What are some of the pressing concerns that we have for schooling in the United States right now? Well clearly, we are very interested in what makes for a high quality classroom. We are also interested in what makes for a high quality teacher. What are the kinds of ingredients in teaching that matter for kids? We are in fact legislating that every child has a high quality teacher in the United States. What does that mean? It is defined in the legislation, but what I’d like to talk about today is really does that means for kids and for teachers in schools. And we spend lots of times talking about the performance gaps that exist between kids and groups of kids as they enter and exit school and we talk a lot about the kinds of morale problems and longevity kinds of things, work force kind of issues for teachers. Many teachers that we talked to feel as though the pressures and the demands of the profession right now exceed their willingness and ability to sort of hang in there. And that should be of a concern to us as well. I come from a School of Education and we are under a lot of pressure to prove that we make a difference, okay? Does it matter that somebody goes through the Curry School of Education? Does that add value somehow to the work that they do when they go out there in the field? And then most importantly and what I am going to sort of harp for a little bit is can we approach all these questions scientifically?

So the context of the decisions that we are going to need to make about schooling in the next several years clearly drive around some federal legislation. No Child Left Behind mandates all these assessments of kids’ performance. We are all living with these things. My junior in high school comes home and he says if they test me one more time, I am going to quit school. Those are the realities that kids and teachers face. No Child Left Behind is mandating a high quality teacher in every single classroom. Keep this in mind as we get further into the presentation because the definition that No Child Left Behind is particularly for elementary schools, which is most of what we are going to talk about, is met in almost all elementary schools in the United States right now. This is because the threshold for the definition of a high quality teacher in that legislation is actually fairly low in the sense that that they have a four year degree and some training and experience in the field.

We are currently obsessed with finding solutions to the kinds of problems that we think the education system faces by looking at what we as researchers call structural kinds of features of schooling. So we try to say, does it make a difference what kind of tests we use? Does it make a difference what kind of curriculum we use? That’s the answer. Does it make a difference if a teacher has a bachelor’s degree or a master’s degree? Does it make a difference how we schedule kids? Block scheduling or not block scheduling? Should we be using vouchers to send the kids to private schools? These are big questions that are driving much of the discourse clearly at the federal level and an awful lot at the state level. Do these things matter would be really important to know and we are going to talk a little bit about that today and these proposed solutions are all again what we call structural changes. They are changes sort of outside the classroom. They don’t really deal at a very direct level with what an individual child experiences and what a teacher is able to offer that child in a given classroom in a given year. You know, in second grade lets say, in a particular locality.

So what we want to do is kind of approach these issues from more of a scientific perspective than from a testimonial or story-based perspective. So that’s the context? What’s some of the realities that we face? Well one of the big realities and there is no shortage of discussion on this and I think probably a little bit more heat than light. The gaps in kids’ performance in any number of ways you want to cut the groups of kids that you are looking at, from the high performers to the low performers, at the end of school, range somewhere between a half and one standard deviation. If you had a test, which has the average score of one hundred and most of the kids fall within fifteen points of that, that means that the gaps in between high performers and low performers are the type that we tend to be concerned of: racial gaps, income gaps, education, family background gaps are between seven and fifteen points on that scale. Most of that gap is there at the start of school. Almost all of that gap is there at the start of school. In fact, roughly eighty percent of that gap is there at the start of school in most of the literature that you look at. In fact, the thing that we know a little bit, particularly in relation to kids in poverty, is that schooling mitigates that gap. And there are periods around which there are actually critical action. The preschool to third grade period is a period in which if kids that are coming from tough backgrounds, low income backgrounds, are in high quality programs for a period of time during that period, you can close the gap somewhat and it stays closed for a little bit. The lines are still parallel after that point, but you can close it a little bit. And around middle school and high school becomes a bigger point for lessening the gap or in fact, not letting the gap get wider, which is what tends to happen when kids from low income versus high income backgrounds are going into middle school and high school. And there are lots and lots of examples of gap closing schools and a lot of people have described those things in a lot of different studies. We are going to come back to this a little bit later.

For most of the time, I am going to talk about where I do my work, which is in the period from preschool to about grade three. When we do national level surveys of kindergarten teachers, they are telling us that almost twenty percent of the kids who walk into their classroom door on their first day of school have adjustment problems that they find difficult or serious that don’t go away after a couple weeks in school. That’s twenty percent of the population having that level of concern. And you know, I trust these teachers. We talk a lot to teachers in the work that we do and I don’t think they are over estimating that one bit. I think there’s a considerable level of concern about kids, mostly around things like following directions, getting along with other kids, working independently, and some issues around academic skills.
Some people have talked about these early school years that go from preschool until about third grade as what’s called a critical period. That is, if you don’t get something done in that period in development, if you don’t set get kids on a particular path during that period of development, on a path of success, then they don’t ever get there. They don’t ever catch up. That strikes me as a little bit of an extreme kind of statement, but there’s clearly plenty of evidence that this period plays a unique and disproportionately important role in kids’ school careers.

Okay, we are going to focus a lot on classroom experiences in most of the talk today and what I’d like to do is to describe for you some steps that we have taken in what is a fairly large-scale effort to try to begin to improve experiences for as many kids as we can in the younger grades over the next couple years. We are going to talk about in reality when we go into classrooms rather than talking to teachers or principals through surveys, but when we actually go in to observe classrooms, what’s the actual nature and quality of experience that kids have preschool to fifth grade. If this is a really important period of time, we should know something about what actually goes on in classrooms so we’ll talk about that in terms of typical classrooms and in terms of variability across the population.

It’s also really important to know whether all these things that we are trying to debate about solutions - teachers’ levels of education, high quality teachers defined by No Child Left Behind, class size, private schooling – do these things really matter in terms of the kind of experiences that kids get in classrooms? If they don’t matter, then we shouldn’t be paying attention to them and we should start focusing elsewhere. And then we will try to find out in what ways do the experiences that children have in these classrooms matter for their development. What are the things that close the gap? What are the things that we can go into a classroom, look at, and say fairly confidently that the kind of experience that this child’s having in this classroom is either going to contribute to success or not going to contribute to their success. And finally, how and in what ways can we work to improve the richness and quality of kids’ experiences in classrooms?

Let’s start with our observations. We are going to talk for a moment about some national level studies that were involved in the observing the actual experiences that kids had in the early grades, from preschool all the way to fifth grade. I feel very strongly and I think this is borne out in an awful lot of studies that have been conducted that if you want to know what’s going on in school, you have to sit in a classroom for a little while. Talking to people is really not the best way to find out the kind of experiences that kids are getting that really can contribute to their growth. I understand why principals say this to parents when they go in, but on the other hand, I disagree with the fact when principals will say at times when you go to ask, how are all the teachers doing in the school here and they’ll routinely say that all the teachers are doing terrifically here. We have got a great school. I understand that and I know that’s why they say that to me as a dad, but on the other hand when principals tend to report that way on research surveys, that information is not information that we can have an awful lot of confidence in if what we are really interested in is trying to understand the gains that kids can make.

So we are going to talk about two different data sets here. One from the National Center for Early Development and Learning and another one from the NICHD Study of Early Child Care. We have observations from these studies. We have observations of about one thousand to twelve hundred classrooms at each grade level from pre-K all the way up to fifth grade. The total sample of observations that I am going to talk about in classrooms, in different classrooms numbers almost five thousand. This is the single largest compilation of information about what actually happens in classrooms in the United States. I like to think of this as sort of taking the country’s temperature in elementary school. It is going out there. It’s a snapshot. We’ll talk about being in classrooms basically for one whole day and what we see in them, but it is a snapshot of what school looks like across the country. It has certain values in some cases and it has limitations in others.

This is just a map that shows the dispersion of the observations in kindergarten, first, third, and fifth grades. Thirteen hundred kids that we followed since birth in that study. And then in the pre-k and k study sites, these are the states in orange and blue. Excuse me, I am too much of a Cavalier, orange and blue. In orange and yellow that are in the pre-k and k observational studies. In each of these studies, in fact, particularly in the pre-k and k studies, we are representing eighty percent of the kids who receive state-funded pre-kindergarten programs. These are mostly the pre-kindergarten programs like Bright Stars in Albemarle County, what used to be located in Jefferson Preschool in the annex that served predominantly low-income kids across the country.

Okay, so the big question is what to observe? You know, you have lots of things you can pay attention when you sit in a classroom for awhile. So what do we observe? Well first off, in all of this work, we rely on, these are large-scale observations, thousands of classrooms so we want to make sure when we go in there we’re paying attention to kinds of things that matter. So the kinds of observational tools that we’ve used in these studies, all of these codes have been selected because they have already been shown to predict to children’s skills. So we’ve combed through a number of different studies that have already been done of classrooms and we’ve tried to pick the kinds of things that add value for kids’ development over and above what families do for development. That add value to kids’ growth during the year that they’re in a particular classroom and that come from a highly rigorous scientific designs so we have pretty good confidence that these are indicators that matter for children’s development.

When we go in to do our work, we are in these classrooms on a day where we arrange the visit with a teacher. It is a typical day or morning where the teacher tells us that we are doing mostly academic work on this particular day, it’s fine for you to come on in. I want to call attention to the fact that all of these teachers are certified teachers so again this is a snapshot in some sense of whether schools of education and school systems are also doing a good job of supporting teachers in classrooms. So I want to make that clear.

We pay attention at the classroom level. What kind of climate does the classroom has? Does it have a positive climate where kids like to, seem to like to, spend time? And the instructional opportunities that are available for kids and we also look at individual kids to find what are you doing right now when I am scanning through on a regular basis in the classroom? Are you paying attention to the lesson? Are you being exposed to a particular kind of activity like literacy or math or science? So we pay attention on a time-sample basis to the setting and activities that children are in, the kinds of things that teachers are doing, and we also rate the quality of the environment on global ratings that run from one to seven. I’ll describe those in a little bit.

The ratings that we use focus in two different areas. These are going to be really important because this is the stuff, I’ll give you the end in the beginning, and this is the stuff that really matters when we go into predicting whether kids are growing academically and socially as a function of exposure to a classroom. So the emotional climate of the classroom we pay attention to. As I said before, is this a place that socially and emotionally kids seem to like to enjoy being together? Do they cooperate with one another or is it a place where there is sort of a harsh emotional tone? Is there evidence that teachers and adults in the classroom sensitively respond to kids’ needs? So they are monitoring what goes on in the classroom. They notice a child may be in some altercation or may not be paying attention and may need some support in the lesson and they find a way to sort of move in there and provide the support the kid needs to keep themselves going. That would be something like sensitivity. How much does this support children’s independence in the classroom? How much are they fostering kid’s confidence in being able to make decisions of their own within appropriate limits? So those are the kinds of things that cluster around emotional climate.

Then the other part of this is instructional climate. One very base dimension of instructional climate is just simply what we call productivity. That is, is the classroom a place where people are doing things or do they spend most of their time in transition? People don’t believe me a lot of times when I’ll talk about a story, but we literally have watched thousands of classrooms and I can tell you in one particular third grade situation that we watched, the bell rang for the beginning of the school day and it was thirty seven minutes before the children were asked to do anything or provided with an sort of structured activity. That got a low rating on productivity. Then you’ll see other classrooms where the teachers do a wonderful job of moving kids through transitions, not in a hurried pace, but keeping the kids busy and active.

The other things that are really important here are the dimensions of concept development where we pay attention to the degree to which the teacher provides the children with rich and stimulating activities that engage their interests and quality of feedback. Does the teacher provide kids with feedback about their performance and opportunities to learn from the teacher’s feedback rather than simply sort of say yes, no, hand the paper in and we’ll look at the paper later. In our procedures, we approach this like the good scientists that we are, I hope. So when we’re sending literally hundreds of people in these studies all across the country in to watch these classrooms, what I don’t want back is five thousand individual stories like we all might generate. I want everybody to be able to look at each of those five thousand classrooms using the same kind of lens so that we can begin to understand the variation across those classrooms and what we think matters for kids. So we spend lots and lots of time, I am not going into a lot of details, but we spend lots of time training people. We use lots of videotape examples. People are not allowed into classrooms to do these observations unless they pass fairly stringent tests.

One of the questions we get a fair amount is you can come into my classroom on one day, but the next day is really different. Well in fact, we have tested that on a number of occasions by having two different visitors go into a classroom. Because we are doing so many of these, we can’t spend lots and lots and lots of time in a given classroom. We have to kind of come in and really do the best job we can within a short time frame so we need to make sure that short time frame represents how that classroom functions over the long haul. So we have gone in on several different classrooms, several hundred classrooms of this data set and observed on multiple occasions and the information I am going to share with you is information that is very reliable across those different occasions. That means when I go into your classroom on Tuesday and I come back six weeks later, the relationship between emotional climate on Tuesday and six weeks later is pretty high so we are pretty confident in that.

Okay, so what do we know when we go in there? Now, I am going to summarize these things. You are going to get the broad-brush summary here, but these are all from published reports that have been reviewed in the literature and they are consistent across pre-k all the way up to fifth grade. So I am going to talk about when I talk in more general statements is consistent across that time period. When there is something specific to first or to third or something like that, I’ll draw attention to that. Well the vast majority of what happens for kids in a given classroom is what we call whole group or individual seatwork. Kids are sitting in a whole group situation or they’re sitting at their desks doing individual seatwork. Well that might not be that surprising to many of us. There are very few if any social or instructional interactions that occur between a teacher and an individual child. When we scale this to look on a per hour basis, there are about four occasions during an hour-long observation during the course of a day in which a teacher and an individual child will have an interaction with one another. That doesn’t change very much given the number of kids in a classroom or the types of kids in a classroom.

Most of what happens in early elementary school is literacy. I don’t think any of us are surprised about that. Clearly, that’s the pressure that’s on teachers and on parents and on school districts. Perhaps the most important thing though out of these data is that there really is just exceptional variation across classrooms. I am going to show you some examples of that. You can go into any given classroom, in any given school, on any given day, they are all running the same curriculum and what happens in Mrs. Smith’s first grade classroom will be strikingly different than what’s happening next door in Mrs. Jones’ classroom. Strikingly different from the child’s perspective. And we’ll talk about that in a little bit.

So how do kids spend their time? Well you get high levels of what we call business or routine activity. So in pre-k or k, you spend a lot of time in snack, lining up, and taking attendance, okay? Thirty percent of your time. Now I don’t know whether thirty percent is the right amount of time or the wrong amount of time, but what’s important about particularly these pre-k classrooms where this is occurring, many of these pre-k programs run a two and a half hour program. You know, a half-day, barely a half day. When you take thirty percent of two and a half hours and you take that out of there as learning time, you are diminishing considerably the investment that the public’s making in that particular program as an intervention for kids who may need more help than otherwise.

In first to fifth grade, kids spend lots of time, roughly again, thirty percent managing materials and routines. There’s an exceptionally high ratio of basic skills to what we call problem-solving focus in the activities that kids receive. I think this is largely a function of the kinds of pressures that have been upon teachers in schools and the kinds of curriculum that are being used in schools so that between kindergarten and first grade, if you look at the ratio of basic skills to a kind of problem solving discussion activity. And when we define basic skills as something like if I am watching an activity that a child’s engaged in, is it basically just a yes, no or a right or wrong answer to what it is that that child’s doing or is that activity encouraging some discussion, some problem-solving, some putting of pieces of information together to come up with a response? So if you look at that ratio, it is seven to one in kindergarten and first grade; pre-k, kindergarten, and first grade. By the time, you’re in third and fifth grades, it’s fourteen to one. You are going to get fourteen more basic skills kind of activity for every problem-solving activity that you’ll receive. This has implications for the kind of learning that we expect to happen in classrooms.

Many of us, many people are very concerned about the degree to which kids are offered opportunities to do problem solving and to put information together. Many of the kinds of things that drive our efforts to put pressure on schools are the kinds of tests. You’ll often times hear reference to the Tems Test and the International Exams in Mathematics and Science that kids take when they are in secondary school and we tend to score about twentieth of all of those countries on those tests. And people often use that information to say we need to teach the kids more basic skills in the early grades and that may in fact be true, but a lot of these tests test problem solving ability when you get to that level and so if we don’t build in some problem solving activity early, some people are concerned that you’ll never really get that kind of skill going even later.

I think this is particularly striking. What’s the ratio of listening, sitting, and watching to doing something? What’s the ratio of passive activity to active activity, where again you’re problem solving. You might be working on a project with kids and you see again in pre-k to k, it’s five to one so little kids get more opportunity to do things rather than sit around. But by the time you are in first, third, or fifth grade, for every occasion that you are actively doing something like a project and working on a particular activity in a more active fashion, you spend ten times that amount of time essentially sitting passively in a classroom and listening to someone talk to you.

Most of the instruction that children get is a whole class where there is myth that basically says that a lot of kids get taught in small groups. We simply don’t see it when we go into lots of classroom. I suspect many teachers do teach in small groups, but in the large numbers that we have, we see very little small group or cooperative learning and that’s fairly consistent across the grades.

This next slide is a little tough, but this shows in a given hour in first grade, the kinds of activities that you are likely to be exposed to. It’s broken down into intervals of sixty minutes. So the activity for the most part is going to be teacher led group activities. The subject areas that you are going to be exposed to, now this is across an entire morning so more than half on average of the kind of activity you are going to get is language arts, which I think fits quite well with the kind of expectations we have for first grade. And you see almost no math, science, social studies. In fact, kids spend more time in transition activities, that is sort of lining up and taking attendance, than they do in math. Those ratios are actually identical to what we see when we go into a third grade classroom and we spend all day with a child following them throughout third grade so we have the opportunity to watch the afternoon where you think most of the science, social studies, and stuff goes on. We don’t see much of it.

The range is what’s really important here. The range in every one of these goes from zero to sixty almost. So that what this means is that there are many classrooms, first grade classrooms, not a trivial number where no literacy activities occurred during the morning that we observed. And there are some classrooms we observed where every minute, there was something that we could code as a literacy or a language arts activity. What we are talking about here is kids being exposed to strikingly different environments in different classrooms. Okay? Strikingly different environments. And again, that range occurs within schools as much as it occurs between schools, between states, between districts. It’s not conditioned on whether you are in the same school or same school system.

Okay, let’s go to those quality ratings now. Remember we were looking at a seven-point scale here, ranges from low to high on all these things. So the low is the kind of thing where you probably wouldn’t want to have your kid and the high is the kind of place where you probably want to have your kid if you are selecting classrooms. And you see here in pre-kindergarten, the emotional climate is really centered around a four, okay? Right in the middle of the road. And in K-5, you see this bump up around grade five so on average, it looks like kindergarten to fifth grades are not bad places to hang out. They are socially places that we code as generally pretty warm and positive.

The pre-k environment is a much tougher environment to be in. That was really surprising to us because you typically expect environments for younger kids to really focus on some of those social attributes and social interactions. How busy are they? What’s the productivity in the classroom? Pre-k, k centered around four okay? Grades one through five centered around five so it doesn’t look so bad from the perspective of whether kids are actually doing stuff although remember what they’re doing is mostly fairly passive in the sense of listening to a teacher or working on a worksheet that’s a basic skills.

The ingredients that come from most of the research that we have been able to do that actually have been shown to contribute to kids’ growth over the course of the year, both concept development and quality of feedback, you see those centered around a two or three. These are quite low levels of the kinds of things that happen in classroom that most of us would sit and say, hey that’s terrific teaching. That’s really teaching that’s going to actually push that kid’s skill level a little bit further. Okay, so that’s kind of the descriptive piece. That’s what’s out there. So a lot of variation. Not a lot of evidence that we’re seeing high quality across the board and some areas of concern, notably this one here in terms of quality of feedback and concept development. Kids mostly passively engaged in classroom activities.

So one of the questions we might ask is, well is it better in private school? Well low and behold in the NICHD study, we had two hundred kids in private schools. So two hundred of these observations of a thousand kids every time are in private schools. So we compared almost two hundred private schools classrooms with seven to eight hundred public schools at each of those different grades. Anybody want to hazard a guess? No differences. On anything. You cannot sit in a classroom if you don’t know it’s a private school and predict that it’s a private school on anything we looked at. And in fact, when we look at whether attending a private school one, two, three, four or five years during the period of kindergarten through fifth grade has any measurable effect on your achievement over and above what your parents contribute to your achievement, there is no effect. Okay? So in this sense, vouchers may not be the answers for kids if what we are doing is using vouchers to try to put kids in a better environment than they might otherwise have had in the public schools.

What’s the consistency of experience across grades? So if you get a good teacher in first grade, what’s the likelihood you are going to get one in second or third. And I apologize to all of the teachers in the audience for talking about good teachers, okay? In the NICHD study, in the grades from k to five, we looked at children who were in the same schools who went from year to year and through our observation system we saw almost no relationship between their experiences from one year to the next. And that’s at the level of those quality ratings, but it’s also at the level of what we coded minute to minute so that if a child was in the classroom that exposed them to lots of literacy activities as a first grader, there was not much likelihood that they would get the same level or dosage of exposure or intensity of exposure as they moved up. Now that sometimes is due to the child. When we factor in as much as we can about the child and what they bring to the equation, we still end up concluding that it looks pretty random whether you’re going to get a particularly good teacher or a good experience, rich experience from year to year.

From pre-k to kindergarten we still have the same sort of thing, but the correlations are fairly low and non-significant so there is a high degree of fluctuation from one year to the next.

Okay, why is this important? Well in a minute, I am going to talk about what classrooms do that really matter. And they really do matter, okay? They matter a lot in non-significant ways, non-trivial ways to kids and they matter most for kids at the toughest end of distribution. That need it the most. Now if you are a kid who gets that kind of dose that matters as a first grader, it matters even more that you get that dose the next year as a second grader. Ted Sanders has done work using state-level data in Tennessee where they test kids year after year after year after year and he’s been able to isolate teachers who…you know kids get tested at the end of kindergarten. They go into first grade and they get tested at the end of first grade. He’s got all the teachers in the state and he’s able to look at that information and be able to detect within that information, teachers who make reliable boosts in kids. So kids who come in not doing so well who almost all the time, if they you exposed to that teacher, you get a bigger boost than if you are exposed to the average teacher in that particular state. Then he has been able to say, okay, what’s the likelihood that if you’re not doing so well at the beginning of your school career, how many of these kind of teachers do you need in a row before once you get a bad teacher, it doesn’t matter so much, okay? And basically his rule of thumb is that you need between four and five teachers in a row that are providing these kind of boosts over and above what the average teacher provides so that if you hit a teacher who doesn’t provide those kind of boosts or is sort of associated with declines in kids performance, that your performance still tracks positively. Okay, four or five years. There is no evidence in our data that suggests that that’s likely to happen.

Okay, let’s look at some profiles of classroom assets. When we look at, in this particular graph, this is quality scaled from zero to seven over here. This is the midpoint, this red line. The light blue is emotional quality and the dark blue is instructional quality. And what we did was subject our ratings to a computer program called Cluster Analysis which means it tries to find a lot of classrooms that are similar to one another and create groups and when we create these groups, we find a number of different patterns. You’ll see obviously here this is the kind of group that I bet everybody in the audience would like their children to attend. That group is fourteen percent of the population in the one thousand observations of first graders that we did. It’s the smallest group among all the different groups that we observed. And notice here that these two groups, which compose almost forty percent of the entire sample have those instructional quality bars fairly low. This particular group of classrooms, twenty percent is pretty low on both of those. This would be a particularly worrisome kind of group of classrooms. One we should be paying attention to trying to help that teacher.
Okay, we did this again in first grade and not dissimilar findings. So we find one group that looks pretty good on everything. Twenty-three percent of the first grades, but then you find again, these couple of groups. This one is actually a nice test case. Not so high on instructional but not too bad on emotional. Same with this group so it would be interesting to find out whether these two classrooms produce learning. They actually don’t produce it as much as we would like. Not as much as this particular classroom. And we see this classroom here. Seventeen percent – quite low on the instructional dimension. If you paired it with this one, you’d have almost forty-five percent of the first grade classrooms in this observational sample. Not particularly high on the kind of ingredients that we think matter the most and have been demonstrated to matter the most for kids’ achievement gains.

So what aspects of schooling predict kids’ performance? We take an approach where we try to look at value added. That is, what is the contribution of the classroom environment to the child’s growth in achievement and growth in social skills? We try to control for family and demographic factors. The kids’ performance on prior tests. All those structural features of schooling that we talked about – the teacher’s education, class size, that kind of stuff. So we are looking at whether and also whether features of schooling matter more for different kids. Different kinds of kids. And what we have been able to detect quite clearly is very robust findings that teachers for academic achievement and social skills when children are exposed to classrooms in which there is evidence of teachers’ sensitive responsiveness. The kind of individual involvement with children that I mentioned before. When feedback to the children is high and frequent. The kids get feedback on their performance and interactions with teachers. That language between teachers and children is rich conceptually. When teachers and children are provided engaging learning formats that draw them in rather than have them sit passively. And when teachers report positive relationships with students. All of those things contribute in a very measurable and detectable way to kids’ performance. Not more than families do, but definitely over and above what families do. Quite reliable. We call this kind of teaching here intentionality. It’s focused. It’s engaged. It’s a sense that it’s directed. That you watch these interactions and you know they are going someplace in terms of learning for that child.
We also see higher levels of social problems under some circumstances. This is particularly the case in first grade in one study. That parents actually report their children to show many more signs, not dramatically more, but significantly more signs of anxiety when they are exposed to highly structured classroom settings that lack social and emotional responsiveness. So if all you get is a high pressure without much warmth and high social value, then kids show some higher signs of anxiety and problem behavior is also increased in classrooms where children are routinely given activities where there is very low engagement value. Where actually the activities call for more passive kinds of involvement on part of the child and teachers report more conflict and we observe less sensitive interactions. Okay. These are not large effects, but they are significant. School matters. Okay? Sometimes I get in front of people and say school matters. They say big deal, school matters? In fact, much of the federal debate is around whether school matters. What’s more important is that these effects that we are looking at matter more for certain kinds of kids. For certain kids. And we’ll talk a little bit about that in a minute. But just in terms of some other findings from this research, we see almost no effects whatsoever on kids’ social or achievement growth for teachers’ education, their experience, class size, public or private schooling. This is all the stuff that generates lots of heat, but not much light in terms of public debate. We see robust, consistent and stronger effects for what happens in families and most of all, what happens in families before kids go to schools that sets them into school on a good, solid trajectory.

We looked at particularly moving from kindergarten to first grade, whether being exposed to particularly high quality emotional environment in first grade or a high quality instructional environment in first grade made a difference for different groups of kids. In this particular case, we are looking at kids who come from families where there is a low educational level. In this case, relatively low in this sample meant high school degree or below and people who came from a relatively higher educational level, which meant beyond a high school degree, most of them college degree. Now look here in this particular case, these are standardized tests of achievement where there has been an adjustment for the child’s prior achievement and then we’ve rescaled them. But essentially what you are able to do here is to compare first grade achievement for children from low education families, this is under conditions of low instructional support and this is the same group of kids who just happened to get into a high instructional quality environment. And you see in fact, they are indistinguishable from the kids who come from the college education and above families in that particular environment and pretty close in that environment too. But remember the low instructional support environments tend to be fairly frequent so you are just lucky if you get into this environment rather that that one. But it matters. In this case, first grade is contributing over three points to that child’s standardized test score. Not bad. If you gain three points a year, a couple years in a row and solidify those gains, you might close gaps.

How about an emotionally supportive classroom? So does it matter if there is emotional support in the classroom? Here we took kids, who their kindergarten teacher said had lots of adjustment and achievement problems in the kindergarten classroom, at the end of kindergarten year. These are sort of kids where teachers sit down and try to decide where these kids go the next year. Nobody really wants them a whole lot. So you have kids with no problems and you have kids with multiple problems and you put them in classrooms that differ on the levels of emotional support and look at the multiple problem kids. Again, standardized tests of achievement over here. These multiple problem kids in a classroom with high levels of emotional support are again indistinguishable from the kids who had no problems the year before. Okay? These are not small groups either. We are talking about in a sample of about a thousand kids, we are talking between a hundred and two hundred kids who are in some of these groups. In this one here, you have almost a four point gain that can be, in part, attributable to that particular environment. Now again, that is a high level of support. Low and moderate levels didn’t make it. You have to have a particularly high level in that particular case. So schools matter. They matter for everybody and they matter even more for kids who need it more. That’s the story there.

So in the elementary grades, there is this generally positive social climate. We see lots of variability. We are concerned about this lacking intentionality. No associations with structural features. Teacher credentialing, education experience, curriculum doesn’t matter what we see when we compare groups on those things. It doesn’t matter. The everyday experiences in classrooms and relationships with teachers, that’s what matters.

So if that’s what matters, then the question is how do you fix it or how do you help make it a more uniform experience for kids not lucky, okay? And make it an experience that may happen for more children. Okay? So the implications that we’ve derived from this and I will talk very briefly about what we are working on now, is that as we look at our data, the focus of educational discourse in terms of school reform and improving a lot of life in classrooms both for kids and for teachers needs to be paying attention, from our perspective, on relationships and interactions. Okay? Not structural features. We should not be spending all this time talking about class size or as much time as we talk about curriculum. Those create platforms okay? They create sort of a common denominator, but it is a fairly low bar that most of what happens in a classroom that we’re learning, gains in learning can be attributed to what goes on in classrooms happens because of what teachers and children do together. So it’s not just the curriculum, it’s how it’s implemented. It’s not courses or degrees, but it’s training and support for teachers. Very different implications for what we do in the Curry School.

We talk about striving toward moving in structural dimensions and implications. Sort of up the scale of quality. We think these classrooms seem to be places that are socially positive, but instructionally a little passive and so how do we meld good social interactions in a warm positive environment with a little bit more intentionality is the real question that I think we face in trying to scale up.

How do we guarantee more uniform and consistent exposure to high quality? In our view, resources need to be aimed at supporting teachers and high quality teaching and in fact we talk about having standards for classrooms and not standards for kids. That in fact that shift in the discourse would be to focus on ensuring high quality of the type that I think we observed rather than high quality of the type that you find out from when people fill out questionnaires of whether they have a four-year degree. And in our model, we rely a lot on standardized observation to provide accountability for classrooms and support to teachers.

So we are all interested in social and academic outcomes for kids and we are really interested in improved teacher outcomes. And in fact, this needs to be even far more elevated as a concern because many teachers are leaving the profession and the shortages are going up dramatically. We spend a lot of time talking about professional development. Whether we are spending enough money. Whether we are testing the right things. Or whether we have the right curriculum and we try to line these things up. But the only way they really line is through what happens between teachers and kids in the middle. This is sort of the point that I have been making. And we think that the data that we’ve been able to present here now from these large observations where we have quite reliable findings about the kinds of attributes of classrooms that matter, provides us a platform to now go in to see whether we can start helping teachers to change those things and support the kind of work that goes on in classrooms. To essentially raise the ratings that we might otherwise get if we were in these classrooms. And so we use an instrument called the Class to do this and we are looking at ways in which the observational assessment essentially filters those kinds of attributes of classrooms that we spend so much time on and in particular we are very focused in the issue of professional development and training.
And so for the next just couple of minutes I am going to talk about how we’re embarking on a new initiative working with the state of Virginia’s pre-kindergarten problems. We’re starting it at a very early phase here because we think that’s the evidence is where it makes the most difference for kids. And we’ve tried to in this way invent a new way that schools of education reach teachers in the field and new ways for teachers to receive support in their classrooms just based on their actual practices with kids, not support that’s based on them coming to me and me lecturing to them in my classroom at the Curry School. So we call this new initiative My Teaching Partner. This is the initiative that is supported by the grant from the National Institutes of Health.

Teachers receive different layers of support in this project. We are interested primarily in trying to impact student growth in language, literacy, and social relationships. These are the key aspects of school readiness. We give teachers a whole set of activities in language, literacy, and social relationships, but as we’ve said, the key thing is how do those things get implemented. Through teacher-child interactions and relationships that we in this particular project assess using this observational tool that we’ve developed based on the work that I just talked about. And how do we provide this in as efficient a way for as many different teachers and we’ve used the World Wide Web as a mechanism for doing that. And this is in many ways an experiment.

Many of you may be familiar with telemedicine where our doctors at U.Va. are able to communicate with doctors in Southwest Virginia about a particular procedure that that doctor in Southwest Virginia may actually be doing on a patient at that moment. And in many ways, we think this is the educational analogue of telemedicine. Where we’re able to contact directly teachers about their practice in their classroom and provide them support and feedback about the work that they are doing.

So the way this works, I am going to run through this very briefly, but everybody gets access to a website and on that website are a whole lot of different resources. We have two hundred and forty teachers enrolled throughout the state. This is actually well more than half the number of children in the state of Virginia that are enrolled in the Virginia pre-school initiative are covered by this intervention. About thirty six hundred kids in these particular teachers’ classrooms. All teachers receive access to activities. One of the challenges in early childhood education is that teachers are having to struggle with this issue of how much I actually teach kids and how much I teach them through play. And this is sort of a battle that goes on a little bit in early childhood education and much of what teachers are told to do – they are told to teach and told to teach literacy, but they are not given a lot of support for actually how to do that. And so we’ve spent a lot of time writing activities that can cover a teacher’s work from September 1st to May 31st and that work with kids around literacy and social relationships.

What’s very cool about this, in our view, is that all of these activities are key to a whole set of extras that teachers can access on the website. And a second group of teachers actually gets access to these extras and they involve teaching tips for how to implement these activities well and some other examples, video-based examples of how these activities look when they are being implemented in a way that we would rate as high quality using the assessments that I’ve described before from our observations. So this would be an example of a teaching tip that a teacher might receive for how she might implement a formal logical awareness activity. But even more intensive is the layer of support we now provide teachers and a third of the teachers are enrolled in what we call a consultancy relationship with people who sit here at U.Va., don’t travel all over the state, but who are able to see because these teachers tape themselves implementing our activities and send them to us. Our people here are able to see teachers as they’re practicing in the classroom and are able to provide them with direct and regular feedback and support about their practice in that classroom. So this is an example of what one of our teachers is enrolled in an activity particularly here and the consultancy relationship is really built around that kind of observation and so what really happens in this particular case is the teacher will videotape her implementation of an activity at a particular time.

Some of these teachers are from very far away from here. They ship us the videotape. Our consultants look at that videotape, edits it and posts it to a private website with some questions for the teacher to look at the edited tape. It’s edited in a particular way. The teacher gets some feedback about her practice and support. She gets to reflect on the consultant’s questions and observations about her practice. So the teacher gets to view this and then the consultant and the teacher meet over the internet in a video conference, a sort of mini-video conference over the internet, and discuss this whole cycle of observation and activities that the teacher has implemented. This cycle repeats itself roughly every two weeks so it’s a continuous stream of feedback and support to teachers. Not a shred of evaluation in it, okay? Teachers do not receive any…we never talk to principals about this information. In fact, most principals…we had a principal talking about this just last week saying that he wished he was more involved in this kind of support to teachers and in fact was glad that his teachers were able to get this kind of support from us.

So thus far, for this two hundred and forty teachers as we try to raise quality in some level or even it out for kids across the state in these pre-k programs. We have over three thousand visits to the site. We have about two visits per week, per teacher. In the consultancy group, teachers since the beginning of the year through January have engaged in six of these cycles of feedback where they have been able to get this kind of individualized personal feedback and support about their practice in their classroom. I should say all these teachers are enrolled with us for two years so this is just the first year of two years. So what we’re really trying to do here is reinvent professional development. Okay? From an education school standpoint to say as an education school, we can begin to bring professional development. Make it focused on an individual teacher’s practice and driven from the best scientific evidence that we have available for what matters in classrooms.

So now you know I told you I was not all that confident in stories, but stories are great when they really say the thing that you want them to say. So I am going to end with a couple of stories that teachers have told us about their engagement in this process.

This teacher tells us, “I think teachers and administrators have a great deal to gain from this type of training and consulting.”
“ This is a fantastic collection of activities. It’s going to be great to have these. I wish I had something like this to help me when I first started teaching.”
“ I was speaking with a previous administrator and told her about my participation with MTP. I told her that I felt this was a breakthrough in teacher training and could have a huge impact.”
And this is a group of teachers. “We had a meeting today with three of us…that are involved in your project and we were all RAVING!!! WE LOVE IT!!! THANK YOU again for this outstanding opportunity.”

So I think in some ways what I’ve tried to describe here in a brief and sort of run through fashion, I apologize for going fairly rapidly through some of this, but that what we’ve done I think over the last ten years is to be engaged in a process of research on classrooms that first, try to describe what’s out there and try to say what is it that matters in classrooms and what is it’s distribution in the population. What are kids exposed to? Is there a problem that we need to fix or isn’t there a problem that we need to fix? Is what we observed related in ways to the kinds of things that we debate about in education? The kinds of things we think are going to fix it. And point of fact, it doesn’t look that way.
We’ve been able to detect within those observations, aspects of classrooms that do seem to matter in very reliable, very robust ways for kids. Not huge effects, but very detectable effects that matter and could matter even more if they are pinned one to another to another over the long run.

And then finally what we’ve tried to do is to say if those are the things that matter for kids in classrooms and teachers do seem to be struggling honestly to try to provide them to kids, what is it that we as an education school can do based on that knowledge to engage in the kind of support to teachers about their practice that they say to us matters for them. And that’s in the end what we’re trying to do here.

All I have now as I said are stories. I hope that in another year or two I can give the talk and tell you that we actually have findings associated with that. I appreciate your listening to me. Thank you.