By Diane Ravitch
Deborah Meier responded to an earlier post about the growing movement against testing.
I said that testing is misused now, as a way to punish (or reward) students, teachers, principals, and schools.
Deborah says that standardized testing is in itself problematic, for all the reasons she gives here:
Standardized testing was not intended to serve all its many misused purposes.
But even as a measure of an individual student it has enormous imitations. The score is "accurate" only within a very broad range==a so-called year on or off, and in 1/3 of the cases wider than that. That's its statistical range of reliability.
However, since it only covers a certain percentage of the skills and aptitudes involved in reading or math it may not reflect how well the students have done on what they primarily DID study.
Nor does it take into account that some work best under timed pressure, and some worse. This wouldn't show up on reliability studies. etc etc.
In short it's even a lousy messenger regarding how one particular child is doing--and rarely useful diagnostically unless the teacher had the feedback--actually items answered wrong and right--immediately so he could match it with the kids, do some follow-up interviews, and try other ways of reteaching it.
That's what good in-class assessment aims at. Such assessment is not, honestly, intended for the purposes of judging students, but rather of judging the effectiveness of one's teaching. And, if we don't keep peering over teacher's shoulders, it might discourage cheating--and might replace all the more critical daily observational and relationship-building skills that good teaching demands.
The greatest moments are when a student actually says--I'm totally lost, or I'm confused, or why doesn't this way seem to work for me, or.... It's ignorance displayed that we should cheer about, not ignorance disguised.