US Study: Merit Pay for Teachers Does Nothing

[Camikaze]

The results of this study surprise me, in that I would've expected some negative impacts, rather than no change. I guess it's theoretically possible that this could happen (there have to be reasons people suggest it in the first place), but I'm surprised that it actually has not failed as utterly as I would expect.

Note: The second poll option seems too mild for what I'd like. Swap the second and third words and it's more representative of my view.

 

[Earthling]

You know, I'm just not totally sure that's true. It is certainly partly true, but if you're willing to put up with 25 years in a crappy school, your final salary really isn't that bad (considering how easy it is to get a teaching degree and get credentialed). TOP teacher salary in Greater New Orleans was in the upper 70's, and in ChicagoLand (for some of the suburbs) it was in the upper 80s.
I think the "social capital gap" has other factors besides raw salary...the "prestige" of the position (pretty low, and these new reforms demean teaching even more), the lack of advancement potential, and quality of life. Right now, teaching is viewed (unfairly or not) as a secondary household income (something the wife does while the man has another job), a temporary job that requires some sort of missionary zeal, or what dumb sorority girls major in.

Yes, but well that's also because that salary progression is still peanuts to many other comparable professional fields. 70k after a teaching degree (even if "easy") and 25 years isn't exactly promising when someone would make that much starting out in your twenties with a degree in engineering, or degrees in law or medicine or so on and go much much higher still. Even the equally "easy" bachelor's in business can probably outpay teaching for more successful folks.
Though, I agree it is true after a certain point not everybody would care, and other factors going into the stress, difficulty, and prestige of the job certainly have an effect. I wouldn't hold out for 70 or 80k to teach for instance but I wouldn't start at 30k either with terrible hours and all the problems and stress to deal with. (Though in all honesty I wouldn't have been really likely to become a teacher anyway. because I don't think I'd do so well working with kids and would enjoy other work more. But some of my classmates, sure, there could easily be a better way that would attract more people like them to teach math and science and so on instead of the current crop of education majors)

 

[Dachs]

Not sure why people are still debating this, but let me make the point again

Still using Wikipedia articles in lieu of an actual argument?

 

[Integral]

The objection about self-selection bias is valid. Of course, once you get past the self-selection filter, it looks like a vanilla randomized control trial.

However, something tells me that the subset of teachers willing to participate in such an experiment is not representative of the set of all teachers.

It also appears that the study had a 50% attrition rate. That's never fun to deal with, but of course whether attrition will affect the results of the study depends on the cause of attrition at the individual level.


Having skimmed the report and being familiar with other research coming out of Peabody, the vast majority of the methodology is very solid. Despite self-selection concerns, you really couldn't have designed this any better.

 

[Dachs]

The objection about self-selection bias is valid.

perhaps


Moderator Action: That really didn't contribute to the topic of this thread.
Please read the forum rules: http://forums.civfanatics.com/showthread.php?t=422889

 

[GoodGame]

So Merit Pay didn't trickle down and boost students' scores? That's not exactly surprising especially in the short term. I'd say it'd be naive to think it'd magically occur in a short cycle. C'mon it was only 3 years.


But, Downtown, you didn't highlight this:

Researchers cautioned, however, that the Nashville experiment does not provide answers to many other questions about incentive pay. For instance, it wasn’t designed to test the hypotheses that pay incentives might serve as a draw to a different population of teacher-candidates or as an incentive for other candidates to stay in the profession—thus potentially changing the quality of the teacher workforce.

I'd say that's more important. That is what I'd expect merit pay to achieve in reasonable time frame. Boosting students' test scores would be a more long term effect of merit pay, I'd hypothesis.

And bottom line, doesn't merit pay make intuitive sense in general? Wouldn't you want to pay better qualified, more experienced, better performing people more money, regardless of profession? Wouldn't you want such people to be retained?

 

[Camikaze]

Yeah, you would want to give better teachers better money as a general principle. But there isn't any metric under which this can be done fairly (that I've heard of). The most basic metric; student grades, is a horrible way to judge a teacher, given how many variables affect such things.

 

[Earthling]

Once you have teachers teaching at a school with the same background, I'd agree, it's hard to make merit pay work very well going from there (and that's what the study mostly found out)

The "fair" way of giving more money to better teachers would be in hiring though, hiring those with better degrees and qualifications and starting with more money at the onset, but that's problematic to implement for a ton of unfortunate reasons (lack of general school district money, unions, etc..)

 

[downtown]

^^thats exactly what Unions want actually. Virtually every union contract allows for more experienced, or more credentialed (more degrees, National Board Certification) to get more money at the start.

So Merit Pay didn't trickle down and boost students' scores? That's not exactly surprising especially in the short term. I'd say it'd be naive to think it'd magically occur in a short cycle. C'mon it was only 3 years.


But, Downtown, you didn't highlight this:


I'd say that's more important. That is what I'd expect merit pay to achieve in reasonable time frame. Boosting students' test scores would be a more long term effect of merit pay, I'd hypothesis.

I think I did mention that back on page two. And I don't see why it would take 3 years to benefit? If Merit Pay serves as a motivation for Teacher Improvement, it should start in year one.

 

[Camikaze]

Unless the idea is structural change withtin the teaching workforce as a whole. That could conceivably have long term benefits whilst not showing anything in the short term.

Although probably not.

 

[scherbchen]

I'm fine with any sort of merit pay system not directly affecting student's test scores. I am fine as long as it gets people interested in the profession. perfectly fine with letting schools decide which teacher to hire once there is a massive overflow of applicants. even perfectly fine with my kids (niece and nephew and a little sister, really) getting mediocre grades in tests as long as they are engaged and participating.

sure it is difficult to actually award merits if you are not going by test scores and those are what most people care about. get into that school, this college, that university, that job, this prep school, this private school. terribly sad.

or just throw another billion or two at some finance bubble that just crashed. what do I know? I just believe it is a decent idea to have somebody who takes care of your kids more than you do a pretty darn good wage. you know what happens to those wealthy guys in Mexico who pay their bodyguards a paltry sum, right?

 

[Cutlass]

So Merit Pay didn't trickle down and boost students' scores? That's not exactly surprising especially in the short term. I'd say it'd be naive to think it'd magically occur in a short cycle. C'mon it was only 3 years.


But, Downtown, you didn't highlight this:


I'd say that's more important. That is what I'd expect merit pay to achieve in reasonable time frame. Boosting students' test scores would be a more long term effect of merit pay, I'd hypothesis.

And bottom line, doesn't merit pay make intuitive sense in general? Wouldn't you want to pay better qualified, more experienced, better performing people more money, regardless of profession? Wouldn't you want such people to be retained?

But that thinking is based on the assumption that many teachers are just phoning it in, and not even really trying. So it would change the behavior of current teachers. And while I'm sure there are teachers like that, it doesn't seem reasonable that large percentages of them are.

And, if there are large percentages or numbers of teachers just phoning it in, then that suggests that the problem should and could be handled by other methods of management.

 

[Ayatollah So]

Can you run the DOE?

Seconded. DT for Secretary of Ed!

But the whole school pretty much blocks out 3 weeks of "TAKS prep" each year, as we drill and kill our kids in the subject areas they're to be tested in.

For the rest of the year I do my best to teach critical thinking skills along with the rah-rah patriotism I'm paid to deliver.

Careful, there. To quote a famous philosopher, reason is one letter away from treason.

Of course the system still sucks. It turns out that if you get your 5th grade students from last years' particularly inept teacher who added no value to her kids' testing, it greatly increases your chance of adding extra value to the normative growth rate of the students--and thus increases your chances of getting a bigger bonus than your fellow 5th grade teachers.

I never thought of that before, but, duh. Oh well, at least that one is likely to average out in the long run. Or not - I guess it depends on the structure of your school.

And bottom line, doesn't merit pay make intuitive sense in general?

Truth comes from the gut, not from books! (Boy, that famous philosopher really comes in handy.)

Wouldn't you want such people to be retained?

Now, that's a whole 'nother ball of wax, which this study doesn't really touch. Merit retention might well be a whole lot more useful than merit pay.

P.S.: Kudos to suiraclaw for that link to Dan Pink's TED talk.

 

[El_Machinae]

^^thats exactly what Unions want actually. Virtually every union contract allows for more experienced, or more credentialed (more degrees, National Board Certification) to get more money at the start.



I think I did mention that back on page two. And I don't see why it would take 3 years to benefit? If Merit Pay serves as a motivation for Teacher Improvement, it should start in year one.

I'm under the impression that certification (& seniority after five years) doesn't predict whether someone will benefit students.

 

[downtown]

I'm under the impression that certification (& seniority after five years) doesn't predict whether someone will benefit students.

You're exactly right, which is why tenure reform (or removal) is a major tenet of US "school reform".

 

[uppi]

I never thought of that before, but, duh. Oh well, at least that one is likely to average out in the long run. Or not - I guess it depends on the structure of your school.

The system could even be gamed - not by the individual teacher but by schools - one way or the the other: assign the worst teachers to the first grade ("It's not our fault, it's the bad kids we get here") and then line up the teachers in ascending order, each one improving on the other and each one getting their bonus. It might even look good for the school ("Look how terrible the kids are at the beginning and how we give our best to make them at least mediocre").

If the school wants to save money it even works the other way around: put the best teachers at the beginning and no one has ever a chance to improve on that.

 

[BuckyRea]

I never thought of that before, but, duh. Oh well, at least that one is likely to average out in the long run. Or not - I guess it depends on the structure of your school.

As soon as they instituted the merit pay system, I immediately began plotting out how I could game the system for maximum dollar rewards. For instance, when they first released the formulas, I figured out that I could make more money over two years if I did a really crappy job teaching to the test in year one, and then showed a dramatic improvement in my kids' scores the following year. The problem is that they kept on changing how they calculate those bonuses & added value rates. In the end I just said to hell with it and did my best to teach all the kids as best I could. How disappointing.

It's all a nice thought, of course, paying more money to the best teachers. But in concrete application it's entirely too apples & oranges to produce reliable results. It's not like a sales commission in which two salesclerks are competing within one store. I'm competing with teachers who teach students at different schools and thus have very different students producing those testing results.

(inserted on edit & rethinking) There's also the problem of what we mean by "the best teachers." I'm an okay teacher when it comes to gifted & talented kids. But it turns out that I totally kill when I work with on-level but English deficient students. I can ramp up vocabulary to grade level like Sally Field flying by habit. But then every time I get a new principal, he looks at my students' testing scores and decides to assign me a couple of honors/AP classes. I can do okay, but that's not where my strength lies. Only he, as a principal, is under a lot of pressure to get his AP numbers up, so he juggles staff year to year to tweak up his results an additional notch.

The system is just begging for corruption, cheating, and favoritism. Mostly it's just encouraging the pursuit of short term rewards instead of looking for long term strategies that will maximize system-wide results.

 

[Earthling]

I'm under the impression that certification (& seniority after five years) doesn't predict whether someone will benefit students.

Mostly because such measures don't really take into account actual factors that apart from personality should reasonably go into teaching success. Certification that doesn't require specific competence at what you'll be teaching won't help so much. If teacher certification required, say, a degree in the field (like a degree in Biology to teach Biology) the correlation would probably be there, but it often doesn't. (I'd agree that just a degree in teaching/addtl certification and seniority there might not indicate so much)

 

[Cutlass]

The other problem being that someone who can get several degrees can most likely make more money than teachers make.

 

[El_Machinae]

You're exactly right, which is why tenure reform (or removal) is a major tenet of US "school reform".

There's something seriously wrong with the educational sciences if a teaching certificate doesn't cause you to be better at teaching. If we don't know how to teach teachers to teach, what do we know about teaching!