<html><body><P>American Enterprise Institute</P> <P>November 19, 2008</P> <P>[Edited transcript from audio tapes]</P> <P><BR> <TABLE cellSpacing=1 cellPadding=1 width="100%" border=0> <TBODY> <TR> <TD> <DIV class=BodyText>3:45 p.m.</DIV></TD> <TD> <DIV class=BodyText>Registration</DIV></TD> <TD> <DIV class=BodyText> </DIV></TD></TR> <TR> <TD> <DIV class=BodyText> </DIV></TD> <TD> <DIV class=BodyText> </DIV></TD> <TD> <DIV class=BodyText> </DIV></TD></TR> <TR> <TD> <DIV class=BodyText>4:00    </DIV></TD> <TD> <DIV class=BodyText><EM>Introduction:</EM> </DIV></TD> <TD> <DIV class=BodyText><A class=eResources href="http://www.aei.org/scholars/scholarID.30/scholar.asp">Frederick M. Hess</A>, AEI</DIV></TD></TR> <TR> <TD> <DIV class=BodyText> </DIV></TD> <TD> <DIV class=BodyText> </DIV></TD> <TD> <DIV class=BodyText> </DIV></TD></TR> <TR> <TD> <DIV class=BodyText> </DIV></TD> <TD> <DIV class=BodyText><EM>Speaker:</EM> </DIV></TD> <TD> <DIV class=BodyText>Anthony S. Bryk, Carnegie Foundation for the Advancement of Teaching</DIV></TD></TR> <TR> <TD> <DIV class=BodyText> </DIV></TD> <TD> <DIV class=BodyText> </DIV></TD> <TD> <DIV class=BodyText> </DIV></TD></TR> <TR> <TD> <DIV class=BodyText>5:30  </DIV></TD> <TD> <DIV class=BodyText>Adjournment and Wine and Cheese Reception</DIV></TD></TR></TBODY></TABLE></P> <P> </P> <P>Proceedings:</P> <P>Anthony S. Bryk: Thank you Rick. It s a pleasure to be back here again. Well, truth in advertising, I m all of ten weeks into this new role at the Carnegie Foundation. And what I ll be sharing with you are our ideas that started to percolate up inside the foundation as we ve started to consult out more broadly, ideas I ll be talking about with the board on Friday, which is my first official board meeting. So we re very early on in this work.</P> <P>In the spirit of Chicago, I call my remarks Make No Small Plans: Building an R&D Infrastructure for the Improvement of Teaching and Learning. And for those of you who may not be familiar with this Make No Small Plans, it s a line attributed to Daniel Burnham, who is this architect extraordinaire responsible for the Columbian Exposition work and very much became the spirit of Chicago. Make no small plans. And Chicago is a great place to be these days -- or be from.</P> <P>We live in extraordinary times. Let s start with some of the observations that were central to the previous AEI paper, which has the word ruminations in it. So I call it the ruminations paper. This combination of economic, social, and technological change, in which we re now a part is challenging the fundamental foundations of what historically we ve called the one best system of public education. We re in this period of renewed mass immigration. We have this new knowledge-driven economy and we have technology as a potentially disruptive innovation that s changing the way we live, work, and will surely change the way we learn and educate.</P> <P>We have this new goal of ambitious academic attainment for all students. And in this context, what we re really looking at is not a call for new programs but fundamental change in the way our institutions that educate do their work.</P> <P>With that kind of big perspective in mind, it leads me to conclude that it s inconceivable that we could have these much better schools we aspire to without a serious renewal of the ways in which we support the development of people, the tools, materials, ideas, and evidence with which they work, and the institutions in which all of this occurs. There really is a pressing need now for more vital research and development infrastructure that focuses on performance improvement. Think about this as an individual human capital initiative. Think about this as an organizational issue. Think about it as an institutional issue, with a strong focus on performance improvement.</P> <P>Quick capsule of the current educational R&D scene: we need it; however, the basic institutional arrangements of public education, the work of universities, the commercial sector, and the interconnections among these are quite weak for carrying out the work we need.</P> <P>On one level, it s a financing problem. We don t spend much on educational R&D and so increased infusion of funds is needed. But it s more than a funding problem. In fact, I would argue that if we simply put more money into the existing institutional arrangements we have, it s not clear how much improved productivity would actually come out. But why is that?</P> <P>This is a problem that exists at the intersection of three big institutions. First, most educational research is conducted in universities. And what we re doing in universities is placing a very high priority on individual contributions to new knowledge. We re concept creators. That s what we do in the university. And we ve gotten quite good at it. But the culture and incentives within universities are not really supportive for the kind of collective work that focuses on engineering systems, practical things that work at some sort of scale. That is, the institutions were not designed with that in mind. So from the institutional perspective of higher ed, we have what might be thought as a gap in these processes of developing usable knowledge to improve practice.</P> <P>The second institutional leg of this is districts networks of schools. Districts tend to operate in a short term, largely reactive mode to state and federal policies. And as policies press in a certain direction, there are these new ideas that represent the new silver bullets and we suddenly march off and say, We re doing X. But what districts tend not to do, because these are longer term commitments, is engage in the systematic longer term development work of developing new practices, codifying them, testing them, and preparing them for more general use. There are, of course, exceptions but they are exceptions. This kind of longer term research and development is generally not thought of as the core work of districts. So inside the district, we can think about this, in some sense, as a problem of long term performance improvement.</P> <P>The third leg of this problem is really in the commercial sector. Now at first, this may sound a bit odd because there is a vision afoot that knowledge is developed in universities and schools; then go take it and implement it. But in fact, a lot of new practices and ideas coming to schools don t come directly out of universities but they come through the things, the goods, products, and services that schools buy through the commercial sector. They represent the third institutional leg of this. And in principle, they play a very powerful role because schools and districts do buy lots of things.</P> <P>However, the marketing decisions that exist in the commercial sector are largely shaped by the political realities of getting things approved by districts or states, being responsive to new federal initiatives about what the federal government is willing to have, what the discretionary resources are used for -- and so on. They re also not in the business of long term research and development of new innovations because there is no guarantee that, even if you build it and it s of high quality, anybody will buy it. What schools and districts buy are so much driven by what states and the federal government are willing to put discretionary resources into. So what that does is create a very conservative marketplace, a conservative set of market forces on the commercial sector. We see this failure, basically, to bring new tools and services to bear. </P> <P>All of which suggests -- just a quick capsule of the old AEI paper and this new vision of research and its connection to practice improvement -- that this work needs to be organized around core problems of practice. Now this may seem very sensible but what else would you organize it around? At least, in the context of universities, for example, much of the research work isn t really organized around core problems of practice. It generally starts with certain theoretical ideas, certain tools we want to work on. It starts with things that we think are interesting rather than problems that necessarily merit our attention. Organize it around core problems of practice, think about this as a task of innovation development where there s a college of expertise that has to be assembled, including researchers and practitioners, that there needs to be an engineering orientation, that it has to be tried out in schools or wherever the learning contexts are. Chances are the early initiatives aren t going to work that well so you re going to retry it. You re going to evaluate it, assess it, redesign it, retry it, and keep getting at it until you get to the point where you ve actually developed something that has the potential of working on a broad scale across large numbers of different contexts in the hands of different sorts of people.</P> <P>We need this kind of engineering orientation. And in doing this, we really need to explore new ways of knowledge development; the clinical expertise of practice connects to the commercial sector because it s really an intersection, those three institutions, that innovations at scale could emerge.</P> <P>All right, that s pretty much where the AEI ruminations paper ended. Today, I want to move off the back of that and begin to develop this modest proposal, a new mission focus, Building a Design, Educational Engineering, and Development Infrastructure for School Improvement. In the early paper, it was called Design Engineering and Development. Someone pointed out to me that that s DED, which sounds like its Dead on Arrival. Now, at least, we re doing the DEED, rather than, hopefully, being dead on arrival.</P> <P>I made an argument that we need this new infrastructure. But what should it point toward? This is a very practical activity but what s the horizon toward which the Design, Educational Engineering, and Development Infrastructure should point? And two big shaping observations, the first of which, is a need to focus on education for a conceptual age, advancing ambitious learning goals for all students.</P> <P>Why is this a key anchor? I would argue it s clearly important that we develop this broad base of basic knowledge and skills in all of our children. But as we think about education for the 21st century, that s not likely to be good enough. Where efforts around basic skills and knowledge are now occurring at very rapid rates on a global scale, to simply educate as well as the Chinese or the Indians is not going to create educational opportunities for all of our children. So we have to do this but we have to do more. What s the more? </P> <P>I ve tried to outline here a few big ideas. One is this capacity to use basic knowledge and skills, to apply knowledge to complex problems, where we re looking at a student s ability to perform competently in tasks that they may not have exactly seen this kind of task before. There s much more demand on social communication skills, working with diverse others, how to use new technologies to inquire, learn, communicate, represent information to others, and so on. I ve added a fourth part to this, developing the habits of mind and heart required to function competently in this new knowledge-driven economy. </P> <P>This involves a variety of things, such as ease in confronting ambiguous tasks. In much of the work, you don t know exactly what the right answer is, and you don t exactly know how to get there. Let s have a disposition to tolerate ambiguity, since the work often involves trying things, failing, and trying again. It involves creating dispositions for persistence in the face of failure, tenacity in this pursuit of worthy goals, that these are habits of minds and hearts that schools also must form in the course of education.</P> <P>To put this into perspective, here s a picture. Actually, it comes from a book by Pat Graham, The History of Schooling in America in the 20th Century. I like it because it displays a classroom, the North Bennett Street Industrial School, 1917, in Boston, which, at that point, was thought to be one of the premiere public school systems in the United States. In addition to the subject matter of the school, if you just look at the picture, it s quite clear that there are certain psychological and social dispositions that are being shaped through this kind of experience. These schools were about preparing docile, loyal, patriotic workers for an industrial labor force. That s not our economic future, an industrial labor force. And individuals who sit patiently for long periods of time in desks who follow the orders of a teacher are typically not the psychological and social dispositions that we need for success in that 21st century labor force. </P> <P>That means that what we do in the very basic social organization of education has to change if we re going to create individuals who can perform competently in their work and in their civic lives in the century ahead. It needs a very different kind of psychological and social experience than what we ve been accustomed to.</P> <P>So proposition one, the work has to aim at much more ambitious goals than we have attempted in the past. Proposition two is, the overall systems of education must become more efficient. My argument into this is, K 12 schooling in the United States does not get its fair share of our nation s human resources and there s a lot of evidence to document that fact. That leads you naturally, if you follow down that path, to the conclusion that teaching needs to become a more attractive profession. If that s what you think your organizational problem is, that, in turn, leads you to think, we ve got to change the career and salary opportunities for educational professionals and we have to make the work conditions more attractive if we want people to think of teaching as being an attractive profession. We have to work on salaries and work conditions.</P> <P>However, we have a colossal public finance problem. All of these things would seem to require us to expend large, increased amounts of money on public education. And it just strikes me as highly unlikely that we will actually see this, at least, anytime soon.</P> <P>Regardless of how important we think improving public education is, as a society, look at what we are competing against: health care costs spiraling out of control, unfunded pension mandates, huge infrastructure needs, a war on terror. Just go down the list of things and -- I did this part of the talk before the current fiscal crisis -- it seems unlikely that we re going to see a major infusion of new funds.</P> <P>If we re going to actually substantially improve our educational system toward these more ambitious academic goals, we have to figure out how to make the whole enterprise more efficient, because only in this way are we going to be able to do something about the human resource needs and the work conditions of the people who work inside these systems. I ll come back to that a bit more. The idea is that it has to be more ambitious and the whole system, in some form or another, has to become more efficient. That s the horizon for DEED.</P> <P>This leads again to this two-headed problem that I ve just described. And oh, yes, we need to accomplish all of this where the basic work task is getting that much more complicated as we now confront large demographic changes that are sweeping across school systems and confronting this renewed immigration in U.S. society.</P> <P>All right, if that s where we have to head, what might be some big organizing ideas for Design, Educational Engineering, and Development work that might actually help get us there? What would be useful operating principles? What I want to do for the rest of this talk is really focus on the how of Design, Educational Engineering, and Development, rather than the what. And by the what, I mean what specific problems of practice are worth investing research and development activities in. That s a talk for another day. This is really more about the how. </P> <P>I would argue that there are four big themes to guide our pursuit of those two twin goals: how to advance toward more ambitious learning outcomes and how to make these systems more efficient in that pursuit.</P> <P>And those four themes are: one, we need to aim toward a science of performance improvement in teaching. Two, we need to focus on usable knowledge for improving instructional systems, its development, testing, and refinement. Three, we need to think about exploiting technology as a source of potentially disruptive innovations. And four, we need to focus, in a very conscious way, on accelerating process of innovation diffusion, looking particularly at the potential of open education resource initiatives. I want to go into each of these four now in a little bit in depth.</P> <P>First observation, again, going back to this idea about the human resources side. Schooling is human resource intensive enterprise. Schools are only as good as the quality of our people, in their capacity to work together. Thus, if we re really about trying to advance more ambitious aims and doing it more efficiently, we have to look at the systems that we use for developing our people. Most of what s done in schools of education is largely driven by professional judgment. And the evidence about efficacy of our human resource development initiatives remains quite weak.</P> <P>So the question becomes can we complement good professional judgment with something akin to building a science of performance improvement? And this phrase, the science of performance improvement I take from the work of Atul Gawande, who has done work on performance improvement in very large scale medical systems, hospitals, public health entities. If you haven t had a chance to read any of his essays, there s a collection of them in a book called Better, and I strongly recommend it. It certainly had a big influence on my own thinking. That s where this idea of the science of performance improvement comes from. And reading through his work, there are a set of underlying principles to this idea of a science of performance improvement. The first of which is, you take the day to day work of educators and put that at the center of the inquiry. What are the basic day to day problems of practice that people are confronting? Those become the central problems in defining the inquiry agenda. The problems of practice improvement are at the center.</P> <P>Two, you ve got to have a commitment to measuring the core activities and the outcomes. Yes, you ve got to measure outcomes. But you also have to think about all the processes that are involved in the production of the outcomes. And you have to gather evidence about them as well because in the spirit of improving performance, performance can break down at any one of a number of different steps. And simply knowing you haven t gotten to the end doesn t necessarily help you to understand which of the components of this may be causing some of the difficulties.</P> <P>The third principle in this science of performance improvement is that the key goal here is achieving greater reliability in whatever innovation you are trying to advance. In systems that are human resource-intensive especially, there is going to be a large amount of natural variability in performance. And we know this in the field of education, there are lots of innovations that have been tried for decades upon decades that typically come up with evidence that they work someplace. Almost everything can work someplace and virtually nothing works everywhere. Innovations have variable consequences. If we understand that that variability exists, then the goal becomes one of trying to create greater reliability and performance. That s the task of improvement at scale. Can we do this with increased reliability?</P> <P>Thinking about this set of ideas, particularly in the context of human resource development initiatives, what would this mean? It means focusing sustained attention to what I call the full causal cascade that leads from the initial design of a human resource development initiative all the way through the intended outcomes, improvements for students. We developed some professional education program. What is it we re trying to accomplish? What is it we want people, the adults, the teachers to learn and be able to do? And how do we know whether or not they can accomplish it on the basis of the professional education initiative we just designed? That s the kind of questions we ask about instructional design for students. But we rarely dig very deeply into that kind of specification when we switch the focus to the instructional design for adults. What do we want them to know and be able to do? And how do we know whether whatever we built is actually accomplishing it?</P> <P>The second part of the causal cascade, assuming that some knowledge and skill has been acquired, does it translate into observable changes in practice? We ve done some professional education. Is this knowledge usable in the context of practice? Can we track this into what happens day to day in classrooms? And then assuming that happens, the next big step is are these practices, in turn, related to measurable improvements, over time, in children s learning gains in those classrooms?</P> <P>So it s taking this enterprise and then breaking it up into its constituent parts and asking about what s the evidence that we have for all these parts? Because the science of performance improvement says, At any place in this, we may have some weaknesses and that s why we re not doing better in terms of what s happening for students, we simply have to understand this if we re going to improve it.</P> <P>It also means a different methodological stance. And here, I may be on the verge of being excommunicated from two different churches at the same time, which would be quite an accomplishment. This neither drives us toward more vigor in doing randomized control trials nor does it drive us more toward a deeper embrace of action research, which are, in some sense, two big competing paradigms about how you might improve practice.</P> <P>The focus, again, has to be on how innovations work in the hands of different individuals, working under different circumstances because back to this observation that there is this natural variability is the phenomenon. If you want to attack natural variability, you have to measure it, you have to know what causes it because that s what gives you the information for trying to introduce more reliability into these systems. </P> <P>So simply getting information, the average effect of an intervention, as it is enacted in some volunteer sample of sites -- it tells you about the average effect but it doesn t give you much information about the actual natural variability. What happens with the treatment intervention and innovation as it actually gets taken up by diverse groups of individuals working under diverse circumstances?</P> <P>In principle, that s the knowledge objective we need. As a statistician, I would say it s the multivariate distribution of the effect sizes dependent on how they co-vary with the characteristics of the individuals who are engaging that innovation and the context in which they re working. That s the information objective we need and that drives us toward more of what I would call systematic forms of naturalistic inquiry, strategies that have potential, over time, to accumulate evidence about particular innovations that could help to lead us toward creating more reliability in the ways in which practice can be enacted and therefore, improved. It really is a different methodological strategy than either the naturalistic inquiry, which tends to characterize action research but basically has no strategy for accumulating, or the more formalized field trials, which can give us very precise information about an average treatment effect in a limited number of conditions but doesn t give us much guidance as to what happens when this moves out into larger scale use and improving its reliability.</P> <P>The next big piece of this, what we re really arguing about, is building this kind of generative capacity for continuous improvement. It s about creating knowledge about practice and being very specific about what this is, testing it, refining it, how it gets represented in various tools, artifacts, social routines, and common rituals because instruction, schooling is about all of these things. It s people engaging in a practice, using various artifacts, textbooks, assessments, and tools in their classrooms, the routines for how you use them, and all of which exists in the cultural matrix that include certain common rituals. If you really want to get improvement in scale, you have to specify these things out.</P> <P>We have to go through this next step of making the knowledge useful. It s not just simply enough to say, This works, and use these tools. It gets to this last point here of what I call the learnability test. Can we develop systems where people can learn to use this knowledge and learn to use it with some reliability?</P> <P>Gawande has a wonderful example of this in his work when he writes about medical practice early on in the 20th century where there s a great deal of concern about infant and maternal mortality during childbirth. Lots of innovations are coming forward. One of the innovations that came forward during that period of time was the use of forceps. And there were various maneuvers about how to use these forceps to assist in childbirth. However, one of the problems with forceps was it was very hard to teach new doctors how to use them because forceps were a matter of feel. And it s very hard to teach feel. And the consequences of getting it wrong are catastrophic. Some things are just hard to teach. They fail at the learnability test. Therefore, they are not good innovations. In this Building a Design, Education Engineering, and Development Infrastructure, the fact that you can make it work in some places is a start but ultimately, it has to pass this learnability test as well.</P> <P>Another observation, schools are only as good as the capacity of people to work together. Improvement at scale, in addition to being a human resource issue, is a problem of social organizations. In addition to developing people, tools, and materials, we must also work to enhance the instructional systems in which educators work and learn. And I want to work on this idea of instructional systems because it s key to the notion of how social betterment at scale might occur.</P> <P>So what does it mean to develop and improve instructional systems? Well, here, the question is you have to really step back a bit and ask the question, What s the work? If you think about this, as I do as an organizational sociologist, if you re trying to figure out an organizational design for an entity that is trying to put out more of some kinds of work, you have to start with some assumptions about what do you think the work of teaching is. If you think the work of teaching is the routine enactment of a set of scripts that someone gives you, that it s basically routine work, it s going to lead you down one path to thinking about the design of organizations to support it.</P> <P>If you think about the work of teaching as a more complex practice, it s really intellectual work. You have to know something about the students in front of you, their particular knowledge and skills, their cultural background, you got to connect to them and you got to know where they are if you re going to take them forward. It s a body of knowledge you have to have here. You have to understand the goals and view for instruction. And you have to understand, have some larger kind of theory of development, how a curriculum, for example, is organized, so how this goal builds on the one before it and how it s going to take you to the next one. You have to have some understanding about the goals and view for instructions. And then you have a whole set of pedagogical strategies, tools, materials, routines around which you have to have automaticity. You have to know how you do these things quickly. And the art of the work is in the intersection of those three things; what you know about students, where you re trying to take them, and how you make in-the-moment decisions about how to draw out of that pedagogical set of tools, materials, strategies you have to try to maximally advance students toward their learning goals. If that s what you think teaching is and that s a premise. Not everyone agrees that that s what teaching is. But if you think that s what it is, then you have to start thinking about what would organizational systems look like that would support and advance that kind of work?</P> <P>And here, what I m wading into is one of these classic dualisms in education, between this notion about scripting instruction on one end, the way we re going to get it really high quality is we re just going to specify it all down to the micro-level detail. We re going to introduce routines to go out and observe and monitor and make sure people are following the script. And it doesn t matter exactly who the kids are or what they know because you are following the routines. That s one character of the one extreme.</P> <P>Then there is this other extreme, which is also I think, equally dysfunctional, which is that, Well, it s just all really complicated. Every child is unique. Every classroom is unique. This is really a kind of craft activity that each teacher has to invent on their own. And that s my caricature of each classroom, a la Leonardo.</P> <P>If that s true, then getting improvement at scale strikes me as an impossible goal. One has to believe there is a more productive middle ground between these two extremes. I tend to still think that, yes, teaching is complex. But I think you can conceive of teaching as an organized complexity. So you d know something about the students goals and view, how these things are laid, how these things connect to each other, and you re working with a set of materials, tools, and ideas of development, more of this idea of an instructional system. And it s in that intersection that teaching occurs. That s the idea of an organized complexity, drawing on the work of Herbert Simon.</P> <P>If that s the nature of the work, what gets enacted everyday is this organized complexity, the next question it leads you to is, how do individuals learn to do this work better? Do we think about this as an implementation of routines problem? No, it doesn t seem like quite the right idea. So this idea that they have to learn how to implement things doesn t seem like the right idea. This begins to connect more to the work on expertise. Because when you look inside this work, this is really about developing expertise, that is, how do you get better day in and day out at making those kinds of instructional decisions that I have been describing? That s a problem of expertise development.</P> <P>Thinking about it that way suggests, what do we know about how experts develop? Experts develop through having lots of opportunity to engage in guided practice with more expert others. You take the basic work, keep doing it over and over again, and you get an opportunity to engage with others who are better than you at that work in trying to improve the work. That links very well to this current reform idea of professional learning communities but I want to put a particular extension on the back. It s professional learning communities organized around specific instructional systems because much of the current rhetoric about professional learning communities is of the form, Well, we ll bring teachers together and we ll talk about their practice, and then comes what I call the ZWIT. There s this Zone of Wishful Thinking, and then something good will happen. In fact, those professional learning communities must really be organized around people engaging in the same work day in and day out. They are actually trying to accomplish similar things with similar kinds of students drawing on similar kinds of resources because that s the context in which expertise gets developed.</P> <P>That leads to this idea of specifying what it means to talk about an instructional system. Well, you have to have some way to describe what you need to know about your students in terms of their background knowledge, their skills, and interests. So part of the instructional system is a systematic way to do that. There are these common set of pedagogical resources and practices. So I m organizing my classroom and for example, when we re doing something called comprehensive literacy it means I might be doing small reading groups. I might be doing writer s workshop. But that s a set of pedagogical practices that exist in some instructional systems. </P> <P>The assumption is we re all doing those things but that s part of what defines the system. We have a shared language, a set of descriptors for those goals and view for students and how they align over time. So we have a common way of talking about what it means for students to develop. And we have some public evidence about what constitutes learning because without that public evidence, there is no evidence feedback as to basically build this cause-and-effect logic inside of these improvement networks.</P> <P>This may seem obvious. Yes, you have to have this kind of cause-and-effect logic but this cause-and-effect thinking about the organization of instruction in student learning is not a pervasive norm in much of classroom practice. The language is much more around, Well, that seemed like an interesting thing that the kids are really engaged in, so that makes it a good instructional activity. But it s not really typically organized around this cause-and-effect logic of I have some goals and views for students. I ve organized the instruction this way. This maximizes the chance that this will happen or someone else has figured out a better way to organize instruction than I did. We don t have that kind of conversation because this kind of cause-and-effect logic isn t central to the conversation. It s largely not central to the conversation because it s typically not anchored in this kind of system thinking that I described here.</P> <P>All right, another observation. Technology is changing how we work, live, and learn and that gets to the third guiding principle for this D-EE-D effort that we really need to look at technology as a core resource for a more efficient and ambitious system. In saying that, I want to emphasize Ken Overs has this point, It s not technology for its own sake. The history of technology development in education is often of the form that some technologist comes up with this whoopseedoo idea, I have invented something -- like the MP3 player -- and if you just use it in your classrooms, things will be really wonderful. So much of what we ve done in this regard is bring technology to the classroom or school door but all that does is set up the problem. It s the question of how do you use it to advance progress on what set of objectives? That s the work that Design, Educational Engineering, and Development needs. Where is this going to add value? What are the pressing problems of practice and what s the argument that bring technology to bear on this that might do something productive?</P> <P>I actually think it can happen. Let me just share a few observations as to why. When you look at most efforts to improve schooling, in some form or another, almost every effort in improving schooling tries to address one of three levers for improvement. One is, you re trying to enhance the quality, quantity of those who teach in the systems in which they work. So we re tying to get more teachers in. We re trying to get better teachers. We re trying to get better curriculum. Lots of reform efforts are in that, or in some form or another, trying to work that lever.</P> <P>There s another lever. We re trying to expand instructional time. It starts with the observation that U.S. schools, in general, tend to have less instructional time than many other countries. You see this lever being picked up a lot by charter schools, charter management organizations that run a longer school day, that run a longer school year. Big urban districts pick this up by adding things like extended day programs and summer school, expanded time. That s the second big lever.</P> <P>The third lever, less used, is to try to change the core work technology, how teachers come together with students in the context of engaging the subject matter. But there are some efforts like this emerging -- not so much with K 12 but especially in the post-secondary domain around online course taking. It s perhaps the best example of this, where the core technology of teaching and learning is now being fundamentally changed.</P> <P>Let s stick to that lever one, enhancing the quantity and quality of those who teach. How could technology help us? Big opportunities around using multimedia, online resources and other technical resources to go with this problem of how we prepare teachers and how we engage professional education. This is a very weak technology right now. We know that there s very wide variability, from recent studies, going from classroom to classroom in the same school in terms of the contributions of teachers to student learning gains. And yet, almost all of the initiatives we have, even the best of them only account for a very small amount of that variability. So you have to believe we could do a better job in that area in technology. There are some useful developments here that lead us to think technology could help.</P> <P>We can expand the base of those who educate through e-mentoring, peer and cross-age networking. And technologies can be very useful in this regard. We can address the expertise constraint, particularly as you try to bring more ambitious instruction into classrooms, exacerbate the problem of the limits of what teachers know and can do around this more ambitious instruction. Now, there s the possibility of bringing this in virtually, a lot of big ways to enhance the quantity and quality of those who teach through technology.</P> <P>The second big lever here is a big low-hanging fruit that the field that, we just simply haven t done enough to exploit. What can we do to facilitate learning in the other 85 percent of time when kids are out of school? Because schools only occupy about 15 percent of children s waking hours. Can we break out of this nine-to-three box that thinks that learning only occurs when you re in school, in that classroom? How can we extend the work out into the home, out into the community during summer so that students can work at their own pace? And a very good example of this is there was an initiative -- actually, that started here in D.C. -- into books which tried to create pen pals outside and all across the Washington D.C. area, where every student inside the Washington D.C. public schools would be connected through a virtual network to a pen pal and they would be writing about books that they would be reading that would be aligned to the current curriculum that students were in. </P> <P>So they were extending the connection of students to adults outside of school. And where this becomes really critical here is that in writing instruction, the binding constraint is an adult who gets to respond to what you write. If you have 25 students and one teacher, each student is only going to get a very small amount of feedback on what they write. If you can expand the number of adults who are connecting to students about writing, in principle, you ought to really be able to improve students work in this area.</P> <P>Another example of this is in the context of English language learner classrooms, where again, you might have 25 English language learners and one teacher. The teacher again becomes the binding constraint because language development depends on the opportunity to create language and have someone who is a primary language speaker respond to what you produce. If you are in a classroom with primarily English language learners, there are very few opportunities for kids to do this. Can extending the networks out, through technology, open up our possibilities in that regard?</P> <P>And then of course, the big one, is this effort of this extraordinary dynamism that s now occurring in intra-space activities where children are, in some sense, self-organizing work around interest and can we figure out ways to harness this kind of incredible outburst of activities that s occurring in children s digital lives and bringing that to focus on the intentional learning goals of schools?</P> <P>Third lever, and this is the biggest one of all, the most ambitious one of all; can we reinvent the basic organization of schools in a 24/7 digital environment? Suppose, which we will have in the not too distant future, when kids live in this 24/7 digital environment, it s everywhere in schools. It s in the community. It s at home. There s this cloud, a wireless cloud that exists over the whole enterprise, where the digital technology is as commonplace as the paper and pen, as many of you who are writing with right now as I look down. Could this allow us, at last, to break out of the structure of schools? Schools are basically -- we ve got a series of classrooms, basic social organization of the work is a teacher and 25 children. If you take that as given, that s the way schools have always been and the way they always have to be, that really constrains the space in which innovation can occur. With this new technology, this 24/7 digital environment, is it possible to break the circle?</P> <P>I think the key here, the opening wedge, is can you figure out ways in which students can work more independently as part of what we call school? We re already starting to see this with online courses, intra-space project activities, use of cognitive tutors. We re starting to see initiatives of this sort. And the key characteristic of the initiative is that there are more occasions in which students can engage in productive, independent work. Why does that matter? Well, if students are engaged in more productive independent work, it then allows us to really open up and re-envision the teacher s role and allows us to do this in ways that people have been talking about, how to make the role of teachers more professional. It creates affordances for this to happen.</P> <P>Why is that? If we want our best teachers in these roles as senior teachers, where they re acting as coaches to junior colleagues and maybe perhaps groups of these teachers now are responsible for groups of students. To do that, to get a senior teacher working with junior teachers, you re changing her work and therefore, you have to change the way students connect to that teacher. You just simply can t pull a teacher out of a classroom 20 or 50 percent of the time without changing the basic social organization of instruction. So the technology now opens that up. It also redesigns the kind of work the senior teachers do. If students are able to work much more independently for parts of the day, then our best teachers can focus on students who most need the kind of work that only expert teachers can advance. So again, right now, it s a binding constraint because I have to work with all 25 kids.</P> <P>And the third part is -- if we want to be able to, in this Design, Educational Engineering, and Development work I described how this needs to be a co-development activity -- you need to be able to figure out ways to get some of your best teachers working with you in these teams. This can t be just the things they do on the side, on the weekends, or in the summer. They have to be a regular part of the team members. The only way you can accomplish that is that somehow, you change the basic way in which you think work is getting done, needs to get done in the school.</P> <P>And embedded in here, because all the way back to that human resources development problem I said before, if we want to be able to pay our best teachers more and we want to be able to change the work that we ask them to do, as in putting a lot more money into the system using the same core technology, there are only two ways to do this: put a lot more money into the system and keep the technology fixed, or if you have to work at the same level of resources, change the way we do the basic technology of schooling. And hard technology opens up the possibility that this second option could work.</P> <P>A couple of these other things. I m running over on time so I m going to go quickly. The technology gives us a potential of solving this century-long problem of what I call the progressivist instructional dilemma which one of my mentors, Pat Graham, called the baked apple versus the pork chop problem. And her analogy goes something as follows, the baked apple, well, you can undercook it, you can overcook it, put a little too much spice in it, and too little spices, it s edible. It doesn t matter what you do to a baked apple. You can eat it. Pork chop, if you season it just right, you cook it just right, it s really succulent. It s a great meal. But if you undercook it, it could kill you.</P> <P>And that s always been the progressivist s dilemma. This idea of intra-space worlds, building on what students bring into the classroom, individualizing and differentiating instruction, these are very good ideas worth embracing; it s just very hard to manage. And the management of that instruction has always been its weak underbelly. Some people figure out how to do it but it simply isn t something that scales and in the century of efforts to trying to do it -- we know what happens when we go down that path. And we may now have a set of tools, again, that actually make the management of these forms of instruction more likely to happen with some reliability at scale.</P> <P>Technology also opens up the possibility of finally solving this performance assessment problem in education as well. We tended to think about assessments as something you add on to instruction. So we do instruction and then we assess. There have been efforts over the last 15 years to develop more performance-based assessment but they re big, they re cumbersome, they re expensive, and they take big chunks of time. And money, like time -- time, like money, are both highly constrained resources in schools.</P> <P>If much of what students and adults do in schools is now encoded in the digital environment, it opens up the possibility of basically flipping the assessment problem from one of something you add on to instruction, to making assessment a problem of how you harness evidence about learning out of the digital record that gets created as a byproduct of instruction. It really opens up the possibility, I think, of solving the performance assessment problem. Having much more ambitious performance assessments, among other things, is really key to getting this assessment-driven accountability pointed in the right direction.</P> <P>And then of course, the last one of all of this, is if we have this much enriched digital record, there s a new information infrastructure that s created as a byproduct of work. And this is really the flywheel, this information that s getting created as a byproduct of the work for this ongoing Design, Educational Engineering, and Development work. Remember, one of the principles I started with was not only do you have to measure the outcomes, but you have to think about all the separate sub-processes, the culminating outcomes, and collect evidence about each of them. At the point I introduced this, you re probably thinking, God, this is a huge data-collection problem, which, if we use the conventional technology, it would be probably insurmountable. </P> <P>But if much of the evidence is now being digitally encoded as a byproduct of the work, it really opens up the possibility that we could dramatically expedite the way in which we learn about improvement efforts. Surely, there is much more because embedded in this is a generative capacity where, as we engage technology -- we know this from other fields and it s actually used -- you should begin to think about new ways to use it that you never envisioned except though the prior uses.</P> <P>Last observation, and I m getting close to closing. Lighthouses don t shed their light very far. We ve had several decades now of efforts to develop model programs and even things that work, those are the lighthouses that don t tend to shed their light very far. They don t tend to get picked up and used very well. And we continue to decry this problem but this is a central problem of innovation diffusion. At base here, we re confronting very deep-seated norms in education. This is not just K 12 education. It starts in primary schools and it goes all the way up into higher education. </P> <P>Our work is fundamentally driven by two norms, about the autonomy of practice in classrooms and this egalitarian ideal that tends to exist among faculty. Autonomy in classrooms means each of us have to invent ourselves. And egalitarian norm means, well, we can have conversations about our practice but it tends to be characterized by this euphemism we share I do X. You do Y. She does C. Isn t that wonderful? And you may go away from this and take some ideas but never do you really enjoin the real cause-and-effect conversation, Well, maybe what she s doing in B is actually better than what I m doing in my classroom, that real kind of critical conversation that s necessary to push practice forward and accumulate knowledge over time. These norms are very deep-seated in the practice and any effort to really develop innovation and have them diffuse at some sort of scale has to have a theory of action. How are we going to go at this cultural change?</P> <P>This is another place where I have begun to think about where technology may be a resource to us. It s around this open educational resources movement -- I ve begun to think about open educational resources as a kind of social organizational intervention. In this open education resources movement, a lot of it is driven by, well, we re trying to get more educational products out, free or at low cost, to teachers, kids, and districts. We are facilitating that sharing activity that I have just described before.</P> <P>But when you look at, Oh, there are initiatives that have been effective, and a good example of this would be something like Wikipedia, when you look inside them in terms of their social organization, there are very specific roles that people take on; there are processes that they engage in. Their mechanisms for adjudicating claims about expertise and efficacy, what s going to go on the corpus, what s not going to go in the corpus, they have a very distinct social organization to them.</P> <P>If OER initiatives were to emerge in education, they would surely span in classrooms, schools, and districts that have individuals now who would basically require individuals to work within larger systems that are generating ideas, testing ideas, and vetting their status within this kind of social community. It would also create visible records of contribution to the improvement of teaching. The reason why the visible records thing is important is that if we can t count it, in bureaucratic systems, if you can t count it, it s very hard to build systems that acknowledge it and to incent more of it.</P> <P>But if you can, if you have records of who has contributed to practice improvement, you can build this into the systems by which we acknowledge and promote people and even reward people. It s one of the things that s great and quite interesting about my tenure at the Graduate School of Business at Stanford is that one way you can get tenure in the Graduate School of Business is by being recognized for your contributions to teaching. And well, how do they do that? The way in which the teaching gets acknowledged in the business school is, they write cases and cases get acknowledged. This is the case to use for this kind of problem. And it gets used over lots of institutions. People buy it. There s a record. So-and-so s case or set of cases are really the definitive case or cases to use in public finance. They give people tenure for that. Why? Because it s got a visible record and it creates something that people can count. If we want a much more vital enterprise working on teaching and improvement, we have to solve this visibility and public record problem.</P> <P>To sum up, I make no small plans and propositions. We need to vitalize this educational engineering and development infrastructure for the improvement of schooling. I would argue you need to anchor this work around trying to do this on some specific problems of practice as they are understood on the job floor of classrooms, schools, colleges, and universities. We have to work at building some things at scale on some of these important problems of practice improvement, which I have labeled POPIs. We have to learn how to do this. We have to create exemplars of this design engineering and developmental work. And we have to study as we do it because we have to figure it out; we have to develop this kind of wisdom of practice to ground these field-building conversations. In that regard, we have to recognize that in doing the work, we re actually engaging in its own design engineering and development process because we are building an infrastructure that doesn t currently exist. That also means, among other things, we can anticipate some of our first efforts will fail, and the fact that they fail isn t the problem. The problem is we have to learn from that for the second round so that hopefully, we get a bit smarter at it.</P> <P>All of this is really organized around this idea about developing people, the tools, materials, and ideas with which they work and the systems within which this work is supported, which leads to one small remaining question, should we talk for another day about the role of universities and their relationships to school districts and the commercial sector. Because this piece here about developing people, the tools, materials, and ideas with which they work, that s the definition of what a professional school of education should be about and how it should connect the sites of practice. But for the most part, that work is not currently occurring. And that s a conversation for another day. Thank you.</P> <P>[End of transcript]</P></body></html>