Summit Materials

Andreas Schleicher

Transcript <a href="/bb4e/video_Schleicher.html#trans" title="http://www.edu.gov.on.ca/bb4e/video_Schleicher.html#trans" style="display:none;">Transcript</a>

Mobile

Transcript

Andreas Schleicher: Thank you so much. I feel very honoured to share some of our lessons from global education comparisons with you this morning. I want to illustrate what our comparisons show different systems achieve, but really put the emphasis on those systems which show what's possible and strive for quality, excellence, equity, and coherence in systems. And, of course, Ontario is part of that success.

This is going to be a perspective — can I have my slides? This is going to be a perspective from 30,000 feet above. It's not going to give you all the detail, but the idea is to give you an impression of the changes that we have seen on a global scale. Some of the changes have been very profound.

In the past, learning was a place we worried how to get children to school. Now, increasingly, learning is considered an activity. It's pursuing students — students pursue it wherever they are in their life.

In the old bureaucratic education systems, teachers were often left alone in classrooms with a lot of prescription what to teach. In the most successful systems today, we set ambitious goals. A clear about what students should be able to do, and then provide teachers with the instruments and tools to decide what they should teach to their students in their situation.

The past was about delivered wisdom. The future is about user-generated wisdom.

In the past, students were taught in similar ways. Today, the challenge is to embrace increasing diversity with differentiated pedagogical practices, and, again, Ontario is a great example here.

The goal of the past was standardization and conformity. Now, it's about being ingenuous, about personalizing educational experiences, about realizing the extraordinary talents of ordinary students.

Education systems have always talked about equity. Now, we measure their success in delivering equity by how successful they moderate the impact of social background on learning outcomes, and we see some data on this later on.

The past was curriculum-centred. Now, it's about being learner-centred.

In the past, the policy focus was on the provision of education. Today it's an outcome, shifting from looking upwards in the bureaucracy towards the next teacher in the classroom, the next school; about creating networks of innovation.

In the past, we emphasized school management. Now, it's about leadership with the focus on supporting, evaluating, developing teacher quality, as it's called, and that includes many aspects from coordinating through curriculum assessment, evaluation, practice, and so on.

And finally, in the past, we considered social background and culture as obstacles to learning. The best performing education systems show you how you can capitalize on the diversity of student interest, capacities, and so on. They see diversity not as the problem of the knowledge economy, but as its source.

Let me show you how the global talent pool has changed. This chart shows you how the world looked in the 1960s in the portion of people having acquired baseline qualifications. We can see the United States far ahead, and then it goes up to Mexico and Brazil where you had just 1 in 10 young people with a baseline qualification.

But then, in the 1970s, things began to change. You could see some countries catching up, moving rapidly forward. In the 1980s, change continued. In the 1990s, once again — and what you can see is in the course of two generations, the global educational talent pool has fundamentally changed. What was the gold standard in the past is now the OECD average.

And you can see, if you look at Korea. In the 1960s, Korea had the level of economic development of Afghanistan today, one of the least developed education systems. Now, they get virtually every young person to graduate: 98%. Korea didn't change its culture. Some people think that it's all about culture. Korea didn't change its culture, didn't change its teachers. It changed its approach to enabling its assets to produce good outcomes.

You can see the changes even more clearly when you look at higher education, where most things are happening now. This chart is a bit complicated; every dot is one country. On the horizontal axis, you see how successful is a system to get lots of young people to complete university qualifications, the graduation rate. Higher education. On the vertical axis, you see how much it costs; also important these days. And, to make the picture a bit more complicated, the size of the dot shows you where the money comes from.

So, look at the United States. Number one in terms of getting lots of people to complete university qualifications. Number one in spending lots of money on them, and the most successful country in mobilizing a lot of resources, so private resources. Here, by contrast, Finland. Moderately successful in higher education. Moderately expensive, and all the money coming from the public purse; very little private money in the system.

The only thing I haven't stressed so far is that this was in 1995, and keep this picture in mind. I want to show you how the world looked differently if you look just five years later, in the year 2000. You could see what was the frontier has now become the benchmark for many countries. If you look, for example, here, Australia or United Kingdom, moving to the frontier. Or Australia, rapidly improving its output. But number one, now, is Finland. Within five years, lots of people completing education.

And the world didn't stop in the year 2000 as some people had predicted, no? Things kept moving on. If you look in the year 2001, the world looked like this. 2002, like this. 2003. 2004. 2005. 2006. 2007. Again, you can see every country has made progress. Every education system is better today than it was in the past, but the pace of change is very dramatic, such that the relative spending of countries has changed very significantly. And what you see is in this world where all work that can be digitized, automated, and outsourced can now be done anywhere in the world, and the yardstick for success is no longer just improvement by national standards; everybody's achieving this, but, really, the best performing education systems.

Look at this. You can see that was the gold standards just ten years before. This is where you find other successful countries; Australia, Finland, and the United Kingdom — sorry, I just want to show you one more point. The United Kingdom actually is a great example here. You see the rapid progress they have seen early in the years, in this charts? Moving rapidly to the frontier, but then they got stuck. They actually set themselves a very ambitious target. They wanted to get about half of the population, every second school (inaudible), to complete a university degree, and they have actually achieved that target. They have made it a reality, a very ambitious target.

Shows you that you can set targets and achieve them. What you can no longer do in this global world is prevent others, to suppress them, and that's why today it is so much more important than just setting sort of fixed ambitions; to make education systems agile, responsive to all the changes, you will see that, again. And if you look at, for the future, we're not stopping here; things go on.

But the challenge isn't just to produce more of the same skills. Education is now so much more than about accumulating subject matter knowledge. You can solve many conventional school tests today in seconds with the help of a smart phone. If you want to have your students being smarter than a smart phone, you need to think about how they can actually enable, create, share knowledge. You have to achieve this. In the fast changing world where we live, everything we call our proprietary knowledge today, every knowledge that we own is a commodity tomorrow, available to anyone else. That's no longer success, is accumulating subject matter knowledge and reproducing it.

We are seeing a shift from the world of stocks, with knowledge that is stucked up somewhere in individuals, depreciating very rapidly in value, to a world in which the enriching power of communication and collaborating flows is increasing. I'm going to show you this with some data in a moment. The challenge, essentially, now, is about new ways of thinking, about new ways of working, about creating new tools for working. And, finally, and I think this is very important, the capacity to live in this multi-faceted, fast-changing world as active and responsible citizens.

Again, let me show you this with some data. I don't have data for Canada, so I'll show you data for the United States. The chart shows you how the composition of the U.S. workforce has changed been '97 and 2000. You can see that work involving routine manual input, work you do with your hands every day in the same way, has increased in the '60s and '70s; people built a lot of factories, but since then, automation have kicked in. Outsourcing have kicked in, and you can see a decline in that kind of work.

When you look at non-routine manual work, things that you do with your hands, but that you do every day in a different way, you can also see a rapid decline, but this has sort of stabilized because there are many, even simple, jobs that are hard to automate, hard to digitize, hard to outsource, no? Somebody's going to clean this room; you can't put a computer. They drive you with a car; can't put a computer. You can't outsource your hairdresser. There are lots of jobs that may not be very difficult, but it's fairly stable. But you know everybody knows that. The reason I'm showing you this chart is a different one.

I want to show you what happened to what we call routine cognitive work. That's basically your capacity to learn something and, later on, to reproduce that knowledge. It's basically about cognitive work that you can put in the form of a script. And what you actually see it that this has been the fastest declining skill category in the labour market. The kind of things that are easiest to teach, that are easiest to test have seen the steepest decline in the demand, and that's real challenge for schools.

Now, where have been the winners? One of those skills categories is what we call non-routine analytic skills. This is not reproducing work you're learned; this is about extrapolating from what you have learned and applying your knowledge in a novel situation that you haven't been before, and, perhaps, nobody has been before. And that's actually, when I talk later about our PISA assessments, is that what we're trying to get at, with PISA?

There's a final skill category, rapidly increasing in demand, and that's about interpersonal skills of all kinds, the capacity to put ourselves in a complex situation, to communicate, to collaborate with others. Rapidly rising increases and, again, schools are challenged. Of course, everybody's curriculum talks about interpersonal skills, no? But at the end of the school year, they're giving students an individual exam or test, and so on, so the reality is still very, very far from those kinds of changes.

Again, at the OECD, we try to measure the success of education systems not by seeing whether students have learned what they have been taught, but by looking at whether they can use, apply, and extrapolate from what they have learned. I'm saying this here because some people say, “Well, this PISA test is terribly unfair. Now you're testing students with things they haven't seen in that form.” True, but life is unfair, no? Because that's exactly what it does to you every day after you leave school, so it's very, very important to do this, and we test every three years roughly half a million of children in OECD countries and the principle economies out of this in those kind of key competencies as you see that here.

Now, here's what the results show from our last assessment in science. Many of you will be familiar with them. The red area shows you countries that did not do so well, yellow is sort of average, and green are the countries that do really well by OECD standards, and you will find Canada among them. You will find many of the countries in Asia there. You can find Finland in Europe, so in every part of the world, you find systems that actually have managed to really do well in terms of the average performance.

I want to focus on this area of the scale. This is where Poland was in the year 2000. This is where Poland is in 2006. Within 6 years, Poland moved from a below average to an average performing country. Again, what you have seen in quantity before is happening in quality as well. Rapid improvement in some countries, and raising the standards, and raising the quality of learning outcomes. And, once again, to people who say, “Well, this is all about cultures. You can't really portray those experiences in your own context,” Poland didn't change its culture. It didn't change its teachers. It didn't change its students. What it did change was the approach to the education system.

But I want to put a second dimension into the picture which we pay great attention to. When you look at the distribution of student performance in each country, there are some countries in which social background has a huge influence on the success of students, where there are lots of winners and losers. But there are other countries where the gap between privileged and less privileged students is much smaller, and this is another important lesson that shows you that systems can actually moderate the impact of social background. Everybody wants to be here where performance is strong and the equity of opportunity is quite well-pronounced. Nobody wants to be there where performance is below average and there are large social inequalities.

And then you can argue: is it better to do well and accept disparities, or do you focus on equity and accept mediocrity? Many people thought, for a long time, that you have to choose between one of those yellow quadrants, but what the best performing systems show is possible, actually, you can do both. And, actually, if you look at this chart long enough, you will find that none of the successful systems has achieved quality by compromising equity, and, once again, you see Canada in the right place of this chart. Very successful.

Now, you can argue, “Well, does it really matter? How much importance should we really attribute to what fifteen year olds know in school?” Well, the best way to find out is actually looking to what happens to them afterwards, and that's precisely what we have done, and that's going to be my last complex data shot. I promise. This chart shows you, on the horizontal axis, the level of student performance when they were fifteen years of age. On the vertical axis, you see how much more likely students were to get into college or to into a well-paying job.

You can see, for example, that those kids, Canadian kids who had achieved Level 2 when they were fifteen on the PISA test, were twice as likely to move into college as opposed to those who had not made it to the baseline level. Those who had achieved Level 3 were four times as likely. Those who had achieved Level 4, eight times as likely, and those who had achieved Level 5, sixteen times as likely. This tells you two things. The first is great news for us. It tells us that the PISA test somehow seems to test things that really matter for kids in their future, but the second thing that it tells you is that what you haven't got right at the beginning of children's career is just very, very difficult to get right. The odds 16 to 1 tells you a story, no? One to 16 is your chance to move up to this area is you are a child that hasn't reached the baseline Level 2 of performance on this test.

Now, you can say, “Okay, they are only nineteen years when we looked at this. Maybe there are those kids who didn't do well caught up later in life.” You can look at what happened to them at — when they were 21, but you actually see that the gap has only become wider. It's not that continuing education and training moderate early educational experiences, they seem to reinforce them.

Now, you can say, “Well, you know, let's do away with this PISA test and let's put in teachers' judgment, no?” The judgments that teachers make every day in school in the form of marks, and you can actually see they also predict something, but they actually predict much less than a universal standard that has been assessed across schools, and that's a very — another very lesson. It tells you that as important as teachers' judgments are in the classroom, they are not enough. You need to have some objective way of looking where you stand, comparatively, to others. You can see the judgments predict less than you have otherwise.

Now, the interesting question, of course, is what have we seen about those systems that you saw in this green quadrant? The systems that demonstrate excellence, equity, and efficiency: how have they got there? The first thing everybody thinks to is money. Our data actually show that improvements in the quality of learning outcomes have gigantic payoff for societies, no? If every country would do what Poland has done, raise its performance by this level over the next 20 years, we're talking about 260 trillion Canadian dollars in additional economic input over the life cycle of these people. What this shows you is that the cost of improvement is a tiny fraction of the economic rewards that economic improvements bring with them.

So then, you can say, “If more education, better education gets you more money; does more money get you better education?” And you can see that it's not as simple. You can basically see that there is a slight relationship between what you invest and the outcomes, but there are lots of exceptions. Some of the best performing systems don't spend so much, and some of the high spenders don't achieve so much.

It becomes more interesting, actually — actually, one of the things I should say is that many of the successful education systems have one thing in common: they invest their resources where the challenges are greatest. And this is one area where there is enormous progress across OECD countries in the last ten years alone. In the past, school systems used to invest more money into the rich kids' schools. Ten years ago, that was the norm in OECD countries. Today, there are only two countries left with that policy. One is Turkey. The other is the United States. Every other country is putting more money into the more challenging schools, but that's the easy part.

You can still see that attracting the brightest teachers into the most difficult schools is still a remaining challenge that hasn't been solved with money. Class sizes have become smaller, but when you look at teacher quality, as we can measure it through our PISA assessments, you see that disadvantaged schools struggle with this, in many cases, as they did ten years ago. So on the macro-level in terms of resources, things have changed. From the school perspective, there's still a lot of work to be done.

But how do countries spend their money? The red dot shows you how much money actually arrives in the classroom. You can spend a lot of money, but not necessarily all of this ends, actually, up in the classroom. And you can see there is a world of difference since I measure this now relative to GDP per capita to bring spending capacity in there, and so on.

You can spend your money in different ways. One is you can try to put up high salaries to attract the best people into the teaching profession. If you look, actually, this is what a country like Korea has done, no? Putting up very, very high salaries, spending about twice the OECD average in terms of teachers' pay with the aim to get brilliant teachers into the profession.

Korea does another thing. They say, “Well, we also want to provide students with a lot of instruction time: long learning hours.” They do that as well. Costs money, drives up costs.

And then Korea does yet another thing. They say, “We don't want our teachers only to teach. We want to build a lot of time into the preparation of lessons, into teacher collaboration, into professional development, into building a profession. Nice thing, no? But costs money, and you can see it drives up costs enormously in Korea.

So how does Korea finance all of this when the red dot is only slightly above the OECD average? The secret is they pay for that with large classes. Getting better teachers, more instruction time, more learning time for teachers, at the expense of large classes.

You now go to the next country on the list: Luxembourg. As expensive as Korea, no? But in Luxembourg, teachers and parents really like small classes. All of the money went into small classes. But you can only spend your money once, and what you see here is that teachers have been left with little instruction time, low pay, and no time to do other things than teaching.

Now, it's very difficult to derive generic lessons from this, but if you look at this chart long enough, you find that many of the high performing systems prioritize getting the right teachers over small classes.

But it becomes more interesting; let's look beyond money. Once you look beyond resources, our analysis suggests that, first of all, schools and countries where students work in a climate characterized by high universal expectations, good teacher relationships, and high teacher morale tend to do better. What our comparisons also show is that most of the high performing countries place great value on ambitious educational standards, give curricula focus and coherence, and place emphasis on external gateways in the system that make it clear to students where education counts.

At the same time, universal standards that define this kind of shared vision of what is important tend to go together with access to best practice and professional development in schools in ways that support teachers, to expand their pedagogic repertoire every day in schools, to personalize learning for all students.

When you look at some of the top performing systems, you often see that they do several things well. First of all, they put the emphasis on getting the right teachers into the school, but perhaps, even more important, they evolve them and develop them once they are in this profession. They invest a lot of resources into continuing professional development of these teachers, into building this community of teachers, and, finally, they do make sure that every student has access to best and good practice of teaching experiences.

You can sort of put those dimensions on a chart like this, and it actually works very well when you map up OECD countries on this. Challenge and support. Where ambitions are low and teachers in schools are poorly supported, nobody would expect much. That's the sort of non-surprising quadrant.

But increasing the challenges through new standards, new tests, new school inspection, new publication of test scores, and so on, without backing them up with better support has often, actually, just led to conflict in the system. Among OECD countries, you find countless tests, countless efforts of more money, taking money away from them, giving them less prescription, giving them more prescription, without having much of an effect.

You also find systems — the Nordic countries of Europe apart from Finland are great examples. They have wonderful school environments. Wonderful working environments for teachers, but they're not very clear about what is expected from students, and they don't have this kind of dimension of challenge, and they're also not among the top performing systems. What you do see, quite clearest in the systems doing best, they get both dimensions right.

Let me just highlight, very briefly, one aspect, and that's teacher professional development. This chart shows you what teachers do in terms of professional development across the OECD countries, and there are different categories. You can see, for example, not many teachers engage in individual or collaborative research, no? It's not very common among teachers across the OECD. Even fewer teachers pursue professional programs that lead toward qualifications that are part of a career path. You know, that get you to a different point in your career. And then, when it comes to informal dialogue and collaboration, yes, every teacher does it.

This is what's happening, but what makes, actually, a difference to teachers? And when you look at this, you can see that the picture is actually quite different. Some of things that happen least are perceived by teachers as making most of the difference in their daily life. Individual and collaborative research; for example, qualification programs that get you on on your career: very, very important.

I want to introduce a second trend among many of the best performing systems, and that's about efforts to enable schools to become the primary driver of the profession of education improvement. In the most advanced education systems, strategic thinking and planning takes place at every level of the system. Every school discusses what the national standards mean for them; moves this vision forward. But let me emphasize again, what relates to results in measurable ways is what happens at school, not necessarily at district or local levels. Can measure, actually, very little in terms of impact. It's really the schools' capacity.

But, again now, our data show that two things go hand in hand in some of the top performing systems. When you look at some of the systems doing well, you see that they have the capacity to intervene when things go wrong. Very, very important. They recognize, diagnose failure. You know, one of the interesting point of data about Finland ...

[Recording pauses and starts again]

... and that's not necessarily to do with test-driven accountability. Some of the best performing systems actually do not have a national test, but they are very, very powerful devices to show schools where they stand and provide resources and data and evidence for those schools to do better. Accountability can have many dimensions, vertical lateral dimensions, but it's about creating this knowledge-rich environment that helps schools understand where they are and where they can be.

Let me illustrate this with the PISA data. This chart shows you sort of the combination we talked about: school autonomy, local responsibility on the one axis, and having a clearly defined vision. I've chosen some very simple proxies for both here, but you do get the same answer whatever you chose.

When schools operate in an environment where they have no clear external gateway and where they have no autonomy, you see them actually performing poorest, on average, across OECD countries.

Schools that have no autonomy, but that actually do operate in a system that clearly articulates what is important, have a quite significant performance advantage: 46 points on the PISA scale. That's more than the equivalent of a school year.

Schools that have lots of local discretion also do well even if they have no framework, but the best performing schools, on average across countries, are those that have local responsibility and operate in this overall framework of an intelligent system. And that's actually one of the trends.

If you, again, look what happened in the last ten years, it has really changed the world. You can basically see how things looked in the past. I mean, you always had to decide how do you create a balance between what you describe national, and the past has been about a tailoristic work environment for teachers and where you build capacity in schools, and this is never a 'one' or 'or,' but it's always the question of combination. When you look at this, even in the year 2000, you could say that most systems were somewhere here, the kind of industrial model of scoring.

Things have changed today. You can basically see this is where many of the education systems are just ten years later. We have only ten years of measurements so far, but then you look at some of the most advanced education systems, and you can really see how schools have become the primary drivers of success.

And one thing I just want to make clear is that this is not about public or private education. We see that in the following chart, and I'll go through this very quickly. This shows you the proportion of public and private schools. Red is public, light green is charter schools, and dark green is private schools, and you can see it looks as if private schools do better, no? They have, in many countries, a performance advantage, but once you account for social background, actually, public schools tend to do better, on average across OECD countries, and in some, quite significantly.

So the issue for public policy today is not how many charter schools you create. The question is how you enable every public school to assume the kind of decision discretion-making authority which you have in charter schools, and that's what the best performing systems show is possible to achieve.

Let me show you — summarize some of the school factors, how they come out. You can actually see — I mean, I show you what we have measured on PISA on the left side, and what we would have measured if all students, all schools, all countries had been equal in their socio-economic context. And you can see, basically, schools that practiced a lot of ability grouping, sorting, tracking, and so on, they don't tend to do so well. Getting rid of responsibility, putting responsibility to someone else doesn't seem to be an effective way to improve learning outcomes.

Schools that are selective do better, no? You can see a big advantage. No surprise. If you can select your clients, you have an advantage. But now, what if you made every school selective in your system? Does it have an effect? And, unfortunately, the answer is no. So it helps the school; it doesn't help the system. Very important distinction.

Now, this is quite controversial, the next point. Making results public, a system of public accountability, is quite clearly linked positively to results in very significant ways because this is very cheap to do. Many of the things we're talking about here cost huge amounts of money. This costs little, and you can see the effect that is there for all schools, irrespective of their social background.

Another point that is important is what we call instruction time. Here, in the case, it was science, and I'm saying this because there are a lot of people who say, “It's not the quantity of learning that matters; it's all about quality.” But outcomes are always the product of quantity times quality, and the quantity alone has a quite sizable, measurable effect.

Out of school tutoring usually tends to be negatively associated with results, but that's basically just a selection effect; let's not worry about that.

One hour of homework gets you little. This is also important. What's the most expensive resource in your education system? Not money. You can put as much money as you want in education. It will always pay off with even more. But student learning time, you can increase, and you can see using student learning time differently can lead to quite a different outcome.

Doing something else than teaching is also associated with results. For example, whether students — whether schools bring science — now, this is about science outcomes, so whether schools bring science to life: exhibitions, competitions, something that goes on beyond the classroom.

Finally, this we talked already about. I can go through quickly through this. This is sort of the level of autonomy, and most of the other things we talk about a lot come out. You see them, you measure them, but they're not really that decisive. For example, it looks as if the more public money you put into the education system, the less well you do, but that's simply not coming out in the net model because that's just the public/private split.

The same is true when you look at competition. We put a lot of emphasis these days on increasing competition among schools, no? You can say it makes systems look better, but, actually, there's no observable net effect, at least not on our measurement system. And the same is true for many of those other variables. I don't want to go through this in great detail.

What I want to do is introduce one last factor into our diagram when my slide comes up. What you see also that in all countries that do well on PISA is the responsibility of schools and teachers to constructively engage with the diversity of student population. There is very little grade repetition, very little tracking, very little streaming. You can't say, “I did the right lesson and have the wrong students, who belong somewhere else.” You can't say, “I have not done a good job this year. Somebody else should take care of it next year.” You cannot. You have none of those devices.

Engaging, integrated systems, but at the same time, those systems don't leave — put everybody in the classroom, but they are capable of personalizing learning experiences to an extent that we can measure this by asking students. That's very, very interesting. Coming out really quite strongly through the system, that countries differ in the extent to which there are good teacher/student relationship, and by this I mean whether a student understands that the teacher expects a lot of them, whether the students believe that the teacher will help them when they are in difficulties. You can shape those kinds of conditions. Individual teachers are also aware of their specific weakness. They have good access to best practice in the school, around the school, on how to improve things, and things like this. Very important.

And let me just illustrate you this with one more of this chart. Remember this? On the vertical side is quality, on the horizontal side is equity, and you can see now, when you map systems into those that are highly stratified, no? Basically, those that track and stream and differentiate students, that try to solve problems by routing students into different pathways, they're all on the left side, no? On the red — marked in red. That tells you that schools tend to be really good in selecting students by their social background, but not very effective, generally, in selecting them by their academic potential, no? Most of the systems that do well in terms of quality and equity know how to embrace heterogeneity in the level of the schools.

Let me put all of those data on one slide. These are all the numbers. What have we shown to matter at the level of the education system? We're talking about systems here at the level of the school. And what matters in terms of equity? Some of the things are unsurprising.

Discipline, a very good predictor for school performance.

You can also say a teacher, no? Teacher behaviour is measured. You know who's the best measure in our data system on teacher behaviour? The students; we ask them and they tell us a lot about what actually goes on in the classroom, and so on.

Parental pressure. We all complain about it, but it's actually a very important predictor for success, and so on.

Teacher/student relationships, the perception that there is a positive relationship, predicts learning outcomes.

Now, when you do look at the ways in which countries deal with diversity, you can see that grade repetition cuts right through, negatively, through the quality, equity.

Tracking and streaming, not so closely related to performance, but quite closely related to negative impact on equity, and so on.

Standards and accountability — just moving these factors a little bit upwards. You can actually see that having a clear vision of what is important in the education, clearly visible to students in the form of high stakes gateways, is important.

Standardized tests, we don't measure much of an impact on quality. They have an important impact on equity of the system.

Pre-primary education, also an interesting feature. You actually do see that what children did before they came to school comes out in their results at age 15. Very interesting, isn't it?

Now, final slide, chart. What have we really learned from all of this? There are seven things I want to illustrate here.

One is when you look at the most successful systems, they have a commitment to education and the belief that every child can be successful. And that sounds trivial, I know, but that's one of the things that you actually see very, very clearly. They have universal educational standards. All students are confronted with the same systems. We are not leveling down the horizontal for students from disadvantaged backgrounds. As opposed to systems that basically believe that students have different destinations that should be served through different systems. It's one of the important features. There's also a clear articulation as to who's actually responsible for those universal standards. One factor.

Second factor is clear, ambitious goals shared across the system and that are aligned with high stakes gateways, examinations, and so on, and also aligned with good instructional systems. There's a well-established delivery chain that goes from those goals towards actually happening in classrooms, and we can actually measure this.

Second, we see a high meta-cognitive level of instruction. Students don't do only well on the test. They understand what is important. They have access to effective learning strategies. They can articulate them. They know what they are doing. A very, very important predictor on learning outcomes.

Third, capacity at point of delivery. We talked about how these systems recruit teachers, attract, develop, retain teachers, no? It's not about money alone. Some of the most successful systems — again, you cite Finland or Korea, no? Korea pays teachers well. Finland does not, but both systems provide a very attractive work environment for teachers. Teachers learn, develop; where teachers are knowledge workers that actually have a career in front of them. Keeping teaching an attractive profession, and, again, I often hear, “Well, you know, we can't do much because teaching is not attractive.” You can changes those things. England, very, very successful in raising the status of the teaching profession. Finland, as well, over a longer period of time. In many of the systems, teaching hasn't always been as attractive as it is. System-wide career development, another of those features coming out through those kind of data.

Third factor. This is about incentives, accountability, and knowledge management. When you hear the word “incentives,” you think about, “Well, do we pay teachers more when they get better test scores?” That's not actually very common in the most successful students. It's more about telling teachers what is important. Telling students, giving those kinds of signals. Encouraging. Intervening inverse in proportion to success. Providing the kind of support. Having a balance between vertical and lateral accountability in the system. Having effective instruments to manage and share knowledge in the system. Having knowledge not only trickling from above to the ground level, but having knowledge also being generated at the ground level and being understood and heard at higher levels. A capable centre. You're very, very impressed when you speak to people in the ministries of high performing systems. They are not bureaucrats. They really know what they want to achieve, and they know how to achieve those kinds of things. Very impressive.

Next point, investing resources where they can make most of a difference. And, again, you can say this is something education systems in OECD countries, all but two have more or less achieved. They put the money where the challenges are greatest, but translating this into a better quality of resources is a big, open agenda, and very, very few systems — you can actually see, perhaps the Nordic countries in Europe have done it. I do think, at least from our data, Ontario has some way to go, so translating resources into quality is a big challenge for them.

Next factor. One of the interesting things is that those systems doing well keep learning. They have always and continue to look up, always looking outward to next practice, best practice, anywhere. These are the countries that have been traditionally most interested in international benchmarking, and so on. When we released the results from the first assessment, on the PISA assessment, the Finnish Minister of Education could have said, “We're number one. Great.” Instead, she said, “Where do you see tomorrow's challenges, tomorrow's threats to our success today?” Basically, this is what it means to continue looking outward, continue looking forward.

And, finally, and that's the most important part of this story. That's about policy coherence. Over long periods of time, I think the Premier made that story very impressive, that something that you see in many of the successful systems, things that count, count over long periods of time. Coherence across different areas of policy. Another here, consistency of implementation. Michael, I'm sure, will tell you a lot about this, and also fidelity of implementation, making sure that the things get done in the right way.

What international comparisons really show you is that success is possible. Success in quality, excellence, and equity. You can achieve all of those things in reasonable periods of time. None of the countries that you see at the top league internationally have been there for very, very long periods of time. These are all fairly recent successes. It's achievable, and it's achievable in a reasonable framework of time, and we begin to learn about some of the factors associated with this success. Thank you so much.

Man: Now, it's my pleasurable task to thank Andreas. Before I thank him formally, let me just say that in your working group sessions, which I'll come back to in a moment, you will see Andreas on a panel, and so any questions that have been raised by the presentation, that's the place where you'll be able to raise them with him.

Andreas, I think of you as the person who put the “evidence” in evidence-informed policy making. This is absolutely remarkable data. Your mastery of the data is phenomenal. Your ability to take very complex data across many, many systems, about many, many factors, and explain it and present it beautifully, is ...

[End of recording]