Note: Jal Mehta, associate professor at the Harvard Graduate School of Education, is guest blogging this week.

One central challenge in American education is building more expertise in teaching. In recent years we have tended to think about this in terms of rating individual teachers, but what if we were to step back and see it more as a systemic issue, in terms of how well the field is organized to promote the development of expertise in teaching?

Presumably the development of expertise is a two part equation: the development of knowledge that would guide work in a field, and then helping practitioners gain that knowledge, either during their training or on the job. We are particularly bad at both parts of that equation.

First, knowledge. University research is fundamentally oriented towards scholarly publishing, meaning that the main audience for one’s work is other researchers as opposed to practitioners. Prestige in research fundamentally accrues to advancing theory, which means that the kind of granular, specific knowledge which could guide practice is actively dis-incentivized in scholarly publishing. Further, the form of research which currently prevails in the American academy is longer on objectivity than it is on relevance--researchers have mostly tried to distance themselves from the people and institutions they are studying in the name of preserving their scientific objectivity. It is also the case that the kind of propositional knowledge that researchers tend to produce is often difficult to translate to the intensely context-specific and interactive teaching situations that teachers face.

Meanwhile, there is (or could be) a lot of knowledge that comes directly out of practice. There is almost no teaching problem that hasn’t been solved by someone somewhere. But there are few mechanisms for this knowledge to be made visible, captured, shared, and vetted. The invisibility of this practical knowledge has been a huge lost opportunity for the field in terms of how to get better.

Second, training (pre-service and in-service). Criticisms of teacher preparation are legion and corroborated both by research and by the experiences of almost all practicing teachers. Training is too theoretical and not closely enough tied to practice, higher ed faculty don’t have enough experience in schools to deliver more relevant knowledge; almost everything that teachers learn is on the job. There is little connection between these pre-service training programs and what schools or districts do (either in terms of induction or just more generally in terms of professional development) meaning that there are generally not trajectories that build common and deeper understandings of a given subject. Once they become full-time teachers, teachers still teach mostly in closed classrooms, lacking the time and sometimes the inclination to collaborate and learn from one another. Districts send out initiatives at a frenetic rate, which leads experienced practitioners to try to keep their heads down, close the door, and ward off outside influences.

Education also does not really have any traditions that formally recognize different levels of expertise within a given school. Perhaps because of union influences, perhaps because teachers tend to be nice people, it is rare to find a school where it is the norm to acknowledge that different teachers are at different levels of expertise.

Compare that picture to a rough sketch of what we know from the expertise literature about the conditions which promote the growth of expertise:

• Clear and specific norms, standards, and a common language through which practice is discussed;
• Intensive feedback and coaching;
• Frequent opportunities to practice in ways that spiral to increasing levels of complexity;
• Participation in communities of like-minded people who help one develop one’s skills and knowledge.

In both fields which emphasize acquiring a knowledge base while in school (like medicine) and fields which feature a more apprenticeship style structure (like carpentry) you can see those components; American education is weak on both the development and transmission of formal knowledge as well as the apprenticeship approach that would enable the growth of practice.

We can see what these things look like in pockets. In particular, teacher preparation that is organized by schools (like Relay in New York and High Tech High in San Diego) has the advantage of being much more clearly oriented towards a particular vision of pedagogical practice. There are also strong professional development providers and independent consultants like Jon Saphier, whose work on The Skillful Teacher has sought to bridge this intermediary space and help teachers more directly. Websites like Better Lesson enable teacher sharing of lesson plans. And, of course, Doug Lemov’s Teach Like a Champion has sold like wildfire, showing that there is considerable demand for the codification of this kind of knowledge.

The question going forward is how we might develop more structures that would support these efforts. Existing institutions are not well-suited to doing this work, and hence I think we need new and different kinds of forms that are more deliberately aimed at building expertise in teaching. Here are five suggestions, three of which feature the work of my students, and two which are my own:

• Help school-based folks to start their own graduate schools of education. When Match was going through this in Massachusetts, they found out that the requirements to establish a graduate school of education included things like having a library and a faculty full of Ph.D.s. This requirement may have made sense at the time, but no longer does. Much as there are applications to become charter schools, there should be a way to apply to run a graduate school of education, and these applications should be judged on their ability to effectively train teachers and produce practical knowledge, not on inputs like whether they have a library or a faculty with Ph.D.s. One of my students, Peter Fishman, is developing a proposal to this end.
  
• Create vertically integrated training. Right now, teachers spend one year in teacher preparation institutions and then are handed off to districts for “induction.” It takes at least three years, on average, to become a modestly competent teacher, and the process would work better if it were overseen by single institutions from start to finish. In a proposal being developed by another pair of my students, Arthur Nevins and Diana Zarzuelo, states could incentivize districts and teacher preparation providers to collaborate and offer coherent proposals that would cover the three years of training. The best of these could be funded and spotlighted, Race to the Top-style, which would gradually increase the pressure for more similar entities.
  
• Good to great. A third of my students, Alex Seeskin, has pointed out that there is much more energy in the space around getting new teachers to master the basics of classroom management and other initial tricks of the trade, and much less that helps teachers who have mastered the basics push towards more complex teaching and deeper learning. Alex is working on an intervention that would help teachers move along this part of their trajectory. The part of the pipeline that covers years 3-7 and beyond is fairly open space, and is critical if we want to build more expert teachers.
  
• Create teacher career lattices. A big part of developing expertise more consistently would be more formally recognizing what teachers need at different stages in their trajectories. Beginning teachers need a lot of advice and direction, intermediates need opportunities to grow and take on new problems, and master teachers should be training new teachers and developing model lessons that could be used for the field. We could imagine a world where these different stages of teacher growth were explicitly recognized (much as you have medical students, residents, and attendings), but we have not done so in education. Programs like the Teacher Advancement Program show how schools can be organized on these principles, but they are currently the exception and not the rule.
  
• Create a conference parallel to AERA focused on improvement. Every time I go to AERA, I feel like the whole is much less than the sum of its parts. I can’t help but imagine that if we took these 13,000 people who are applying their collective intelligence to understanding schooling and more directly connected them to the folks who they were studying, much more could be achieved than through our current paradigm of practitioners acting and researchers documenting and evaluating. The most hopeful work I see out there are folks who are trying to shift to a different paradigm.--in particular, Anthony Bryk’s work around networked improvement communities as well as the work around design-based research. What these approaches share is a commitment to using rigorous methods of inquiry, working with practitioners to shape questions and answers, and feeding results into an iterative process of social learning and improvement. Since I’m not hopeful that AERA as a whole will move in this direction, I’d like to see the creation of a parallel conference, call it the Applied Educational Research for Improvement conference (AERI). At AERI, all of the presentations would be collaborative endeavors between researchers and practitioners and there would be a commitment to using what was learned in real time as part of improvement efforts.

These are just a few ideas, and they don’t address the full range of problems identified above. I’m working on a project on this topic - if anyone has other ideas or resources, feel free to be in touch.