The Moral Majority
Last Monday's Teachers' Letters to ObamaTeach-In on Testing was jam-packed with information and new ideas. Lots of good questions, too--including one from Elaine Romero, U.S. Department of Education Ambassador, who asked guest expert Yong Zhao how teachers in China were evaluated. (Answer: by the things they do to increase students' learning and accomplishments, but not by standardized test data.)
Monty Neill of FairTest (the guy who's often been accused of never meeting a test he liked) took a very moderate stance: there are reasons to use the occasional benchmarked test, he said, but let's be certain that we're assessing the right things--and let's use the results in ethical ways.
The capstone of the evening for me, however, was a presentation by Doug Christensen, who framed the issue of assessment in terms of teacher professionalism. Testing should be teachers' work and responsibility, Christensen said. If student learning is our measurable product, it makes no sense to detach those ends from the means--the teaching processes of explaining content and setting learning goals for kids. Redefining the measurement of success, taking it out of teachers' hands, represents de-skilling of the professional work of teaching.
At first glance, the cycle of instruction seems pretty obvious--a bit of the old educational blah-blah, perhaps. But you'd be surprised at how easy it is to wander away from the essential elements:
#1) Knowing your students, evaluating their prior learning and capacity
#2) Planning and setting targets for learning
#3) Delivering instruction
#5) Looking at results, using them to determine what to do next
When testing is positioned as "accountability"--checking up on teachers' work--it short-circuits the instructional cycle, eliminating the quality assurance that the system is working. Instead, testing becomes a threat, and can disrupt the process in some pretty predictable ways, as teachers begin to focus the score as the product.
When teacher evaluations become dependent on test scores--a concept embedded in the Blueprint to re-authorize ESEA--the link between teaching and knowing whether your instruction has been effective is corroded. Someone else has taken over the process of evaluation. And what to do next becomes whatever it takes to raise the scores.
Are teachers proficient at setting high goals and developing valid assessments? Not all of them, certainly. But taking the process of assessing out of the hands of teachers does a disservice not only to teachers but to students. The teachers are not honing their own skill at developing consistently useful assessments, which help them tailor the next set of lessons as well as curriculum for the next year. Students are taking generic, high-stakes tests that may or may not reveal what they've actually learned and do little except compare them to other students.
The point that Christensen drove home was that assessment is not a primarily a technical issue, or a political concern. It's a moral issue. We have lost sight of meeting individual student needs. We are unclear on the purposes of the work we're doing--are we supposed to be helping kids create a toolkit for building a satisfying life? Or are we just worried about the numbers and how they reflect on our school? What happens when professional judgment is devalued and real estate agents use school achievement data to sell houses?
Speaking out about best practice in testing becomes an ethical issue--a matter of taking a principled stand for the well-being of students. It's the right thing to do.