Programme for International Student Assessment (PISA)

Launch of PISA 2006


Speech by Angel Gurría, OECD Secretary-General

Tokyo, Japan Press Club, 04 December 2007


Good afternoon, Ladies and Gentlemen,

Thank you for inviting me to speak at the Japan Press Club. It is an honour to be in this prestigious building, and to have the opportunity to address such an influential audience. I am delighted to be able to share with you the new results of one of our most widely known and cited research products, the latest report from PISA, the OECD’s Programme for International Student Assessment.

PISA examines how well individual national education systems are doing in equipping their young people for the world of tomorrow. In the highly competitive globalised economy of today, quality education is one of the most valuable assets that a society and an individual can have. Skills are key factors for productivity, economic growth and better living standards. Effective and innovative education policies open enormous opportunities for individuals just as faulty educational systems result in declining standards, exclusion and unemployment. They also underpin healthy and vibrant economies. That is why education plays a central role in OECD’s agenda.

Dramatic changes in the global talent pool over recent decades oblige countries to assess the educational progress of their young people in a global context. Today, countries like China or India are delivering high skills at moderate cost and at an ever increasing pace. Other countries – including the developed countries that are members of OECD -- cannot ignore these competitive pressures, on pain of harming their own future well-being. 

Governments are aware of this challenge and I am happy to present the results of our latest PISA assessment where no less than 57, countries participated, up from 41 in 2003 and 28 in 2000, covering close to 90% of the world economy.  In my remarks today, I will begin by showing where countries stand in terms of the science knowledge and skills of their 15-year-olds. I will then highlight where education systems can be, by showing you what the best performing education systems achieve in terms of quality, equity, and efficiency. And I will conclude with identifying some of the policy levers that PISA identifies for raising quality and improving equity.

There are many reasons for doing the global PISA launch in Tokyo. One of them is certainly the remarkable progress that Japan has achieved in preparing for the educational challenges of a globalised world. In the 1960s, Japan ranked 14th among OECD countries in the proportion of its population with tertiary qualifications. Today it ranks second, just after Canada.  But there is more to be proud of. As I mentioned, PISA 2006 focuses on the science performance of 15-year-olds. And it shows that Japan, along with Finland, Canada, Australia and Korea, achieves not only high performance but also offers an equitable access to learning opportunities. Students from all socio-economic spheres are given an opportunity to realise their potential, and they take these opportunities up. These results are a wake-up call for other countries that are doing less well in this respect. But OECD countries also need to look outside the OECD area, where we find three of the 5 top performers in PISA 2006 (Hong Kong China, Chinese Taipei and Estonia).

But PISA is much more than just a ranking. It also tells countries about their strengths and weaknesses compared to their peers. Japan is a case in point: Japanese students did extremely well when asked to use scientific evidence: reproduce knowledge, interpret evidence, draw conclusions and identify the underlying reasoning. But they struggled to identify scientific issues and to figure out the features of a scientific investigation, in short they have difficulties to apply their knowledge to novel situations.

This is an important finding. Students who learn just to memorise and reproduce scientific knowledge and skills may find themselves ill-prepared for tomorrow’s job market. The current curriculum reform in Japan aims to strengthen students’ competency in the investigative arena so there is recognition of this problem. Successful countries in this area, such as Finland, New Zealand, Australia, the Netherlands and Canada, may provide useful points of reference and offer best practices.

We have devoted most of the attention in PISA 2006 to science, but PISA looks at other competency areas too. In mathematics, which had been our focus area in 2003, Japan maintained a high achievement level of 523 score points, even if the lower performance of girls dragged overall performance down slightly. In reading the results remained broadly unchanged since 2003 too, but at a much lower performance level, 498 score points. Accessing, managing, integrating and evaluating written information appears to be the greatest challenge for Japanese 15-year-olds.

But looking at averages only is not enough. How skills are distributed also matters. High-level skills are particularly important for countries like Japan that operate at the frontier of technological development. On average across OECD countries, 1.3% of 15-year-olds reach Level 6, the top level of the PISA 2006 science scale. This group can consistently identify, explain and apply scientific knowledge in a variety of complex life situations. Japan has twice as many top performers as the OECD average (2.6%). This is a very good result but it could be even better. In New Zealand and Finland the share of Level 6 performers is at least 3.9%, three times the OECD average; these countries also do better in the percentage of students reaching the next best level. This is a very important finding because, even if PISA cannot establish the causal nature of the relationship, the proportion of Level 5 and 6 performers at age 15 is a good predictor for a country’s research intensity; it  explains 70% of the OECD cross-country variation in the share of researchers in total employment.

But science education is not just about those who move on to become brain surgeons and particle physicists. It should also give citizens the ability to participate fully in society and in the labour market. This requires baseline scientific competency at least at PISA Level 2, which requires competencies such as recalling single scientific concepts and using results of a scientific experiment represented in a data table as they support a personal decision. Many countries have a serious problem with low performers; across the OECD, on average 19.2% of students perform below the PISA baseline Level 2.  On this score, Japan again looks good with only 12% of 15-year-olds not reaching Level 2. But again Finland looks better with only 4%. Our experience in many countries has been that it is extremely difficult to reverse patterns of poor performance, and as skill demands in labour-markets increase, the social costs of poor educational levels are high and increasing. Therefore, even if the proportion of low performers in Japan is still low compared with other OECD countries, monitoring this will deserve continued attention.

In PISA 2006 we also looked at students’ attitudes towards science. Why is this important? Competing successfully in a globalised world increasingly depends on countries’ ability to innovate. This in turn will require major investments in scientific infrastructure and the ability to attract qualified individuals into science-related professions. Governments have to secure broad public support for scientific endeavour. Science and technology have enabled remarkable achievement over the past 100 years, but addressing these challenges successfully will require countries to make major investments in scientific infrastructure and to attract qualified individuals into science-related professions, as well as to secure broad public support for scientific endeavour and the capacity of all citizens to use science in relation to their lives. Peoples’ attitudes to science thus play a key role. In general, Japanese 15-year-olds report a fairly strong level of appreciation of science, even if it is not as strong as it is in many other OECD countries. But they often attach less personal value to science than their peers in other OECD countries. They fail to see the opportunities that science can offer for their own lives and their motivation for future science learning is low. Only 8% of Japanese students say that they expect a science-related career at age 30 (OECD average 25%), the lowest proportion in the OECD.  Last but not least, while doing well in the PISA test, Japanese 15-year-olds were least confident in their science abilities among OECD countries.

We reach similar conclusions when it comes to science and the environment. Japanese 15-year-olds report a below-average level of awareness of most of the environmental issues, with environmental awareness and science performance closely linked. It also seems that many young Japanese may not yet have fully grasped the seriousness of the environmental challenges that we face, as they tend to report an above-average level of optimism regarding environmental issues, and the less they know about science, the more optimistic they report to be that the environmental challenges will be successfully addressed. 

I would like to mention another very important dimension of all our work at the OECD: gender differences. Here, the PISA 2006 results are very encouraging: In 22 out of 30 OECD countries, boys and girls perform equally well in science.  But will this result in gender parity later on in life, in career choices, in jobs, in salaries?  Of course, we do not know what study choices Japanese 15-year-olds will actually make when they are older. But PISA does show that Japanese girls are much less motivated to learn science than boys. Girls rarely participate in science-related activities, even though Japanese girls do well on academic tests. This is an important policy concern.

Knowing how things are is important. But how can we make things better? The results from the OECD PISA Assessment leave us with the question what schools and school policy-makers can do to raise performance and to moderate the impact that socio-economic background has on student performance.

Some people say that strong educational performance is all to do with money. And indeed, the results from PISA show a positive cross-country relationship with expenditure per student. But the relationship is far from straightforward: Finland, New Zealand, Korea, Japan, Australia and the Netherlands do well with moderate expenditure, while top spenders like the United States and Norway perform below the OECD average. The PISA results also show that, across the OECD area, student performance has generally remained flat between 2000 and 2006 while expenditure on education in OECD countries has risen by 39% in real terms during this period. 

So money is important but not sufficient to raise educational performance. It matters at least equally how educational resources are invested. An adequate supply of teachers and quality of educational resources at school are associated with better learning outcomes. But more importantly, there are a number of school policies and practices that are crucial for performance without being necessarily tied to resources. Let me just highlight three of them – institutional differentiation, autonomy, and accountability, because they feature so prominently in national education policy debates.

Differentiation at an early age damages equity without improving quality. In systems that separate children early in secondary school, students’ performance by the age of 15 depends more than average on their socio-economic background. And there is no systematic benefit in terms of the average performance. This is an important policy lesson, perhaps less so for Japan but, but certainly for many of Europe’s education systems.

Private schooling is another form of institutional differentiation. Looking only at performance, students in private schools outperformed those in public schools in 20 countries. Only in three countries, public schools showed better results than private ones. But once you take account of the socio-economic background of students and schools the picture changes. Public schools then have an advantage of 12 score points over private schools. Private schools do of course offer an attractive alternative for parents looking to maximise the benefits for their children, including those benefits that come from the socio-economic level of schools’ intake, but more private schooling is not automatically associated with better overall outcomes. In Japan, we found no performance difference between public and private schools before the socio-economic background of students and schools was taken into account. But once we took these factors on board, students in public schools outperformed students in private schools. As a caveat, let me add here that we should, of course, also consider that there is significant variability in Japan’s private schools. While some perform very well, others are geared to educating poor performers who did not succeed in the public system.

On the second point: autonomy. Another feature that the best performers in PISA share is that they have devolved responsibility to the frontline. PISA suggests that countries giving more responsibility to schools tend to perform better. Giving schools more autonomy in formulating the budget, and letting them decide on allocations within the school tends to go hand in hand with better performance. This remains true even after accounting for socio-economic background and as other school and system level factors.

The third point is accountability, and improved accountability is a fundamental counterpart to greater school autonomy. Accountability has to do, among other factors, with how education systems use results from assessments. In many countries, this is controversially debated. Some see assessment results primarily as tools to reveal best practices and identify shared problems in order to encourage teachers and schools to improve learning environments. Others extend their purpose to use the results to support contestability of public services or foster market-mechanisms in the allocation of resources. And it is widely debated to what extent information on student performance should be made available to parents and the public at large. PISA shows that schools posting results publicly tend to perform better (even after accounting for all other school and socio-economic factors). This effect is strong across many countries. This suggests that external monitoring of standards, rather than relying mostly on schools and teachers to uphold them, can make a real difference to results.

PISA itself has encouraged countries not to take internally assessed education standards for granted. We can already see that the discipline provided by subjecting schools to external assessment with publicly visible results produces strong effects. Of course, such issues are very sensitive and need to be carefully addressed, particularly in a country like Japan that has just reintroduced a national assessment system this year, after 50 years of absence of external monitoring. However, the long-term perspective of improved transparency in schooling outcomes is important.

School systems continue to face the challenge how to improve equity without threatening quality. Given that resources are finite, the answer is not straightforward. Will reducing resources for socio-economically advantaged students and schools harm students’ performance more than improving resources for socio-economically disadvantaged students and schools would improve results? Even if this were not to lower the average score, it is possible that it would reduce the number of high-performing students, which in itself is undesirable.

PISA tells us that the most important factors for success they are not the ones most closely associated with finite material resources, such as the distribution of good teachers. Rather, what matters is how schools and the school system are run – for example, the amount of time that students spend in class and the extent to which schools are accountable for their results. Delivering such advantages to one student is not obviously at the expense of another. This, in itself, is an important conclusion from PISA. It underlines once more that quality, equity and coherence in educational standards are indeed achievable policy goals.

By way of conclusion, let me say that educational policies should provide the basis for our children to succeed. Successful learning experiences involve enabling environments at school, at home, everywhere. To get it right requires a deep understanding of how the system works. PISA is one of the tools at hand to improve performance, not only for policy-makers but for all of us striving to give our children the best education we can. But getting it right also requires courage to take the right measures and to reform when needed. The OECD stands ready to help – both with the analysis and the often difficult aspects of making reform happen.



Also AvailableEgalement disponible(s)