Hear it: The audio version
So one tests to figure out where to start and then they test again to see if they got it. Now this is very straight forward right? You teach addition and at first they say that 2+2 equals sandwich. Then after a rigorous unit in addition they say 2+2=4. Whew, they got it. Testing helped you see that they got it. You may now move on.
Then there is comparative testing. If you teach English and they can’t pass a spelling test before the studies, but then they can afterwards this all works fine. Furthermore, if you have six classes of kids all learning the same spelling words, you can rate them one against the other. Teacher A’s class has an average score of 95% while teacher B’s kids have an average of 75%, but teacher C’s group is looking at 45% as the class average. At a glance we know that teacher A rocks and teacher C sucks. We want our kids with teacher A. I mean the name alone should sell us on this class. We know that even with this very simple type of concrete scoring comparative testing is trickier than that. But really, is it? I mean these dunderheads can either spell or they can’t. “Back in my day we learned spelling and all my white, English-speaking, suburban peers could spell dammit.”
Now here are a few other ways of looking at the scores. If teacher B is teaching a class that has a group of mainstreamed, special-needs kids then the average is not telling us all that much. They will test differently, and the scores may or may not be altered to accommodate their innate abilities. Maybe class C has a majority of ESL (English as a second language) kids. While a student may have super spelling skills in Spanish or French, these do not translate all that well. Maybe class C has got four kids in it, all of whom are either there because they are prone to carry knives, or have been prosecuted for selling drugs. I am guessing that their interest in spelling tests may even be less than my concerns about attentive driving. Perhaps Teacher A got the five worst spellers to stay home the day of the test, or she gave a review moments before the quiz, or she gave candy motivators as rewards…
There are countless other scenarios that lead you to believe that this comparative testing has its limitations.
Later I will mention how this gets even more complex when you look at fields that are harder to test, like how well a kid plays trumpet, or does drama, or plays dodge ball. Regardless of the area we are testing, two things will hold true: Every single student brings their own story when they arrive at school;
m and the average score of the whole is a numeric average of the whole. Errk. What? Did that sentence just say that the average is the average? Well, yes. So the reader sits there and says, “I am so wasting my time reading this.” Well yeah, that would be true unless you factor in that national comparative testing is done by the government and they want all kids (Lake-Wobegone style) to be above average.
States want to be above average. Because states with smart kids are above-average states. On a cynical note one can only surmise how much of this is based on career advancement possibilities for state-level professionals whose test is the score average for the state. Superintendents, who make more than twice as much as the average teacher, want to be above average, and well they are. Making that much money for sitting at the district office makes them above average right up front. With all that time on their hands while they consider retirement or a political post, they need to think of ways to make their principals better so that they will also be above average.