“The Flight of a Butterfly” or “The Flight of a Bullet”: The Impossible Dream of Transforming Teaching into a Science

According to many policymakers and researchers, teaching should be more like the “flight of a bullet” rather than the “flight of a butterfly.”*  Using the latest social science findings, they are determined to re-engineer teaching to make it more efficient, less wasteful, and far more effective than ever before.  Behind the current passion among policymakers and politicians for using test scores to evaluate teacher performance (and pay higher salaries) is the current “science” of value-added measures (VAM) that leans heavily upon the work of William Sanders. But these smart officials have ignored the long march that researchers have slogged through in the past century.

Before William Sanders, there was Franklin Bobbitt in the 1920s, Ralph Tyler and Benjamin Bloom in the 1950s, Nathaniel Gage in the 1970s and 1980s, and many other researchers.  These scholars believed that teaching can be rational and predictable through scientifically engineering classrooms; they rejected the notion that teaching can be unpredictable and uncertain–“the flight of a butterfly.”

In How To Make a Curriculum (1924),Franklin Bobbitt listed 160 “educational objectives”  that teachers should pursue in teaching children such as “the ability to use language …required for proper and effective participation in community life.” Colleagues in math listed 300 for teachers in grades 1-6 and nearly 900 for social studies. This scientific movement to graft “educational objectives” onto daily classroom lessons collapsed of its own weight by the 1940s, and largely ignored by teachers. Elliot Eisner told that story well.

By the early 1960s, another generation of social scientists had advanced the idea that teachers should use “behavioral objectives” to guide lessons. Ralph Tyler, Benjamin Bloom and others  created taxonomies that provided teachers with “prescriptions for the formulation of educational objectives.” As Eisner pointed out, teachers generally ignored these scientific prescriptions in their daily lessons.

In the 1970s and 1980s, Nathaniel Gage and others sought to establish “a scientific basis for the art of teaching.” They focused on teacher behaviors (questions asked, how students are called upon,etc.)–the process of teaching–and the products of effective teaching, student scores on standardized tests. This line of research called “process-product” continued the behavioral tradition from an earlier generation committed to a science of teaching. Using experimental methods to identify teaching behaviors that were correlated to student gains in test scores on standardized tests, Gage and others came up with “teacher should” statements that were associated with improved student achievement.

The limitations of establishing a set of scientifically prescribed teaching behaviors soon became apparent as critics pointed out how many other factors (e.g., the content of the lesson, students themselves, the classroom environment, the school) come into play when teachers teach students. Again, teachers generally ignored the results from “process-product” studies.

And here in 2010, the re-engineering of teaching through science again seeks “the flight of the bullet.” Who among researchers and policymakers ever mentions the artistry of teaching? Evaluating and paying teachers on the basis of student test scores through value-added measures dominates reform talk.

Researchers and policy advocates now prescribe teaching behaviors that will yield gains in student achievement. Doug Lemov’s Teach Like a Champion lays out 49 “should” behaviors extracted from research in high teacher-turnover, largely minority and poor urban schools. The aura of “science” hovers over these  prescriptions. Ditto for VAM.

Even though many researchers challenge the use of VAM metrics and methodologies to evaluate teacher effectiveness,Washington, D.C., New York City, and Los Angeles–note how most top-performing suburban districts in the nation have taken an oath of silence on VAM–have embraced it. The “science” of measuring teacher effectiveness, in short, is sharply contested–and, yeah, that bullet zigs and zags.

The point of all this is to be clear that, yes, some parts of teaching can be improved through scientific studies. Empirical findings time and again have improved teaching from decoding skills in reading to classroom management. But what has been learned from science is not the lion’s share of what constitutes daily teaching. As Philip Jackson said in 1968:

“teaching is an opportunistic process [where] … neither the teacher or [her] students can predict with any certainty exactly what will happen next. Plans are forever going awry and unexpected opportunities …are constantly emerging. The seasoned teacher seizes upon these opportunities and uses them to … his student’s  advantage.”

Surprise and uncertainty greet teachers daily even for their best-planned lesson. Experienced teachers know this in their bones and in finessing the unpredictability of classroom life (or flopping) know that few researchers, especially among VAM-obsessed ones–care for such artistry because it cannot be connected to students’ test scores.

Those who still dream of engineering  classrooms into rational places where empirically-derived prescriptions help teachers become effective have failed to grasp that daily teaching is a mix of artistry, science and uncertainty.

*Philip Jackson, Life in Classrooms, 1968, pp. 166-167.


Filed under dilemmas of teaching, how teachers teach

8 responses to ““The Flight of a Butterfly” or “The Flight of a Bullet”: The Impossible Dream of Transforming Teaching into a Science

  1. teachingbattleground

    I’m reading Lemov’s book at the moment and I quite like it. I can see how it fits into a “scientific”, technocratic approach to teaching, however, I think there is a distinction to be made between different technocratic approaches.

    For some researchers (John Hattie’s recent book “Visible Learning” springs to mind), teaching is like medicine. We can only evaluate teaching methods from precisely measured results. Until the students take a reliable written test we are likely to be blinded to the effectiveness of our methods, and even when analysing test results then we have to be careful not to be fooled by the placebo effect. Anything we observe directly in the classroom is likely to be misleading, in much the same way as (we now know) most of medicine was unreliable until people started improving methods for objective evaluations of treatments.

    Lemov’s book, while still technocratic, sees teaching as more akin to engineering than medicine. It may be focussed on technical methods which can be studied systematically, but we can see those methods working directly and applying them is often on the basis of improvisation and professional judgement rather than according to a formula. Teaching remains a craft not a science (although the “science” of testing may help us identify the master craftsmen).

    I can see why people might object even to this approach, perhaps teaching should be akin to a performance or a friendly conversation rather than a technical craft. However, I still think it can be a helpful way to view teaching, and shouldn’t be lumped in with the extreme positivist approaches which view teaching methods as precisely measurable commodities to be applied as treatments accoding to the latest data on effect sizes.

  2. texasadministrator

    Shouldn’t the title be … or “The Flight of the Magic Bullet”? In education, programs and new techniques are sold as if a new and better process as just been discovered that will make learning quicker and easier. Seems our elected officials and researchers are always in search of the “Magic Bullet” to student learning. Unfortunately, learning takes effort.

    Here’s a couple of suggestions: Teachers, prepare good lessons that are relevant to students, and students, work hard and study. Principals, provide a safe and orderly environment that’s conducive to learning and support your teachers.

    That might not be the “Magic Bullet” everyone’s looking for but it will hit the mark every time.

  3. Cal

    VAM and teaching “methods” are quite different. One involves inputs, one outputs. I agree with you that teaching itself is not a science, and am equally tired of the different approaches being sold as the new new thing.

    But if you consider teaching itself an art or a craft, and an unpredictable one at that, you can still measure the results. That is, whether a teacher follows Lemov religiously or breaks every rule in the non-existent book, he or she can be evaluated by the student’s progress.

    We can debate whether or not they should be, or how to do it accurately, but that’s a different issue.

  4. Diana Senechal

    A big problem with the focus on “results” is that teachers (and schools) are by no means in agreement over educational goals. One cannot fairly compare the results of two teachers legitimately trying to do very different things. When the test is essentially non-curricular, similar results can reflect widely different learning. This is often the case with ELA in particular.

    For grades K-8, and even beyond, in most states, there is no required literature on the ELA tests or in the ELA standards. One eighth-grade teacher might teach Aristotle, Sophocles, and Shakespeare; another might have students read chapter books of their choice and fill out Venn diagrams and prediction charts. Both teachers may be addressing the standards and preparing students for the test. The students in the two classes might score similarly on the ELA test yet have learned profoundly different subject matter.

    One might argue that they should be learning to read, in any case, so the results should still mean something. But this gets murky in value-added ratings. A teacher might be rated “average” just because her students didn’t make great gains beyond what was projected–but they might have already been scoring well in previous years, and they might have learned far more than the test even tested.

    It is fine to have general standardized tests–but it seems absurd to compare teacher results without considering what the teachers actually taught and why.

    • larrycuban

      Thanks, Diana, for the reminder about the nature of the standardized tests used for VAM in general and ELA in particular . An important point that a few researchers mention but advocates generally duck.

  5. I wonder if the thing we ought be studying is not teachers’ actions, but teachers’ thinking. That is, what reasons do the teachers give for certain actions. I bet if we identified common thinking (or perhaps “guiding principles”) among the best teachers, that would be much more useful than a list of “teacher shoulds”.

    Then, teacher prep programs ought work to instill students with the understanding and dispositions necessary to think in the complex ways the best teachers do.

    Just some “thinking out loud”. As always, great post.

  6. Monica

    “Those who still dream of engineering classrooms into rational places where empirically-derived prescriptions help teachers become effective have failed to grasp that daily teaching is a mix of artistry, science and uncertainty.”

    And do these same reformers raise their own children in this manner? Do they engineer their family life into empirically-derived prescriptions? I sincerely doubt it, since such an idea is preposterous when considering parents and their own children, yet it is seen as rational when considering teachers and 20 (or many more) students. As I say to my pre-service student teachers, “when you are in the classroom, anything can and will happen.”

    Wonderful post… thank you.

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 )

Google photo

You are commenting using your Google 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 )

Connecting to %s