Like educators in nations whose educational systems outperform ours, U.S. teachers should be evaluated on our ability to teach and test what really matters.

The more I think about the current rush to set up quick-and-dirty teacher evaluation systems based primarily on results of misused standardized testing data, the more I realize that we are losing sight of the real prize: our children's learning of important things, and developing the professional expertise of our nation's teachers. That expertise includes being able to teach well and to measure student learning accurately.

During the season of testing-induced madness around the country, I'm reflecting on something I wrote a while back that was also quoted in our new book, Teaching 2030.

For years, one of my favorite classroom assessments has been to tie my opening activity fo the semester to my final exam (a composition). Students start the class by telling me (in writing) about their past experiences with writing, types of writing they have done (in and out of school), and their views on what constitutes good writing. For the final exam, I ask them to revisit that piece and explain what has changed as a result of their experiences in this class. They have to document examples of their own growth as writers. Thus, I have an exit essay that can be graded using the rubric adopted by the English faculty, but I also give the students a tool that guides them through a reflection of what they have learned and why. Student work samples like these (which can be digitized, stored, and analyzed over time) are also extremely valuable to me as evaluations of my own work and of how the class could be improved or changed.

The purpose of my classroom writing assessment is so students and I can measure the amoung of individual progress made by each writer. They all start from different points and end with various levels of proficiency as writers. I can generate reports, based on our school-adopted rubrics and learning outcomes, that show where each student is in relation to those outcomes and how far each student has moved over the course of the semester.

If the scoring instruments that I'm using within my classroom are of high quality, then I as an ethical professionally trained expert should be able to use those instruments to evaluate my students' work accurately and fairly. [Hint to policymakers and pundits: This is what 'good' teachers do]. Why is that too big a leap for our society to make in thinking about what makes an effective classroom teacher? We make exactly the same assumption for doctors, professional sport referrees, and auto mechanics. Do some of them make mistakes in their judgments? Yes. Are some of them unscrupulous or inept. Yes. Do we question the entire enterprise because it includes imperfect assessmsents or some poor performers? No.

We're asking the right policy question when we ask: "How can we better prepare the nation's teachers to conduct, evaluate, and use classroom assessments (formative and summative)---and to share that information in a format usable by parents, schools, employers, and other interested parties. This is the broad vision of accountability that we need.

I've spent some time mulling through the "Learning About Teaching" report of initial findings from the Gates Foundation funded Measures of Effective Teaching Project.

The team of researchers working on the project includes a wide range of views and perspectives, including some people whose work I highly respect. The team was very careful to address many of the major complaints against use of value added measures. For example, with assistance from RAND, over a two-year process, the teachers and students were randomly assigned, and no teacher in the study was made to teach in a subject, grade level, or school to which s/he had not been assigned during the previous year (when baseline data was collected). In the real world attempts to use value-added measures, this is often not the case.

To its credit, the team also sought to incorporate several other types of measures. For example, the study introduces an instrument to help measure teachers' pedagogical knowledge, not just their content knowledge.  Also, the inclusion of the TRIPOD student survey developed by Ron Ferguson of Harvard is unique, as is the survey of the participating teachers to collect data on their working conditions.

The team states that much more analysis and more conclusions are forthcoming, but I thought some of these preliminary findings are of particular interest:

1) "Teachers who raise student performance on standardized tests are also raising the students' conceptual understanding of the subject matter."

I prefer to state that in reverse: When teachers are allowed to teach students to understand the subject, to learn deeply rather than just to memorize content, the students will be able to perform well on any type of assessment or task that may be placed before them.

2) "...the above pattern implies that schooling itself may have little impact on standard reading comprehension assessments after 3rd grade. But literacy involves more than reading comprehension. As the Common Core State Standards recently adopted in many states remind us, it includes writing as well. In fact, English teachers after grade 4 generally focus more on writing than teaching children to read" (pg. 20).    Depending on how that point is taken by policymakers, it could help or hurt the movement for more comprehensive approach to literacy in schools.

Again, the research team admits that this work is just beginning and from this one study, we should be cautious about drawing too many important conclusions prematurely. The researcher in me finds their approach interesting; the teacher wonders when the other shoe will drop.