March 03, 2013

Testing Teachers

There has been a lot of controversy about whether student assessments should be used to evaluate K-12 teachers.  The media is full of debate about this topic (e.g., Wall Street Journal) and the Seattle Times has had many editorials pushing for the use of assessments such as MAP to evaluate teachers.  And several foundations, supported by rich contributors, like the League of Education Voters and the Gates Foundation, are pushing teacher evaluation through student assessments.

I would like to argue that using student assessments to evaluate teachers not only has issues, but is putting the cart before the horse.   First, we need to test teachers in a robust way to evaluate their knowledge of the subjects they are teaching and only allow teachers with strong subject knowledge to teach.

If you are going to put a teacher into a classroom, he or she must have a strong knowledge of the material they are teaching.  In fact, a teacher can only be effective if he or she knows that material at a considerably higher level than what is being presented in class.  A deep knowledge of a subject allows one to simplify concepts accurately, to understand how the material will fit into the student's future learning.

I know this is true because I put it to the test in every class.  When I teach atmospheric sciences 101, I do a far better job because I have a deep knowledge of the subject, allowing me to rephrase the material, correct errors in the textbook, and to provide better answers to insightful student questions.  K-12 teachers need to be able to do the same.

Thus, someone teaching elementary school math must completely master at least middle school math, a middle school teacher completely master high school math (algebra through calculus), and a high school teacher must be proficient at college level math (e.g., calculus, linear algebra, differential equations).

Is this the case now?  The answer is certainly no for many teachers..   Many elementary school teachers are uncomfortable with even elementary school math, and many don't have facility with middle school math (e.g., working with fractions, easy algebra).  Believe me, I have talked to quite a few teachers that have confirmed this and noted this in my kids own teachers.  Middle school math teachers are often volunteered for the job and lack the necessary background.

Competency in the material they are teaching is the LEAST we can expect from teachers.  Why not ensure this foundational knowledge?   Lawyer's have bar exams, certified accounts have exams, engineers have exams, and doctors have medical boards.  Even hair stylists have to pass competency tests.   Doesn't it make sense we do the same for teachers with whom we trust the future of our children and society?

Yes, Washington State does have a teaching competency exam that all teachers must pass (WEST-B), but that exam is too easy and with too low of a passing level to be effective. 

Teachers should be seen as educated and trained professionals, worthy of a substantial degree of trust.   But first they must demonstrate they deserve this trust by showing their mastery of the subjects for which they are providing instruction.

Lack of mastery and deep knowledge of the subjects have impacts on curricula that can be profound and negative.   For example, elementary school and middle school teachers who have weak math backgrounds are drawn to "reform" or "fuzzy math" approaches that are light on content and rigor, but heavy on discussion, group work, and "real life" examples.  And some of these teachers become curriculum coordinators and "math coaches" in their district, pushing the same weak math books.    This is how terrible books like Everyday Math get adopted and used for years, leading to declining student skills and huge rates of remediation in high school and community colleges.

I think it is extraordinary that challenging competency exams in K-12 are generally absent in our country.   A lot of the blame for the low priority on subject knowledge needs to be placed with college and university schools of education, including, sadly, my own University of Washington College of Education.  Such colleges of education generally downplay subject knowledge while stressing arcane educational theories and social engineering.  If you want to prove this to yourself, look at the curricula of some local schools of education or talk to some teachers...and be prepared to be depressed with what you learn.
To illustrate, the only requirement for math competency at the University of Washington Elementary Education Program is that the student pass (credit, no credit) Math 170 (Math for Elementary School Teachers), a class that stresses reflection papers and class projects, and uses a looseleaf textbook (Reconceptualizing Mathematics, by Sowder, Sowder, and Nickerson ) that stresses that math instruction "requires far more than simply knowing the “math facts” and a handful of algorithms."   When you see a book putting down math facts and algorithms, and stressing the "deep knowledge of mathematics" that is offers, you know you got a problem on your hands.  In fact, looking through the book it is clear that few students would gain mastery of elementary school mathematics by reading it.  Science competency for UW future elementary school teachers?  Don't ask..

Let me be clear---insuring subject matter knowledge is just the first step for insuring a quality teaching experience in every classroom.  Teachers also need the right personal skills, empathy, and knowledge of effective teaching approaches, coupled with reasonable size classes and sufficient classroom infrastructure, to insure each student can learn at his/her potential.

Assessing teacher competency by using student performance on standardized tests is problematic and difficult for many reasons, including the variations in student demographics and family background, the tendency to teach to the test, and other reasons.  Why not start with a step everyone should be able to agree on?   Test teachers to insure they have completely mastered the material they are teaching at a level far beyond the level for which they are giving instruction.

It is just common sense.


  1. in my personal experience, a teacher's social intelligence, charisma and empathy are more important than their knowledge base. many accomplished teachers have no idea how to convey their ideas.

    i have nothing to back this up, but i would guess that a great communicator with a vague grasp of the subject matter would be a better teacher than a socially inept expert.

  2. Great commentary.

    Oops, teachers unions and one party rule of WA for decades. You can forget about any meaningful reform.

    Oh, and make sure to vote for more taxes so teachers can retire at 55 while the rest of us work until 70 or older.

  3. Reform has happened JeffB, teachers now retire at 65 in WA state.

    Knowing your content is important but I disagree that knowing kids and how to teach them is not equally important. There are more skills to learn in school than content and group work is an important life skill.

    There are many who know their content deeply but cannot deliver well so their students struggle to learn it or have to teach themselves. Think college professors, so we have to help students learn to learn.

    I also believe there is more to math than practicing algorithms and doing things by hand that calculators can do. Once I learn how to divide the calculator can do it so that I can tackle more difficult problems. Children not only to understand math beyond number crunching but also to learn to love, or better yet, continue to love math.

  4. There is nothing so articulate and charismatic than fluency in what is being taught, and exemplified in the teacher's own experience. It is an utter waste of time for a pleasant teacher who knows no French to "motivate" students to learn French. The same is true for math, for music, for history, for writing, for basketball. One does not "reason it out" with help on homework from family members. One learns by knowing what elements to drill on and memorize, what elements to practice in performance, and what portions come by reason with a practiced intuition and familiarity with the material.

    It only takes one really bad--indifferent, incompetent, blundering, terrorizing--teacher in a string of good ones to destroy a subject--socially, conceptually, pragmatically. So, yes, test on the subject matter 4 years beyond the instruction, and test as well for social competence, and make sure those designing and doing the testing have already scored a high pass.

  5. There are people who still fail the WEST-B, or at least freak out about the math section. A friend lorded over the fact that he aced the math section while I missed only one question.

    However, there is also the WEST-E, which goes a bit more in depth in regard to the content the teacher will actually cover. This test is also ridiculously easy.

    Western Washington University provides a somewhat more robust math education program for elementary educators, offering a couple methods classes catered to the subject matter they'll cover. I think there was one for secondary and a couple for elementary.

    When I went through the secondary ed math program at WWU, I felt underserved with a single math methods class, yet five classes on teachable moments and textbook analysis. But I was also reassured repeatedly that "you're going for math, you'll be fine."

  6. A little off topic

    Teachers, not the state or federal government ought to be testing and grading students in math, reading,writing, perhaps history etc.

    They should be using weekly quizzes, and use those to assess at level of proficiency. These tests or quizzes should be so frequent that they are not a cause of stress.

    Principals should occasionally and unannounced send a proctor to do the weekly tests, and replace the teachers tests with the principals tests. Purpose: confirm the teacher is accurately assessing students.

    The State should very occasionaly send in a proctor and do the same, purpose to see if the principal is accurately assessing.

    Statisticians could determine how frequently principals and the State assess. Fraud would be almost impossible. Stress would disappear. Data would very likely improve.

  7. Cliff, I think swapping tests or test subjects is not the issue.

    Compared to college level courses, many students in public schools are not motivated to perform well, particularly in lower income schools. Every teacher can tell you how many of her kids show up "ready to learn", and in lots of schools that number is less than 50%. Kids not ready to learn sometimes haven't eaten, often haven't slept near enough, or are so attuned by TV and video games that school is hopelessly boring and remote to them. To get past all that, it's necessary to form a relationship with the kids so they want to perform well. That mean strong social skills, being a good listener, knowing how to lay down the law, and being a compelling communicator.

    Beyond that, some students are actively disruptive in a way that's unmanageable. In Seattle in particular, the non-administration of special ed coupled with the capacity crunch means that lots of kids are being placed in regular classrooms that shouldn't be. They are set up to spectacularly fail, and in failing they can wreck the class for everyone else (including the teacher). In many schools teachers are evaluated by how much trouble they make for the administration, and that means not sending kids to the office. Every teacher has stories of the "bad class", which was usually triggered by 1 or 2 kids that were unmanageable.

    Beyond all that, you need a teacher that knows their material and has lesson plans that engage the students to become creatively involved in learning more about it. Unless you are teaching APP / spectrum though, that part of the job is secondary.

    If you're looking for a test that tells you when there's a problem, I'd suggest the evaluation teachers have of their principal, the administration, and each other.

  8. Would-be teachers are supposed to get math in high school, then more math to get their BAs. It's really sad that we still need to test them on middle school level math.

    For the most part, middle school math and science teachers have more math and science in college -- graduating with degrees in math or a math-based science.

    Do math and science teachers get higher pay than humanities teachers? $70,000 salary (at the top of the career) would look pretty good to many humanities graduates, but isn't that great for a good math, science, or engineering graduate. So the best of them tend not to go into K-12 teaching.

  9. An elaboration on the tests required to get your math certification here in Washington. Passing the WEST-B math test is required for anyone starting a teaching program (aimed at any level) and must be passed before any student teaching (or what ever phrase is used by that particular program) is started.

    There are different WEST-E exams that are required, depending on which teaching endorsement is being pursued: a general math for a k-8 endorsements, another for "mid-level" math (to teach math 7th & 8th grade math) and a third for "mathematics" (to teach HS math). The last allows you to teach math through high school, in theory including calculus.

    Coming from an engineering background (20+ in aerospace), I must admit I was dismayed by the shallow knowledge level being taught in our courses, but I will also admit that it was an online program, independent from my Masters in Teaching program.

    Are these adequate? I know that passing the general level math WEST-E prevented several of my Masters in Teaching cohorts from getting certified in Washington.

    I don't know the passing rate from my Math certification program. The program I just went through was identical for both MS & HS levels; the discriminator was the WEST-E that was passed. With my engineering background, I passed the Mid-level math WEST-E before I even started the endorsement program. In retrospect, I could have probably passed the HS math WEST-E at that point as well, but since the program was required to get the endorsement, I thought I'd take advantage of the refresher. As it was, when I did take it, the testing facility 'glitched' my calculator so that it was unusable; I passed anyway, but with a much lower score than I would have had if I hadn't had to take time to do all my calculations by hand.

  10. Cliff,
    Love your commentary. I think you hit the nail on its head. There should be a rigorous qualification process for teachers to teach math. I would also proposed that math teaching should be tiered (or seggregated) by kids who are talented in math and therefore do not need to "discuss" math and those who need fuzzy math. Why don't we allow both approaches and let the kids/parents select what's good for them. For those people who love fuzzy math, let them have at it.

    I agree that we are wasting tax dollars. The problem however lies in the administrators, who determines the curriculum and you might be interested in finding how their salaries (and your tax dollars). Teachers are not who are wasting our tax dollars.

  11. I just went through the Washington teacher testing process, and as RainyCity1 explained, the WEST-E tests are much more comprehensive.

    But, I think it was a little disingenuous to present the example WEST-B question that you did. There were a few questions at that level on the WEST-B, but they ranged from that level to basic algebra and trigonometry. Teachers have enough to deal with without the insinuation that they struggle with questions about triangles.

    Lastly, all teachers should be able to competently and mathematically explain shapes to, say, a 5 year-old. I've known a lot of math, physics and environmental science graduate students that would have trouble giving an off-the-cuff explanation of, say, a rhombus, that could educate and engage a kindergartner. I'm also willing to admit that even though I have multiple degrees in biology it took me a while to remember some of the very basic human physiology on the WEST-E biology test. The triangle question was an example of the easiest questions on the WEST-B, and I'll bet that no one got that one wrong (95% confidence interval), but, it's also an important question to ask.

  12. Thank you, Cliff. This report about the fractions questions on the TIMSS test shines a light on the area that seems to be causing the most trouble for kids and that is not just a little confusion, but a complete misunderstanding of fractions. Super interesting how they can see by which wrong answers were selected (especially U.S. and Finland) that only 29% of our kids understand the basics of fractions. Without fractions, decimals and percentages (the 3 weak areas for teachers and thus student learning) kids don't progress through high school and college without trouble. Regardless of what the ed colleges are doing, school districts could screen the elementary school teacher applicants for math content and pedagogical skills pretty easily. Just through attrition alone, new hires with math skills could turn things around in a few years. I really hope something like this can happen. Thanks, again.

  13. Wonderful post Cliff. I am a convert to your approach. I wonder if it could make it past teacher unions?

    The K-12 teachers I remember most were the talented ones. My high school jazz band instructor was a saxophone player ... and he played violin for the Wash DC symphony.

    I know that's overkill for the argument here, but the idea of teachers than know their craft is a great, and maybe doable, start.

  14. Without taking into consideration the WEST-E, the argument you're making is factually inaccurate since the WEST-B is only supposed to cover basic skills and is given to everyone who wants to be certified whether they hope to teach pre-school or high school. The WEST-E became the standard in 2000 and the content is set by the Professional Standards Board. The test covers a lot of ground and does require more advanced knowledge in your content. Looking at the "Prepping for the test" section it looks like the elementary test covers algebra and geometry which would seem to address your desire to have elementary teachers knowing middle school concepts.

    I also wanted to point out that WA has the fourth highest group of National Board Certified teachers in the country. Part of receiving this certification involves responding to a constructed response exam (not multiple choice) that covers a wide variety of content and pedagogy questions. For secondary teachers, the focus is on the content while for elementary the focus is more on pedagogy. While it is not required to pass each of these questions to become certified, you do need to pass or come close to passing all of them if you receive certification and we have a very large group of teachers who have achieved that. I would imagine that in districts that support National Board candidates (Seattle being one of them), you probably have at least one in most department or grade level teams. They should be given leadership opportunities to help keep everyone on the same page and pushing for increasing student success.

    Expertise and learning cannot be fully measured by an exam. Often math teachers, particularly secondary math teachers, were good math students and thus I'm sure could score well on the test with some review. Knowing the math doesn't mean they know how to help students learn who don't have a natural aptitude in the subject. To learn how to do that is really the career task of teaching and one that can't be measured on a singular test, whether that test is given to the student or the teacher.

    As a Washington state teacher, I invite high standards. I have them for myself and I have them for my students but please at least try to fact check before you jump on the teacher-bashing bandwagon.

  15. I have three points to make:

    1. The testing process to become a teacher is much more than the West B. There’s the West E and the EdTPA.

    2. Math content knowledge is sorely lacking, but there is (at least at UWB) a much more comprehensive look into how math works, the multiple ways to solve equations, and why it works. Dividing fractions is simple, but why it works is pretty interesting. Can you prove Pythagoras’s Theorem Algebraically Geometrically, and beyond? Personally, I have gone beyond Calculus, Linear Algebra, Matrix Algebra, but deep understanding of math isn't enough to be able to teach it. How many college professors in mathematics could teach fifth graders?

    3. What is the purpose of teaching math? Are we training excellent calculators? We have programs like wolframalpha that can solve very complex problems. There are innumerable computer calculation tools. Is computation more important than problem solving? Understanding why we use mathematics is becoming more and more important. Algorithms and math facts are on the way out, because they don’t take number sense, and instead rely on following steps that calculators already far exceed in speed and accuracy.

  16. Testing teachers in their subject area, what a novel idea, but isn't that what the colleges of education are supposed to do?

  17. Cliff - You are my weather idol. But you are a little misguided on this one.

    I was typing out a big long response and now I dont think it matters. But, you are welcome any time to teach a math lesson at one of our classrooms amd see what is really needed to be an effective teacher. It is too easy to judge through a window. Go substitute, get your hands dirty, and then I'll listen.

  18. When I read your post, I though - OK, it's hard to argue against teachers having a good grasp of the subject matter they are teaching. I thought I would read through the comments to see if anyone would give it a try, and was surprised to see a few had gone there.

    Mathematics isn't exactly a dynamic field until you reach PhD level work, so we're not asking teachers here to hit a moving target which would require life-long learning habits, just learn to a specific target level so that when a student asks a question that doesn't come directly from the book or lesson plan, they can answer with authority and provide insight.

  19. A few of you mentioned the WEST-E test, which is more rigorous. From my reading of the requirements, an elementary school teacher does NOT have to take that exam.

    And the comments about "deeper understanding " of math provided by reform/fuzzy math...I think that is without merit. Reform math has produced students without the skills and without good math intuition/understanding. Math is a tool that folks like me use to solve problems. It is a language for working with scientific problems. I suspect most of the "deeper understanding" folks have never really used math on a professional level...cliff


    Link to Zeichner article. Available via UW libraries

  21. Having suffered too many classes in history or math or some other subject being taught by graduates of physical education programs, I think the teachers indeed be tested and the administration assign teachers to classes for which they are trained.

  22. I should be more clear. I am all for paying teachers more and administrators less. After all teachers are the product. But only when they test in to it with full merit. Because the days of entitlement and the rest of us carrying the burden of the public sector are waning. And as for retirement at 55. Still a problem because many long time teachers and other public employees still qualify for PERS. I'll pay their retirement right after they pay for mine. Oh wait, I have to save and provide for myself, unlike many I know who left the public sector early and are now living lavishly with multiple homes on the rest of us.

  23. I agree with the premise that teacher's need to have extensive content knowledge, but think that a sole focus on this component of the complex, skilled, and challenging work that is good teaching oversimplifies the situation.

    It is generally acknowledged that effective teachers must not only have content knowledge, but also:

    General pedagogical knowledge, defined as broad principles and strategies about classroom
    management and organization that transcend subject matter;

    Curriculum knowledge, with particular grasp of the materials and programs that serve as
    tools of the trade for teachers;

    Pedagogical content knowledge – that amalgam of content and pedagogy that is
    unique to the province of teachers;

    Knowledge of learners and their characteristics;

    Knowledge of educational contexts, ranging from the working of the group or classroom, the
    governance and financing of school districts, to the character of communities and culture;

    Knowledge of educational ends, purposes and values, and their philosophical and historical
    grounds (See Schulman, 1987).

    These categories speak to the complexity that is inherent in preparing teachers for the challenges that face them in today's changing classrooms. Expertise in content knowledge is critical to good teaching, however, it is not enough to ensure effective teaching for all learners. Teacher education programs face the extraordinary challenge of preparing teachers in all of these areas with increasing pressure to shorten programs and with (as with all higher education) decreasing resources.

    The most productive discussions of reform will recognize these difficulties and bring together communities, policy makers, and educators to jointly make the decision about the future of public education.

  24. I'm currently in a teaching program where I'm receiving my BS along with my education degree.

    If you want to instill a test like the bar exam there must be an incentive to reach this higher competency. Pay teachers for what they are worth like you would a doctor or a lawyer.

    Another aspect to consider is the lack luster teaching methods I've received from college instructors. Some are stuck in the stone age of what is a presentable way to bring material. I guess that is why they call them professors, they simply profess.

    Also until the common core standards are established their is still an incredible amount of inconsistencies of what is to be taught and achieved by students.

  25. rainycity1 is correct in what it refers to the WEST-B and WEST-E. I have an engineering background, too, and wanted to get into teaching. Passing the WEST-B was required to get into the teaching program (or to pass it before the student teaching was started, depending on the program). The WEST-E is to get the specific "endorsements" (i.e., a k-8 teacher who wants to have an endorsement in math needs to pass the math WEST-E, an endorsement in Physics requires another WEST-E, etc)

  26. Ms. Kirking,
    You have left several message...I want to be clear...content knowledge is not the only requirement for good teaching, but I would suggest it is critical one that is not given sufficient emphasis...cliff

  27. Using student assessments to evaluate K-12 teachers is not for the purpose of evaluating competence, but annual performance. Every employee needs to have a proper annual performance review that is based on what you did that year. Student assessments is one way to perform that annual evaluation.


Please make sure your comments are civil. Name calling and personal attacks are not appropriate.

Lightning Returns to the Pacific Northwest

 Lots of thunderstorms, some approaching severe levels, have hit eastern Oregon and Washington during the past day.....and there are severe ...