Arne and Michelle vs. Larry: The Statistical Battle

By Robert Plants
Editor, Schools for the 21st Century

I opened my newspaper this morning to an article titled “ACT Scores Dropping but More Students Are Prepared for College.” I asked myself how is this possible when other reports say that schools and teachers are not preparing students for future learning.

But I’m getting off my topic, which is the research-based finding that “more than 90% of the variation in student gain scores is due to the variation in student-level factors that are not under the control of the teacher.”

Another interesting note from those who have evaluated value-added methods and particularly the one used in LA is that there is opportunity for as much as 26% error in teacher ratings. If you want to put it another way, 26% is a large labeling error to make regarding someone’s chosen vocation. It sort of opens one to litigation in my mind.

Add to this discussion the idea that a teacher’s credentials would help produce better results. In my personal experience, that would mean the consideration of yet another test score, the Praxis, which is nothing more than a glorified ACT test.

In case you missed it, we have a President who is content on relying on a big city superintendent to run his Department of Education. So we are now going on almost ten years of NCLB. All we have to show for it are tests, tests, and more tests and standards that set minimums but are treated as maximums. Now that NCLB’s failures are becoming apparent (it took them ten years to notice), we are turning to the real root of the problem — those darn teachers.

I think the thing that scares me most is that we have Arne Duncan and Michelle Rhee, whose solution as urban administrators is to just fire the teachers who don’t measure up. That’s right, fire them based on accountability systems that violate basic tenets of good, solid research, that is, the use of multiple measures of data and academically accepted measures of what is significant and what is not.

Okay, so you say that I’m just criticizing and not making any contribution to what might be more effective than student testing, teacher’s value-added scores, or the use of pre-service certification measures (god forbid). Well, in my mind, the answer is already out there. In the last week, Valerie Strauss, in her Washington Post blog, featured “The Best Kind of Teacher Evaluation” (8.17.10) by Larry Ferlazzo, a popular instructional technology blogger, who wrote about a teacher assessment system used in his school. The system laid out in the article seemed very fair and balanced, using multiple measures over multiple periods of collection. The beauty of the system is that it gets away from high stakes summative assessment measures. Have we forgotten what formative assessment is as well as its value?

My point is that, unlike Arne Duncan and Michelle Rhee who advocate firing teachers at the drop of a test score, districts should instead be using formative assessment measures to identify teachers’ weaknesses, provide the training to correct those issues, then do this over a specified period of time. Then, and only then, should they recommend that a teacher who performs poorly look for a new job. Oh, and when districts do that, they should also help the teacher. Most companies do when they lay people off.

So this is what Larry proposes: A positive approach where faculty are observed by knowledgeable supervisors who know them, the school, and the students. This helps account for some of the uncontrollable variables. The school emphasizes that improving the teachers improves student performance. It’s not rocket science. This school is using different forms of data to inform and not drive teacher performance.

This system relies on other positive formative measures as well, including teacher involvement in determining their assessments, results from student evaluations, and teacher self-reflection. Ever heard of the reflective teacher? It’s a standard concept in much of teacher education.

So I will end with this: Why are we listening to Arne and Michelle and not Larry?

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s