Merit pay based on student test scores is a flawed idea

(The following post is taken from my Daily Kos diary.)


It would be a dream scenario for any high school basketball coach- arriving at practice every day with an athletic front line, averaging six feet, nine inches tall (and able to handle the ball) and a backcourt consisting of a lightning-quick point guard who can penetrate and dish off or shoot, and a nothing-but-net three-point shooter.
And even better, all five players are battle-tested seniors with three years of varsity experience.
A coach could guide a team like that to the state championship.
After that, there might be a letdown. The next year’s squad is raw, inexperienced, and does not have anyone over six feet, four inches, and no natural point guard.

Naturally, no one with any intelligence would expect the coach to be able to top the previous year’s record.
That would be impossible…yet the same thing is expected of teachers in the United States every year.

When pontificating politicians and pundits rail against the so-called “failing” American public school system, their answers to repair it always are led by an over-reliance on the results of standardized tests. In Missouri, where I teach eighth grade English, and in many other states, our elected officials are indulging in the tried-and-true national pastime of election year grandstanding at the expense of the teachers. Efforts are being made in our state to offer merit pay, based on standardized test scores, to teachers, on the condition that they give up tenure. The legislature approved a bill last year allowing this to be done in St. Louis schools.

And while I firmly believe teachers should be held accountable for the education received by those in their charge, standardized test scores are not the way to do it.

Too many variables go into students’ success on standardized tests.

STUDENTS’ HOME LIVES- While teachers work miracles every day in our public schools, trying to convince students who come from broken homes (or who have no homes), are physically, emotionally, or sexually abused, or come from homes where the parents are abusing drugs or alcohol that algebraic equations, parts of speech, and frog dissection are necessary for them to get ahead in life is virtually impossible.

PREVIOUS TEACHERS- A student’s learning can easily be derailed by running into a subpar teacher, or perhaps one who has to be gone for an extended amount of time due to sickness. The merit pay systems being considered never take into account that the educational process builds from year to year. If there is a detour along the way, it can have a negative effect on a student. Should it also be responsible for a teacher losing a job or a pay increase?

VERTICAL SCORING- Returning to the analogy that opened this post, we are not testing how much students learn from year to year, which would seem to be the only measurement that makes sense; we are comparing this year’s students with an entirely different group of students who took the class the previous year. Just as you have basketball players with different characteristics and capabilities each year, the same holds true for students. It not only is an unfair comparison, but it is a comparison that is absolutely meaningless. Yet this is the comparison politicians plan to use to hold teachers’ livelihoods hostage.

While teachers (and the American public) would be ill served by making these comparisons the basis on which our schools are staffed and our teachers rewarded (or shown the door), there undeniably has to be accountability in education.
So how can it be done in a way that works for the teachers, the taxpayers, and above all, the students?

USE LONGITUDINAL SCORING- Instead of comparing apples and oranges, determine how much the same students have learned from one year to the next. While there would still be flaws in this system, it would still offer a realistic picture of whether a student is moving forward or digressing in his or her education.

JUDGE TEACHERS ON THREE-YEAR RESULTS- If test scores fall by large amounts, take action, but there needs to be a built-in protection that keeps teachers from being penalized for years that they don’t have those all-star lineups of students available.


TEST STUDENTS AT BEGINNING AND END OF YEAR- If we test the students on what they are expected to know at the beginning of the school year, and then test again at the end to see what they have learned, that would be a far more accurate method of determining what learning has taken place.

ADD OTHER MEASUREMENTS TO THE MIX- Administrator evaluations and peer evaluations could be used.

Sadly, there is no perfect way of evaluating teachers. That being said, the biggest mistake we can make is to base everything on test scores. If we take that step- and we are leaning dangerously in that direction- our schools will be all test preparation all the time, and the joy of learning, a joy that has produced our best and brightest, will be only a dim memory.