Monday, June 25, 2012

Why Teacher Appraisals Should Not Directly Include Student Test Scores

I can immediately think of four reasons why student test scores should not be counted as any direct percentage of a teacher's performance appraisal: (1) it is inherently unjust to judge person A on the basis of person B's performance; (2) any such formula is by definition an instance of the fallacy of composition; (3) it will destroy teacher morale, and therefore end up costing taxpayers more at a time when we haven't money to pay for those increases; and (4), worst of all, the backwash effect of any increased emphasis on testing will continue to degrade the educations our students are receiving.

(1) I am completely willing to be judged on my performance, but given my experience at Locke High School, where I heard of students making funny face designs out of their Scantron answer sheets on tests which held no stakes for them, I believe the potential for abuse is too great to make important decisions about people's employment and careers on such a flimsy basis.

(2) The most valuable class I had in college was Philosophy 9, Practical Logic, at UCLA. One lecture dealt with logical fallacies, and one of those included was the fallacy of composition, which assumes that the whole is equal to the sum of its parts. It isn't; the arrangement of the parts matters as well, and it is often difficult to completely identify all of the parts in any complex activity, of which teaching is one. Imagine Mrs. Lincoln being asked to give her review of Our American Cousin: we can imagine her dutifully filling out a form about the play's plot, acting, direction, and so on; there would not likely be a category about one's date's safety included on the form, but the Ford Theatre's lack of security had a decisive influence on her experience of the play ("Other than that, Mrs. Lincoln, how did you like the play?"). Similarly, there are far more aspects to successful teaching than are being discussed in any of the reductive formulas being passed around, and the last thing we need is a formula-based simplification of the teaching profession akin to the simplified reductions in curricular learning we have seen as a result of the formula-based assessments that proliferated after No Child Left Behind.

(3) I have read that, wherever such formulas have been tried around the world, they have been intensely disliked, and it's hard to see why we should now expect an exception. Teachers' morale is already at a historical low, as recorded on surveys going back to the 1980s, and as we drive young people away from entering such a badly appreciated career, the only way to attract and retain people in a job whose intrinsic satisfactions have so badly declined will be by compensating teachers with higher salaries, and our states and taxpayers do not have big pots of money lying around waiting to be drawn upon to pay those higher wages.

(4) I used to teach Advanced Placement English as well as classes preparing those students who had already failed California's High School Exit Examination multiple times so as to help them pass that test, so I am well acquainted with teaching to high stakes tests; but many subjects (art, music, vocational, and others) are not oriented towards such tests, nor should they be; and we have entirely too much testing going on already, particularly in our increasingly miserable primary schools. A predictable consequence of this appraisal proposal will be having teachers whose job security and salaries are dependent on their students' scores not put students first and do what is best for them, but instead put themselves first by turning their classrooms into appalling test prep factories. I have already noticed a change in my younger son's school, from which he graduated this week and from which his older brother graduated eight years earlier. The older teachers in my older son's school knew how to teach interesting lessons and units that did not end in some sort of standardized test, and so the best part of the school year was after such tests were finished, in May. By contrast, the younger teachers and new leaders in my younger son's school seem to have no idea what to do once the testing is finished; the last five weeks of the school year are a round of parties and games and are devoid of learning, a complete ripoff of the taxpayers, and this apparently has met with administrative approval.

I know some will think I am switching sides here, but I believe reform has gone badly off the rails in the last few years, and I urge the teachers in Los Angeles and similar communities to reject this bad idea being promoted by well intentioned people.


  1. Just two quick points. It is not only that the teachers don't know what to do after the test, but that the students feel the course is over after the test. Just go on twitter and see what students were writing after the Florida EOCs, "Dawg, why I gotta go to school. We took the test." Can't really blame the students since they are called "End of Course" exams. An easy solution would be to move them to the last week of school instead of mid-April.
    Last point, I am not so sure the people promoting the use of value added models are "well intentioned." In Florida, it is being used to move people off of professional contract status on to annual contract status. You can read more about it in my blog

    1. Thank you, kafkateach. I agree with your point about moving end-of-course testing to the ends of the courses. In countries like France, which has one of the biggest end-of-course tests (but only, sensibly, at the end of the entire course of secondary education), the tests really are the last time the students go to their secondary schools. And when I mention "well intentioned people", I am thinking of my many friends who are behind this movement to include student test scores in teacher appraisals. I certainly don't know everyone.

  2. I agree completely. Once the test is done, (and now we test 4 times a year, Reading and Math), the teachers and the students are done.
    My students don't complain about working, because I continue to teach in ways that engage my students, so they are willing participants.
    However, it's sad to see teachers have one hour recess after testing is over, when there is still so much to learn, and teach!

    1. Lisa, we need more teachers like you! Teachers who are dedicated to genuine student learning. If I had my druthers, your students wouldn't be taking any tests at all -- that is, unless you designed and your colleagues had decided to do so, and were using them formatively, to better your students learning. Actually, I'm not against summative testing, but it should be limited to the end of the school year, and probably isn't a good idea in primary schools at all. I support the Finnish approach to comprehensive elementary schools.

    2. This comment has been removed by the author.

  3. Bruce, you make some excellent points; however, I disagree with on almost every one of them.

    (Disclaimer: I'm not the best at writing as I did skip 80% of my classes in high school; prepare yourself for cringe-worthy tense agreements and excessive comma use, among many other shortcomings that will I'm sure become very apparent -- especially as an instructor of English yourself.)

    Point 1: While some students do use their Scantrons as their own personal publicly-funded art canvas, I do believe we need to take certain factors into consideration.
    o First one being age: is a third grader as likely to become an impromptu artist as say, a junior in High School? Due to a lack of empirical evidence here, I can only use my own experience as a source of reference here. As a third grader, I cared a great deal about my scores and was diligent about having my score reflect that; however, when I found out that my scores don't actually affect me directly in anyway, I simply stopped caring and made designs out of my scores as well. How can we make a control for this? Perhaps by taking the average scores of the students who have previously taken the course over a ten-year ( or some other arbitrary number) period and use that as a baseline for each course. In doing so, we could account for factors that undoubtedly influence the test-taking habits of students. What class does the teacher teach, is a kid in an AP Physics course as likely to fudge their exams as their peers in a remedial Algebra course -- probably not. How does socioeconomic status come into play? Multiple studies have shown students of higher socioeconomic stance are likely to do well and prosper in an academic setting. Again, to compare an AP Physics course with that of a remedial Algebra course: are students in the AP Physics course as likely to reside within the same tax bracket as those in the remedial Algebra course? Because of this, the instructors of these courses should NOT be be rated in a relative way, but rather in an independent one based on how students have historically done in the same --or similar-- course over x amount of time.

    Point 2: I suppose I addressed this in point one. Not only should teachers across different disciplines be judged relative of one another, but schools shouldn't be either. The teachers at private school ABC that costs parents $37,000/year should not be be rated on the same scale as teachers in the high-crime rate, urban areas that require students to first walk through metal detectors before class everyday. The worry of being shot and/or bullied would certainly detract attention away from the academics and orient it towards personal safety. This would also be taken into account by making historical averages for the course be the baseline for each rating.

    Point 3: No doubt is moral important in almost every setting; however, do we definitively know what is the cause of the low moral? Are the young, new teachers being discouraged because of those that already have tenure -- and in some cases witness the abuse of said tenure status? No, there is not a pot of gold at the end of the rainbow but by rating teachers and hiring/firing accordingly, the more experienced, higher-paid teachers could be replaced by less-experienced teachers that can demand less pay and may ultimately provide better results. Something also worth mentioning is that if the average salary for teachers in Chicago is $76,000/yr, while the national average is $42,000/yr. Does that discrepancy not merit some form of higher standards and higher stakes for teachers in Chicago?

    1. Point 4: First and foremost, there is something intrinsically wrong with the current system where students cannot pass the exit exam. Not being able to pass that exam should be indicative that that particular student should NOT have been allowed to go beyond 8th grade, if even that far. Students are seemingly pushed through the system every year simply because of the convenience of it for both the student and the school system. How detrimental is this to the student and the system as a whole? The value of the high school diploma has been on a decline now for decades. Granting them to students that cannot "Identify the triangle" within a group of shapes or to be able to "Identify definition of a noun"--both being questions I remember gawking at myself as a student myself when the exam was first introduced-- expedites the rate of this happening while simultaneously degrading the meaning of being a High School Graduate.

      Honestly, to snark at public schools being "test prep factories" would be to ignore the current trajectory of academia in this country is on. Exams are administered for students to be granted credit for their AP courses; students then are tested again to get into college through either the SAT, SAT2 and the ACT; from there students are again tested throughout college --with most of those exams being multiple choice. From there if students decide to pursue interests involving graduate school -- something that is becoming more and more important -- they then need to take more exams(i.e. GMAT, GRE, LSAT, MCAT, etc). Certain careers require additional exams beyond this: after taking the MCAT and spending four years in medical school, students are only granted their license to practice medicine after taking a series of exams called the Step1, Step2 and Step3.
      At the end of the day, would most prefer to be in these test-taking factories and ultimately succeed; or to be in a dynamic setting that does NOT only focus on test taking, but ultimately end up falling behind?
      Of course, a system that is both dynamic and adequate in test preparation would be preferred, but to achieve that we would need to go beyond just Chicago and change the academic system entirely in our country, if not the world.

  4. Van, thank you for your comments. I will try to briefly respond to each. Your replies to points 1 and 2 are similar, and I don't think we disagree much, but my solutions would be to (1) develop fewer and better tests, to replace our annual cheap ones, and not to externally test at all in primary school, and (2) because I agree with your points about differences between courses and neighbourhoods, to not legislate teacher appraisal at the state level, but leave these matters to local districts. With regard to point 3, my main concern is with respect to the teacher supply pipeline: I believe that talented young people will shy away from entering a profession that is currently on the receiving end of so much abuse, and would never recommend my own children, both of whose parents are teachers, to become teachers themselves under current conditions. And as for your reply to point 4, the system you call for in your final sentence, one "that is both dynamic and adequate in test preparation", is one I have been striving to build for years now, and continue to do so; you may wish to click on the link next to my photo above to get a better sense of what I am trying to do.

  5. This comment has been removed by a blog administrator.