Monday, May 25, 2015

Bell Curve Beatdown

If you are only going to read one blog post this month, it should be this post by Jersey Jazzman about standardized testing. Come for sentences like this one:

This can't be stressed enough in the testing debates: we design tests not based on objective criteria, but on socially constructed frameworks that assume some of us are above average, some of us are below, and most of us are in the middle.

In other words, standardized tests are not designed to answer the question, "How well do these students understand this material." The test manufacturers believe they already know the answer to that question-- some students understand very well, most understand moderately well, and some don't understand at all. If the test results do not confirm that pre-determined result, then the test must be defective, and we have to redesign it.


It's hard to state how contrary that is to common teacher sense. If every student in my class fails a test, I know I need to reteach because I didn't get the material taught. If every student in my class does well, I do a little happy dance because we all nailed that stuff. But in either case, a test manufacturer just blames the test and sends it back for redesign.

And the test manufacturer believes that curve can never change, creating a Sisyphusian task -- we are supposed to make all students above average, and we are supposed to prove it with an instrument that will always, must always, show that only a few excel, a few fail, and most are average. In other words, the standardization crew demands that teachers change the bell curve when they themselves believe that the bell curve can never, ever be changed. Or as Jersey Jazzman puts it-

We're insisting that all children demonstrate high performance on a test that, by design, only allows a few children to demonstrate high performance.

Go read the post. It's a great explanation in plain language of the technical reasons that the standardized testing game is rigged for failure as well as why you have had the nagging sense that the whole testing business is crazy-making and not actually measuring educational effectiveness at all.
This can't be stressed enough in the testing debates: we design tests not based on objective criteria, but on socially constructed frameworks that assume some of us are above average, some of us are below, and most of us are in the middle. - See more at: http://jerseyjazzman.blogspot.com/2015/05/standardized-tests-symptoms-not-causes.html#sthash.39shNQ3R.dpuf

1 comment:

  1. I've never thought that the bell curve made any sense. It's totally artificial, and this post explains why it makes no sense. It's the same idea as the bottom 5% of schools being labeled failing, but even if you got them into the top 25%, there would still be a bottom 5%.

    Instead of using norm-referenced standardized testing, we should be using criterion-based assessments. I've always thought that if all my students do everything I ask them to and work as hard as they can, most of them should get A's and B's because if I teach to all kind of learners, the learning should be accessible to them and they should be able to meet my criteria.

    JJ touches on another issue, a sociological one: why does there have to be an elite class? That's also totally artificial. If a person works hard, as hard as anyone else, no matter WHAT it is they're working at, shouldn't they have the right to a decent wage, decent housing, good health care, workplace rights and the time to raise their children?

    I

    ReplyDelete