correct decisions in doling out merit pay. This might be an argument for some form of school privatization were there not so many other arguments against that dubious concept."
Rob, actually the answer is quite simple: scratch merit pay for teachers, give Principals huge bonuses for successful schools and fire them if they don't get results. Make the bonuses high enough to get the most talented people and be serious about what you have to do in order to earn the bonus. Good Principals are underpaid. There are too many crummy Principals.
Posted by: William | Oct 17, 2007 10:07:07 PM
I think my dentist analogy was poorly phrased. I was suggesting that dentists wouldn't get to charge their patients based on cavities; they would be paid a set wage by some dental oversight board and dentists who had patients with good teeth with no problems would be paid handsomely and dentists who had patients with bad teeth would be paid poorly. Because clearly, the patients who had bad teeth are the responsibility of the dentist. It's his/her fault for all these cavities.
Posted by: The Critic | Oct 18, 2007 10:04:20 AM
The reality is, any of these issues can be worked through towards the implementation of a merit pay system; teaching is not so difficult, or so misunderstood, or too ambiguous that such a compensation model can’t be applied. It’s like other customer facing professions in the sense that success isn’t always completely in the control of the employee and that bad customers/students are what really drive outcomes.
And what will you solve? How will this improve education? Will you get better students or will you just put extra burdens on teachers?
No one has yet managed to present any comprehensive argument that demonstrates how some ideal, magical candy floss spun form of merit pay will improve education in any meaningful form. Can you point to a study that demonstrates its efficacy? Can you argue from facts instead of suppositions how this will suddenly transform American student performance?
No. No one has done that. All they've done is come here and say why they think merit pay is a good idea, but why do they think it's a good idea? Because it sounds good? Because it's all about "accountability" and who can be against that? Seriously. What effect will it have on education that will be beneficial enough to implement this complicated scheme?
Posted by: The Critic | Oct 18, 2007 10:10:10 AM
The Critic,
Now you’re on to something here. I would say that it’s never a bad thing to implement good management practices; so implementing merit pay is a good thing. What will you get from it? Implemented properly, you’ll get better, more effective teacher. What’s the effect of this? Maybe some help on the margin, but as I’ve stated before, this would have to be one of many small changes that combined may push the dial towards improvement.
But, just because merit pay alone is not the answer doesn’t mean it’s something that’s not going to yield improvement, even if it’s small. Plus, it comes at little cost and doesn’t require a huge, up front investment. So from my view, there’s really no reason not to implement a merit pay system.
Posted by: DM | Oct 18, 2007 11:28:44 AM
The comments to this entry are closed.
