Can Data Threaten Good Teaching?

During some TLN discussion about scripted curriculum and the growing emphasis on teaching 'prescriptions' based on data analysis, science teacher-coach Anthony Cody wrote:

(While) there is not necessarily a conflict between more thorough data systems and respect for the professional judgment of teachers…there is a problem with data, which I am aware of as someone with some background in science.

In order for data to be meaningful, as many variables as possible need to be controlled. If you are to compare students in different classrooms, the instruction needs to be as similar as possible. So as we go forward, if we view the solution to our educational troubles as being the generation of meaningful data, then we will tend to seek to standardize curriculum, instruction and assessment as much as possible, so the data will have the greatest possible validity.

But there are some huge problems with this approach. One large problem is that we are attempting to standardize all these factors in order to generate reliable data. But it is not clear that the new data is substantially better than the data we have had in the past, or that we have interventions that are appropriate responses to the data.

But the bigger question I would raise is the hidden cost of all this standardization. From the edu-technocrats’ point of view, it is no problem to expect the teachers to follow this or that timeline, or use this or that scripted curriculum. But as accomplished teachers, we know there is a huge cost paid by the teacher and students. The teacher has lost discretion to craft a sequence of lessons that reflects her best knowledge and her understanding of her students -- and the students have lost access to her special knowledge, and her ability to respond to their interests, strengths and weaknesses.

I think the standardized teaching model is having trouble. Its flagship vehicle, No Child Left Behind, is in grave jeopardy, and the serious problems of equity in our schools have not been substantially shifted, in spite of the "bright light" high-stakes test data have shone on them. As the model begins to sputter, its proponents look to rescue it by intensifying the standardization, making the systems even more thorough and reliable, unwilling to recognize the flaw in the premise.

The flaw in the premise, from my point of view, is that we are dealing with an immensely complex human enterprise. In a single classroom with 20 learners, we have a wide range of abilities and aptitudes, interests and backgrounds. No matter how good the scripted curriculum, no matter how sophisticated the tests, the decisions made by a strong teacher based on skillful classroom assessment practices will, on the whole, be better than those that would be made as a result of the standardized system.

In order to take advantage of this expertise, of course, you must empower and trust that teacher. You must invest in that teacher (and the development of more teachers like her), and create the conditions that sustain and support her. You can ask her to be accountable, but only if you give her the support and autonomy she needs to be effective and in charge of the learning for which you are making her responsible.

So every time we hear of these data-driven solutions, we need to be prepared look closely to see what is being sacrificed in order to create "reliable data," and we need to be prepared to argue forcefully for the importance of the teacher's judgment in assessing students, in designing and delivering instruction to meet the needs of our students, and in guiding our schools to meet the needs of all of our students.

Comments

[Bob]

I share many of your thoughts about relationships between databased instruction and good teaching. Yet, I cast it differently.

Presumably, all teachers use scripts known by many names, such as lesson plans, models, curricula guides, textbooks, assignments, etc.

Databased scripts show how instructional procedures have yielded anticipated results for samples of teachers. Some scripts give priority to transmitting specific, measurable units of information in predefined steps and time frames, because other teachers have used ways to bring students along irrespective of students' personal situations.

Thus, with databased procedures, teachers know and can adjust instruction promptly when students fail to meet learning criteria for each step.

Tedious, yes. Misused, probably. Misstated, sometimes. Effective, depends on the teacher.

I'd suggest that increased student learning rates, with or without scripts, define the art of good teaching.

[Anthony Cody]

Bob,
I am afraid I disagree on a basic level that a lesson plan created by a classroom teacher is equivalent to a script, and therefore we might as well use "databased" scripts. My point was exactly the opposite. A lesson plan authored by the teacher, even if it draws on a variety of published curricula or borrowed ideas, is the unique response of that teacher to the conditions in her class. Honoring the teacher's capacity to make this judgment in effect challenges teachers to know their students and learn what will work with them. A strong teacher develops a reportoire of instructional responses to students, and is able to shift gears quickly to take advantage of learning opportunities, and overcome obstacles to learning. This capacity is not demanded by scripted curriculum, and therefore it is not *built* by scripted curriculum. Instruction has been rendered less intelligent as a result.

[Rick]

When top-down fails, try bottom-up. It seems to me that technology advances of the last several years mean we can start with an individual teacher's ideas, encourage them to publish those ideas on the web/intranet, then create opportunities for networking and looking for commonalities with other local teachers. If you require the publishing, and make a range of permissible formats to choose from, you support self-determination while still getting a measure of accountability/standardization. Could be a very time-consuming and labour-intensive way to go about things, though even very small diversions of money from defence ought to make it make it easy to do a lot more than we do now.

[Anthony Cody]

No matter what we do, there is a great deal of work involved in getting many more students to succeed. I think what has been done in Nebraska with the School-based Teacher-led Assessment Reporting System provides us with some idea of what this might look like. they allow the local districts to set the standards, and the teachers are involved in designing how student learning is measured. As a result, teachers have become much more expert at all different sorts of assessment, which makes them far more powerful and effective in the classroom.

[John in NC]

Some of Rick's ideas are explored in this article from last fall's Journal of Staff Development:

http://snipurl.com/wurtzel

A quote:

"Transforming high school teaching and learning requires a new vision of teacher professionalism based on core commitments to improving individual and collective practice and student outcomes. Improving practice can only be done by teachers, not to teachers.

"But when teacher professionalism is defined as autonomy — freedom to make decisions about what, how, and sometimes even whom to teach — that autonomy does not support instructional improvement. Robust teacher professionalism offers a new teacher job description that places accountability for results and the use and refinement of effective practices at the core of teaching."

Controversial ideas, perhaps, but worthy of debate.