The problem, of course, is the statistical assumptions. Just as one might flip a coin three times in succession and get three heads, a teacher might be randomly assigned an unusually high number of disruptive students who affect the performance of their classmates, or an unusually high number of bright and nurturing students who boost their classmates. In any given year, an unusual number of students might suffer through abuse, divorce, or other traumas at home that cause their performance to nosedive through no fault of the teacher. Alternatively, several abusive parents or stepparents might be removed from the picture, boosting average performance with no credit due to the teacher. Using the raw “value-added” scores to rank teachers has the potential for significant abuse.

God forbid teachers are graded. Their unions will not permit it.

Those who earn their living grading others cannot themselves be graded.  What hypocrisy.

My wife teaches second grade.  She is committed to excellence in teaching and in students.  The teachers who get her kids in 3rd or 4th grade are always glad - BUT - students are not widgets.  Each is unique and each has a unique setting.  Teachers are far from the only variable involved in kids learning, as Dale Rye notes above.  There is no fair objective way to “grade” teachers.  They are simply far from the only factor in kids’ learning.

Dale, you missed the “based on changes in test scores from year to year”. If the students were “bad” students coming in, then that would have been reflected on last years tests. In contrast to flipping a coin three times, if one flips a coin 20 times (e.g., 20 students), you very rarely will get 5 or fewer heads or 5 or fewer tails, say. This is a fairly level playing field. Of course, teacher unionists and their liberal apologists will give the argument that it is not a perfect measure (and there isn’t any perfect measure), so let’s continue with not rewarding good teachers over bad.

Imagine the power you’d have felt in the sixth grade if you and your classmates knew that bad test scores would get your teacher fired. 

Wooohooo!

This system is not bad in theory. It is simply bad in practice, because the number of abuses and the ease with which they are undertaken will nullify the theoretical benefits. The Rule: There is no substitute for good character and for sound academic training, but the application of these obvious elements to instruction cannot be systematized. Students are not products, the educational system is not factory production line, the application of manufacturing standards to education is incommensurable, and teachers are not workers on the line - unless this is what you want them to be, in which case the system above is excellent and will yield the result you want.
  Bad teachers need to be fired. But how can you tell a bad teacher?  By having the department chairs and outside evaluators write essays on the teacher’s MO over the process of several years. If the summary of all evaluations is negative, then the teacher should be fired without having to surmount endless obstacles. There is no substitute to personal observation and the written assessment. This is much too simple, much too obvious, much too demanding careful assessment, much too time consuming to be adopted.  Larry

Businesses grade employees on their performance all the time.  The military seems to be able to do it as well.  It seems that the only group that cannot be evaluated on their performance is teachers.  Yet as I stated above teachers are expected and required to grade students on their performance.  Amazing.

I taught high school for almost four years. I put my heart and soul into that job. The kids couldn’t care less. I asked my students every year, “how many of you didn’t try on the STAR test last year because it doesn’t count against your grade?” Most of the class raised their hands. After numerous instances of me finding graphic drawings involving me, being cussed out almost every time I went to confiscate a phone, being yelled at by parents over the phone, admin not backing me up on discipline, and on top of that…being evaluated based on someone else’s effort on some 4-hour test once a year… no way worth the pay. I left that job despising teenagers.

#9, see#8. This is one reason why his example is meaningless, because in a business, if employees simply refused to work, they would be fired. But not here. Business models will NOT work for teaching because this is NOT a business. I spent 21 years in public education trying to explain this (among other things) to school committees who went right on thinking that if THEY could run a business, they could surely run a damned school.
  ˆncidentally, one more thing that #8 ignores is the peculiar inevitability that the football team and the hockey team will’ get the money they want and the English Dept will get the leftovers. And this is true whether the football teams has a winning season or not. Some business. And this is why evaluation must be personal, extended - and exercise in the humanities, not an exercise in business administration.  L

Larry, do you really mean that a teacher can have no measurable deliverables?  If students can be graded on what they have learned (that is been taught by their teacher) why can’t the teacher be graded/evaluated on the teaching?  If you can do one you can do the other.  If teachers cannot be graded then neither can their students.  Why even bother with report cards?

10.  I think you’re right, but the key thing that test scores, particularly value-added scores, can add to the mix is the ‘baseline’ from which the ‘human’ evaluation needs to be compared.  In other words, what do we do with a teacher that is profoundly effective in the class room re: bx management, care for students, etc., but is consistently ineffectual in raising test scores?

Michael, two years ago I was evaluated seven times during the school year. Five of those times I was given outstanding reviews by my assigned mentor, department chair, and my university-assigned evaluator. Those did not count. My official evaluator was the director of HR. He visited my class for 20 minutes in the fall, and my class was running well that day. The class was interactive, and people were actually asking questions and doing their work. I met with the HR director who presented me with his findings: I was marked “needs improvement” on everything, which is considered a terrible review. He gave me a list of things to work on. I applied all of what he suggested to my class and methods. Two months later, he visited me again (on the day we came back from Christmas vacation). Well, later that week, in his office, he told me that I showed no improvement, explaining the things he saw that were “wrong” in my class. I handed him my previous evaluation and showed him everything that I “fixed” from my last evaluation. I also asked him how he could say that I showed no improvement when he didn’t refer to a single item from my previous evaluation.  I also caught him contradicting himself during the meeting and pointed it out.

Needless to say the meeting did NOT end well. The man seemed very satisfied when he handed me my “you will not be hired back next year” notice in March.

Another teacher I was working with spent three years at that same school, and received good reviews. If he came back the next year, he would have received tenure… but he was let go without explanation.

I have heard many horror stories of incompetent administrators doing spot-evaluations (15-30 minutes, once or twice a year).
Are these the types of evaluations you are referring to?

if your question means, can excellence or a lack thereof be reduced to a set of numbers. The answer is, no, obviously not. This is like asking me if a painter’s quality can be reduced to a set of numbers based on how many paintings he has sold, and at what prices, how many hours he works, the difference between his asking price and actually selling price and whether Sotheby’s has ever shown an interest in handling his work. Are teacher’s like artists? Good ones are; the best always are.
  Can third rate teachers produce good scores? Why of course. White males produced high SAT scores year after year when taught by the mediocrities who staffed public schools 50 years ago. But how can that be? Because the white males INTENDED to test well. And if they did NOT intend to test well, as is the present case, the best teacher in the world will not produce high test scores. This has been examined again and again over the last fifty years. And if, to boot, some classes have mainstreamed kids who destroy a class while other teachers do not have them? (I have bitter knowledge of this misery) And if the town is the worst kind of blue collar that despises intellectual performance? ( and like most teachers, I have sour knowledge of this condition too)  Finally some teachers teach solely for the test (and very successfully) and some teachers teach to educate. Is there a difference?
  Grading math is right or wrong (commonly). But English essays SHOULD be graded X/Y over Z, when X is content, Y is style and Z is the holistic result. Thia becomes a single grade on a report card, but this obviously fails to tell the substantive truth, which is why guidance counselors write letters to go with applications.   
  There is one way to make the Race To The Top and Teaching As Manufacturing a Product work: Standardize all curricula and all tests and require all teachers to teach to the test. Then you will have established a common baseline from which judgments can be made.
Nothing else will do.
  The alternative is to treat teachers as scholars, especially in the Humanities where assessment is hardest, and assess their work in essays by multiple assessors who discuss each class with each teacher so that a BEHAVIORAL baseline can be established. This takes hours and hours, and, would you believe it, competence in academics and adolescent behavior. Pass/fail ratios, for example, will be used ONLY IN THIS CONTEXT. Other than that they are meaningless.
  A father said, “When I got married, I had five theories and no kids; now I have five kids and no theories.” This is the gospel, but America, brainlessly, persists in thinking that the thermometer tells you what the weather is. Larry

The actual teachers here have spoken to the practicalities of how most teacher evaluation systems work in practice. I’m sure this is long since a dead thread, but I would like to address Robroy’s statistical argument in #4. One assumes that class sizes are as likely to be odd as even in number. If Individual A flips a coin 21 times, he cannot get 10.5 heads. There is exactly the same chance that he will get 11 heads as there is that Individual B will get 10. Does that make A 10% better than B? If A and B flip the coin 20 times each, there is the same chance that A will get 9 heads as that B will get 11. Does that make B 22% better than A? In the case of “value added” evaluations, the risk of random variations affecting the outcome extends not only to this year’s scores (which the teacher can do something about) but also last year’s (which is beyond the current teacher’s control). Suppose that A gets a class of 21 that had 11 heads last year and gets 10 this year, while B gets one with 10 last year and 11 now. B isn’t a better teacher; he’s just luckier. Unless we are talking about the Las Vegas schools, gambling odds shouldn’t be a determinant for teacher pay.

13, maybe the evaluations were not fair, but that is far different that what I see is the claim or assumption that teachers, alone among professions, are so unique that their job performance cannot be objectively graded or evaluated.  Any one in the military is subject to constant evaluation by their immediate superior.  Officer and enlisted efficiency reports are a way of life.  Indeed an unrated period is a bad thing.

I just find it bazaar that a profession that earns its living by the evaluation of others insists that its own evaluation is impossible.  I can understand not wanted to be evaluated, but I am amused that teachers claim that their profession is so unique that evaluation is impossible.

Each teacher is not evaluated only based on a standard set of evaluation points, they are also subject to the evaluator’s own opinions on how the class should be taught. There is a myriad of opinions and methods of teaching many different styles of classes. One evaluator may say that I am doing fine with a method, and another may think it is a showing of poor performance.

Well, if it can’t be determined that teachers can teach anything maybe they are over paid.  Come to think of it we don’t need to grade students either because if we can’t determine that teachers are teaching them anything, we clearly can’t determine that students are learning anything.  And I guess that means that we don’t need schools for teachers who can’t teach and students who can’t learn.  I want my property taxes back.

I agree that teachers need to be evaluated. The problem you run into is when you tie student scores to evaluations, or you have administrators with their own teaching philosophy evaluating someone with a differing teaching philosophy.

I don’t know what to suggest, but the current system of evaluations stinks

#16, using the military makes my point for me. Here the behavioral standards are uniform across an entire service, even if they are often boneheaded (as I am told some indeed are.) The point is that a baseline can be established that defines competence. So I said you could get what you want if you standardize all curricula and all testing across the country, and require all teachers to teach to that test system. Do you really favor this?
  Now, a teacher’s job performance can be assessed with some objectivity, but this objectivity is often inseparable from subjective elements. This is why multiple assessors, strong themselves on competence, are required. The assessments will tend to cancel out mere subjectivities. But this means that each assessor MUST examine first hand each class, because the number of significant variables far surpasses the most powerful computer to grasp. Just think about it. Today, X is a sweetheart. Tomorrow she has fallen in love with Y two rows over and so has worn a sweater put on with nail polish to catch his attention. She gets attention all right. There are more hormones flowing in the aisles than water in the Kennebec. Do you know what this does to a class? And if the evaluator is there?
Will this alter test scores? It sure will that day. And the NEXT day it is something else again.
    Or, eight students in the class a really stupid, but they are there because the parents want them there, and this teacher has a rep for getting along with dummies. So year after year…. What does this do to test scores? As an ex department chair, I can tell you, for that teacher, it is ruinous. Good rep, bad scores, bad pass/fail ratio.
  And here’s a final point which no one likes: A department chair can tell, in two years, who’s good, bad and mediocre, even without written assessments or test scores. You can tell just by listening to a class, by smelling the atmosphere.
  You grade students only because you are forced to. At the half year, the teacher declares three categories Excel,Pass,Fail. That’s all.
Fail warnings go out at the quarters.
  Schools are encouraged to throw out the dungbombs. Parents are given their educational tax dollars back and told, “Go put this kid somewhere else. We don’t want him.” Parents are told at the outset, every year, “We don’t care about brains; we care about a willingness to learn.” No willingness, he’s toast. Eventually parents will catch on.
    And the evaluation system I outlined above is faithfully applied.
  THIS will work. But it is beyond most Americans to grasp this principle, that they themselves don’t know squat about skillful teaching - until they do it themselves and then suddenly the homeschoolers (whom I favor) realize they just don’t know what they need to know. Consider how well so many homeschooled kids do, and nary an assessment between ‘em. What can the reason be?  Larry