I'm really not suggesting we stop using statistics bcause we do, without question, need a way of understanding trends. I'm a real big proponent of detecting and using patterns in time as well as space.
The problem is when we're given statistics without context as the sole support for some position or decision.
Today's NY Times has another article about the "miracle" of the Dallas, TX school system's statistical gains, exposing what looks more and more like a scam.
These graphs compare Dallas' results on the Stanford Achievement Tests with Los Angeles' and those of the nation as a whole. The national comparison in particular strikes me as problematic…not only does it show that a passing grade (70) on the Texas Assessment of Academic Skills would put you in a pretty low percentile on the Stanford Achievement Tests, but that the percentile you'd fall into has gotten lower over the past few years.
Of course they leave no child behind. They're backing up over them. And this is in addition to the scandal of their false reporting on the dropout rate.
And Dallas isn't alone in this. California, or San Francisco at minimum, has institutionalized the practice of getting kids they consider problematic to "test out" of high school rather than graduate.
Now, one may choose to dispute theses statistics. But doesn't that further my point, that context is vital?
I've long held that I was educated in spite of Dallas Public Schools, not because of them. That was the point that I became convinced that the average person is pretty stupid, overall. (I'm not sure if I was right, but I'm not sure that I was wrong, either.)There is another scandal brewing here in Dallas being covered by alterna-lefty reporter Jim Schutze about how DISD is hiding bad teachers by not releasing class-by-class numbers, putting out amazingly bad curricula, and just being generally incompetent.I don't think that they are any more or less incompetent than, say, Atlanta -- they just don't waste quite as much money as Atlanta doing it.