The big headline from the recent Gates/Scholastic survey of teachers is that only 28% of teachers see standardized tests as an essential or important gauge of student assessment, and only 26% say they are accurate as a reflection of student knowledge. Another question reveals part of the reason this may be so - only 45% of teachers think their students take these tests seriously, or perform to the best of their ability.

We have been stuck in an accountability rut for the past decade, with most reform initiatives revolving around test scores of one sort or another. Even the concept of student learning has been subtly redefined to mean "achievement on end of year tests."

There are several assumptions that have driven this.

One. Standardized tests are an adequate, if imperfect means of measuring learning.

Two. Teacher performance can be captured by the growth of their students on these tests.

Three. There is a significant number of ineffective teachers who are evading detection by current evaluation practices, who will be accurately identified when we begin using test scores or VAM systems as part of our evaluation systems.

These assumptions are fundamentally challenged by what this survey reveals.

First of all, if fewer than half of the teachers believe their students take standardized tests seriously, then how can these tests be used as the basis for any important judgments at all?

But there is another dimension to the survey that is even more important, because it reveals a doorway to what we all might agree is our common goal, helping our students learn more.

A total of 92% of the teachers surveyed reported that formative, ongoing assessments were absolutely essential or important in measuring student achievement. Furthermore, more than 90% of teachers actively use student performance data of this sort to differentiate instruction, target interventions for students in need of help, and otherwise improve their teaching.

Clearly teachers are keeping track of learning, and using this information to guide their instruction. What is more, we are doing it far more effectively than the state is, through its annual tests.

Another piece of the reform storyline is that teacher opposition to performance pay and VAM-based evaluation systems is based in a fear of accountability. But the survey showed some interesting results here as well. Fully 85% of teachers agree that student growth over the school year should be a part of a teachers' evaluation! However only 36% agree that we should rely on standardized tests for this data.

Is it possible that teachers do not fear accountability, but rather the misapplication of clumsy and unreliable measurements like standardized tests, that their students do not even care about?

As I wrote recently, it is possible to incorporate student learning into teacher evaluations through a variety of indicators, which do not have the terrible effects we have seen come from attaching ever-higher stakes to standardized tests.

We need to be careful that we do not simply trade one set of tests for another. Many districts are implementing frequent "benchmark" tests, and calling this "formative assessment." This is a sleight-of-hand trick. If we attach high stakes, such as teacher evaluations and student grades to these tests, they could become even more crippling to good teaching than the annual standardized tests. Formative assessment is, as Dr. Myron Atkin explained here a while back,

...working with a student, or a group of students, to develop a course of action that helps bridge the gap between current student knowledge and the desired educational goal. Providing feedback that is usable, detailed, and often individualized is at the heart of this kind of assessment. Formative assessment, so defined, is a pivotal element of everyday classroom teaching. It occurs throughout the school day. It requires collaborative involvement of both teacher and student. And it isn't something purchased from a vendor that can be used in an identical fashion anywhere, like an instruction book or a cooking recipe.

Strengthening this sort of assessment ought to be at the heart of professional development, and the information we gather about how our students are learning should be brought to bear in our evaluations as well. 

How about a different set of assumptions to drive improvement:

One. Teacher-created, classroom-based assessments are the richest source of data about student learning available.

Two. Teacher performance can be captured by an evaluation system that looks at how teachers are assessing learning, as well as what those assessments reveal students have learned.

Three. Teachers are seeking to be more effective, and embrace real accountability, when it comes with responsibility, autonomy and support.

The reforms we would see flow from these assumptions would look very different from the ones we have been struggling over for the past decade.

What do you think? Should teacher-driven formative assessment data be honored? Would a shift in our assumptions create a new direction for reform?