teacherken's musings on everything

One of the more controversial proposals to improve education has been the idea of giving teachers merit pay. There is an article in today’s Arizona Republic which discusses the use of such a program in one small (6,300 students) district in that state. You can read the article [Rewarding best teachers: In Casa Grande, they earn more if they get good grades http://www.azcentral.com/arizonarepublic/local/articles/0911teacherpay11.html] if you think it necessary. In this post I will address the issue of merit pay with respect to the compensation of teachers but also explore the issue with respect to other professions covered by civil service regulations at local, state, and national levels.

Traditionally teachers are compensated according to educational background and numbers of years of service. Most school districts will have multiple salary “ladders”, such as bachelor’s degree, BA + 30 credits, MA, MA + 30 credits. Some will add additional ladders, such as MA + 60 credits and/or doctoral degree. For each ladder, the level of compensation depends on number of years of service, with higher compensation paid for longer service. This model of compensation is very similar to how most governmental employees are paid, the difference being that education is not the determinant of which salary ladder is applied, but rather the classification of the job. Thus in my previous career as a data processing professional for local government (Arlington County Virginia) a computer programmer was compensated at a lower scale than a systems analyst, who received less than a supervisory systems analyst, and so on. While within fields the job requirements could easily be differentiated, such an approach sometimes caused problems when jobs in different fields were compared -- how does one compare the job requirements of a librarian against those of a sanitation worker, for example. Those jobs which were traditionally staffed by females - such as librarians - usually were compensated at a lower level than those that were traditionally male, regardless of things such as educational requirements, responsibilities of the job, etc. While I was at Arlington, we went through a complete reclassification of all civil service jobs which attempted to address this issues (and to a large degree did), but then, Arlington is politically a very liberal community (in my 23 years hear the only Republicans to carry the County have been one incumbent Congressman, two County Board members, on incumbent Commonwealth's attorney, and two members of the school board. Even Reagan in ‘84 did not carry the County. We have never had a Republican statewide candidate win, nor state legislator .. those Republicans who did win were well-known local figures. Currently, with 6 in the state legislature (at least in part from the County), 5 each on County Board and School Board, and 5 Constitutional offices, there is exactly one Republican on the school board).

One may note that this model of compensation is exactly paralleled by the Federal GSA schedules, and also the military pay schedules. In the latter, one gets a promotion in rank, through which one is paid on a higher pay ladder. Additional compensation for the military can be achieved for specific items such as flight pay or combat pay. For some job on Civil service scales there is additional pay for working 2nd or 3rd shift (although this usually does not apply in managerial and supervisory positions, as these are considered exempt).

Many of these compensation systems have some measure of financial reward because of performance, as well as the possibility of financial punishment. In the military, the way this occurs is through promotions and demotions. There is no merit increment to the pay, nor in most circumstances are there bonuses for specific action - here the recognition is usually through the awarding of medals and citations. One will note that some citations recognize that the individual does not work alone, and hence are group or unit citations.

In some civil service environments one can e paid bonuses based on performance - so-called performance awards. These are one-time awards which do not affect the base compensation (unlike promotions). My wife, who is a Federal employee, has received several of these. In all of these systems (not yet talking about teachers), the method of evaluation may be in part against the standards for the job, but in most cases (excepting in general the military) there will be a specific individual performance plan against which the employee will be measured.

No let’s look at teacher compensation. In recent years there have been explorations of changing the method of compensation to recognize some difference in performance or job requirements. Thus there are districts which offer a higher level of compensation for working in harder to staff schools, or schools in greater need of improvement. This additional emolument applies only while one works in such an environment, and as such is roughly equivalent to combat or flight pay in the military. Some districts have gone to school-wide or school applicable bonuses for improvements in test scores (I will address test scores anon), the difference being that school-wide applies equally to all members of the school while school-applicable are distributed within the school with the principal having some level of discretion as to the distribution of the awards. These have been controversial in both applications, among other reasons for the dependence on test scores, but also because of the issue that some principals who have discretion have been known to abuse that discretion. Of course, in the latter regard one can find similar complaints in the military and elsewhere in the civilian sectors at all levels of government.

The use of test scores is controversial for many reasons. If compensation in anyway is solely or largely dependent upon test scores, the pressure to manipulate and/or teach to the test seems to increase monumentally. Thus increased scores may or may not indicate an improvement in student learning. We have seen cases in places like Texas where he improvement on state test scores was more an artifact of manipulating who took the tests rather than any increase in student learning. Even in good schools in good districts, there have been cases of principals and teachers whose compensation depended upon improved test scores (in already high-performing schools) who felt the pressure to show improvement was so great that they cheated.

Okay that’s the background. Clearly a merit approach can be done without exclusive reliance upon test scores. The Casa Grande system for example is far more complex. As the article notes

Traditionally, teachers are paid based on years of experience and educational degrees. But Casa Grande teachers also can earn an extra $2,300 a year by meeting 67 goals based on what they know and show they can do.

And despite the traditional objection to such plans among teachers,

But Casa Grande believes it developed a model that works: Involve teachers in the planning, set reasonable and clear goals, offer lots of staff training and tweak the system when needed.

Note the differences -- this is not a plan imposed from the top, but includes teachers in its planning and design. The goals set are clear and reasonable, staff training is offered, and the system is not locked in but capable of modification. Also note that the additional increment is not permanent, but contingent on continued meeting of the goals. BTW, I would contrast that last point with the 4,500 per year (from state and local district) that those in my system receive for having completed National Board Certification -- once one achieves it, it is an ongoing (so long as the funding exists) additional increment. I will discuss this more anon.

Let me offer a few more snippets from the article

Teachers are measured on such things as establishing rules in the classroom, stimulating interest and involving students in the learning process, and building relationships with parents.

New this year, experienced teachers can opt to design their own projects instead of being evaluated on the 67 goals. The idea is to encourage teachers to test new methods that could boost student achievement. If the data show positive results, they get the bonus money. Starting teachers with no previous experience and a bachelor's degree could earn $35,175 at the end of this year in Casa Grande if they earn all the performance pay available, said Brenda Tijerina, personnel director in the district. Without any kind of bonus, that teacher would earn $31,800. Statewide, beginning teachers last school year on average made $28,100.

The key to dramatically changing a pay plan is getting teachers involved, education experts say.
In Casa Grande, administrators worked with teachers and other employees to develop the pay system.
The eight-person team consulted other districts and an educational consultant to create a plan they believed everyone would accept.
They've changed it over time. This year performance goals were expanded from 17 to 67 to make expectations clearer to teachers.

In the early 1990s, schools around the United States faced increased scrutiny from the public, and "a number of publications talked about the sad state of American education," said Lisa Gross of the Kentucky Department of Education.
"People started really paying attention to what was happening in their local schools," said Gross, whose state schools also offer incentives for teachers, but they're not tied to evaluations as in Casa Grande.

While the public wants results, some teachers are skeptical.
Performance pay might work in small districts like Casa Grande, but it can present more challenges in a large district, as it did in Cincinnati where teachers voted to reject a performance pay plan.
"It could be that they have developed a sense of community and mutual trust," said Sue Taylor, president of the Cincinnati Federation of Teachers, commenting on Casa Grande's pay plan. "It wouldn't work in our district maybe because we're so much larger. Also, we have this absolute belief that whoever evaluates us must have expertise in our area."

The question of the evaluation and compensation of teachers has been ongoing. Here I might note that the Denver schools started a process of merit pay a number of years ago on an experimental basis. Their plan was set up with the cooperation of the local teachers’ union. Rochester a few years ago set up a procedure by which there was parent input into the the evaluation of teachers -- it was a part, but not in itself the determinant of the evaluation of that teachers. The use of test scores as ONE (among many) factors in the evaluation of teachers is far less objectionable than exclusive or predominant reliance upon test scores as has been proposed by some (who usually lack real understanding of what occurs within classrooms).

One reason for the emphasis on evaluation and merit pay is the perception / belief that (1) there are a high number of less than fully qualified teachers in the classroom, and (2) there is a difficulty recruiting and retaining more qualified teachers, especially in certain fields (eg math and science) where pay outside of teaching is so much higher. One can question the accuracy of both perceptions as less than fully accurate, but the concerns that arise even from partially distorted perceptions is nevertheless real. And clearly if teachers are to be paid more - a position for which there is some general acceptance - the people as a whole want to know that they are getting something in return for the additional taxes that must be collected.

I will not in this piece offer a definitive answer to this issue. I have done some serious research on the issue of teacher evaluation - it was the underlying topic of my now abandoned dissertation. And there may be some justification for varying somewhat the compensation teachers receive so that it is not exclusively based on education and years of service.

let me first note several things about myself, and then raise a couple of questions I think need to be addressed. First, by most approaches to evaluation and/or that are used for so-called merit pay I would come out very well indeed. My students do quite well on standardized tests, even though I do not teach to the test. I have a high level of student involvement. My track record on communication with parents is outstanding. I often have parents ask to have their children assigned to my classes because of older siblings and/or neighbor children who have had me (we do NOT do that kind of class assignment), and my evaluations from observations - both by administrators from within my building and from people from outside - have consistently been outstanding. I have a track record of professional involvement: attendance at seminars to improve my knowledge and skill, including several to which admission is by competitive process, presentations at professional conferences including at a state-wide level, helping train student teachers and mentoring new teachers, serving as a leader on some school-wide efforts at improving and/or maintaining our excellence. I have two masters degrees and most of a doctorate, and will either this November or the following year obtain my national board certification.

And yet, in general I oppose such differential compensation on a merit basis, even though I would financially benefit. Why?

First, teaching is a collaborative process. One might argue that the performance of elementary students is largely the responsibility of the primary teacher. While I have never taught elementary, I would question that, as often the most significant teacher is the art, music or phys ed teacher who is able to connect with the student when the primary teacher cannot. Without pursuing this further, I note that the secondary situation is quite different. In Middle school, where I taught for 4 years, a team of teachers shares students. If I am one of 4 core subject teachers for those students, there is no way I can be solely responsible - positively or negatively - for the performance of those students, even within my core subject area. At a large high school (2,900+) such as that where I now teach, there is far less of such a sharing approach. And yet, within our department there is a fair amount of cooperation and division of labor without which none of us would succeed. Thus we are teaching AP government to 10th graders for the first time. I he three sections, another teacher has one. We are doing joint planning, dividing up the responsibilities, etc. Neither one of us could do all of it on our own. How much of our success will be individual and how much will be as a result of our joint efforts? And if you argue that we should share any additional compensation, does that mean I should get 3/4 of the additional because of the difference of student load, or should it be 1/2 because we are equally sharing the planning? But what then about the support we receive from others in our department? There are 9 other teachers who already teach AP social studies courses, who have been generous in helping us get organized and learn from their experience. There is a school-wide support system for AP. We have an administration that is terrific at supporting teachers in using their best judgment on academic matters. Our school system paid for a one-week AP training session for both of us, and our PTSA has paid for a one-day session for me, and for supplemental materials for both of us. Are not these part of the compensation and support we have already received?

I am concerned that to reward teachers based on specific performance criteria will undercut any attempts at our acting like and being viewed as professionals. In general, lawyers and doctors are not compensated based on such specific criteria. They are expected to act like responsible professionals in their fields. My own approach towards teaching is similar. Thus I have a bias against such differential compensation.

That bias is not overwhelming. That is, I can be convinced that some measure of merit compensation is justifiable. I would expect it to depend upon multiple measures, and I would demand that it be designed and administered at least in part by teachers, so that it not undercut what should be our sense of professionalism. It should include teachers having ongoing professional development plans, and should require support and training to achieve those professional goals. Test scores could be a part of the evaluation, but here there should be some measure of improved learning - if not strictly value added, before/after test scores. There is no way I should benefit from performance on the state test on government when 95% of my AP students could pass that test before taking my course, and that is unfair to the teacher who has special ed students who have little chance of passing the first time they encounter the state test.

Experiments such as Casa Grande will occur. I am enough of a realist to accept that. I hope that there are a variety of approaches that can then be fairly evaluated. And if schools and systems decide to follow such examples, I hope they are wise enough to make the necessary local adaptations and not merely buy and/or insert a system from elsewhere which might not be appropriate to local conditions.

This bloviation has gone on long enough. I strongly encourage others to comment on this subject, whether or not you are an educational professional. This is an important issue, and it deserves our attention.