It's not a factory.

Schools matter quotes a letter to the editor of the NYTimes:

"The Times notes that No Child Left Behind was unpopular "partly because it requires schools to administer far more standardized tests …". ("Experts Say a Rewrite of Nation’s Main Education Law Will Be Hard This Year," January 28). Education Secretary Duncan has announced that the Race to the Top national standards plan will include national tests linked to the standards, which means far more testing than we had with NCLB."

Okay. Let's stop here and discuss this for a second. NCLB was not vilified because it required testing, but because it mandated changes to the schools based on those tests, changes that would not fix the issue that those tests were thought to identify but didn't.

more:

You can't fix a school in the way you can fix a car. "My alternator jammed and burned up and I fixed it - I replaced the broken part." The same cannot be said for a school. Which teacher do you "replace because he's broken" if a subgroup didn't perform? Which aspect of that teacher is "broken" in the way my alternator was? Or do you toss the administration out on its ear? (must rethink this last)

Second, education is not an assembly-line with control of the raw materials and control of the process. If there is a substandard material, you can't toss it in the recycle bin. If the parts don't get stamped correctly, you change the procedure. Education, on the other hand, is a long process of stuttering advancement performed for a short time per day at arm's length on willing and unwilling students with varying abilities and desires, minds of their own, part-time jobs and full-time lives.

It's not a true assembly-line if the raw material is fighting you all the way and would rather not be there. The alternator in my car doesn't get to argue that the battery was at fault because it did its homework and that electrical testing equipment is flawed and its mommy was going to complain to the principal.

It's not an assembly line if the product itself, the supplier and the maker can be at odds and don't agree on the best way to make that product better (somehow value-added).

If there is a mistake down the line, you cannot "recall" the product and start over. If you notice something that went wrong, there is no real way to apply those lessons directly to the next batch. Your product research is flawed and rarely gets to you in any meaningful or useful form, anyway.

You can't control the product: it has it's own motivations and desires.
You can't control the preliminaries: other teachers, parents, peers, environment mold that kid before you have a chance.
You can't know the parameters: the "data-driven" decisions are based on data derived by others and is very suspect.

Here's a couple rhetorical questions for you:

1. Can a teacher be responsible for students he's never had? I'm one of many in this department. Every year, they're all new faces with new dynamics.
2. If I'm teaching pre-calculus and calculus, do I bear part of the responsibility for a broken school? My kids have all passed the test (if they cared to). If you give me a calculus class, I'll teach calculus to them.
3. How about placement? How can you consider me "broken" if I got that kid in 10th grade in a course he couldn't handle? Am I responsible for his failing the test?
4. Why is it that math is the touchpoint for success for all students? If you gave those tests to adults, you'd get the same 30% passing rates yet those adults are successful. The art teacher and the history teacher may be a much bigger influence and might be far more important or interesting to that student.