Monday, November 1, 2010

So, how DID Finland do it?

I’ve written before about how we should emulate the way Finland pulled itself up, from a stultified, centrally controlled, Soviet-Era educational system to one in which all children perform spectacularly on valid international tests of student achievement. Here, I’ll share what they did and how.

Their success is real

First, let’s establish that their success is real. The Programme for International Student Assessment (PISA) tests are sponsored by the 30-member Organization for Economic Cooperation and Development (OECD), although even more non-OECD nations now participate. These tests in reading, mathematics, and science were administered to 15-year-olds in 41–65 countries in 2000, 2003, 2006, and 2009 (results expected in December 2010). Each testing cycle focuses on one of the subject areas, with minor assessments of the others. The international average given is for the OECD (that is, developed) nations.

PISA tests are the gold standard of achievement testing. In order to do well on them, students cannot simply recall factual knowledge; they must extend what they know into unfamiliar settings to solve problems. Students who do well on them have also done well in life — the ultimate test of the success of schooling. In the 2006 cycle, Finland’s students were the top performers in math and science and number two in reading. (For comparison, U.S. students ranked 25th of 30 for math, 21st of 30 for science, and last in reading.) In the 2003 cycle, Finland was at the top in all three categories.

Student success is consistent across all schools

Differences in student performance within schools are generally taken as reflecting natural variation in inborn talent. Variation between schools, however, is an indicator of social inequality — in most places, there is some segregation by socio-economic class among schools, and the wealthier students’ schools have more resources. PISA offers sophisticated statistical analysis of performance controlling for class. Analysis of data shows that, unlike ours, many nations’ students achieve not only high average scores but also scores that do not vary much by socio-economic status: they have realized the ideal of uniformly high student achievement that No Child Left Behind (NCLB) set for us. Again, Finland is tops when it comes to equality of opportunity and performance. And, please note that Finland is no longer homogeneous: recent immigrants, mostly from poorer countries, speak more than 60 languages. In some urban schools, half of the students are from immigrant families whose native tongue is not Finnish.

Here are some examples of results, comparing Finland and the U.S. In the 2006 Science test, mean performance was 563 for Finland (the top performer that year) and 489 for the U.S. (and 500 for the average of OECD nations, on a 1000-point scale). An indicator of unequal opportunity leading to unequal results is the “between-school variance explained by the index of economic, social, and cultural status of students and schools.” This variance was 19% of the total variance in tested countries for the USA but only 1% for Finland. In the 2003 Mathematics test, mean performance was 544 for Finland and 483 for the USA (and 500 for the OECD average). The between-school variance explained by these factors was, again, 19% for the USA and less than 1% for Finland. Even without any controlling for these variables, performance varies by less than 5% overall among Finnish schools. Nearly ALL their students do quite well.

This success does not cost a fortune

I am always wary of examples of success in schools that spend so much more than average as to be irrelevant in the real world of shrinking funding for schools and other civic priorities. For example, the SEED school in Washington, DC, that was put on a pedestal in the recent documentary “Waiting for Superman,” is a residential school that spends $35,000 per student per year — an unattainable ideal for the vast majority of students.

But Finland has transformed its public schools to offer equal opportunity and to achieve consistently excellent results without spending a fortune. OECD figures show that Finland’s total expenditures on educational institutions, expressed as a percentage of Gross National Product, have actually declined over recent decades. Pasi Sahlberg, a Senior Education Specialist at the World Bank and an Adjunct Professor at the University of Helsinki, produced an analysis that shows this, in “Education policies for raising student learning: the Finnish approach.” He found only a weak correlation between student performance and “cumulative expenditures per student from age 6–15.” This figure for Finland was at the OECD average, whereas U.S. spending was about 20% above that average. Note, of course, that Finland’s comprehensive social welfare system provides much more support to less wealthy families than does the American system.

AN ASIDE: The dataset and tools for manipulating it at are truly amazing. I encourage you to look into them yourself. When news media do reports on PISA results, as they surely will in December, their summaries and assertions may not be valid reflections of the data, but you can judge for yourself. PISA provides a Data Analysis Manual. You can even “Take the Test” of sample questions yourself to see how much more is demanded of students today! Or, use the Dept. of Education’s International Data Explorer interface at If you feel under-prepared to make statistical judgments, a report on US performance in 2006 from the National Center for Education Statistics can be found at

* * *

To summarize, results on this well-respected test series demonstrate not only that Finland’s students are top performers, but also that their performance varies much less by economic, cultural, and social status than in our country. And they do not “throw money” at education to get these results.

So, what do they do differently that we could learn from?

Sahlberg, cited above, notes that Finland has not adopted the “market-oriented reform strategies” that have overtaken much of the rest of the developed world, including the U.S.: “Consequential accountability accompanied by high-stakes testing and externally determined learning standards has not been part of Finnish education policies.” Instead, cultivating leadership, emphasizing teaching and learning, encouraging creativity, ensuring equity, enabling and trusting teacher professionalism, and “intelligent accountability” have wrought this “miracle.” It was accomplished beginning in the 1990s, when Finland was undergoing “a severe economic decline characterized by a major banking crisis.” Sound familiar? Yet, ten years into the process, Finland began to be ranked by the World Economic Forum as one of the world’s most competitive economies, a distinction it has maintained since then. It is also ranked as one of the least corrupt nations — something we may have difficulty emulating.

We do not have the culture of trust, the respect for public institutions, and the shared values of honesty and equity that Finns enjoy, and that enabled their progress in transforming their educational culture. But, clearly, this values-based approach has worked for them — and reform efforts based on competition and coercion have just as clearly failed here, as elsewhere. Student performance on valid tests such as the PISA ones has not improved, despite the nationwide push that began with NCLB and continues under the Obama administration education policies. Collateral damage from these policies continues to increase, as measured by student drop-out rates, cheating on high-stakes tests, rock-bottom teacher morale and steady defections from the profession, elimination of flexibility and creativity in teaching methods, and a narrowing of curriculum down to the most basic of “core” subjects. What we are doing is not working. What do we have to lose by trying an alternative path?

The first step, it seems to me, is a national conversation about shared beliefs and values. No Mission Statement can make any sense until we establish that context first. Do we truly believe that all children can learn, given enough time and the right support? Do we truly believe that all children should have the same opportunity to succeed, which opportunity must be embodied in the uniformly excellent teachers, facilities, programs, and resources available to them? Do we truly believe that a well-trained and experienced teacher actually has some professional expertise worthy of our respect? Do we truly believe that a well-educated and superbly functioning adult needs more than the “three R’s” to get there? Do we even believe that our society will be better off if all of our adults are allowed and enabled to meet those standards? These are the kinds of beliefs and ideals that underlay the Finnish success.

The policies that flow from such consensus on values would be very different from what we have now. It might seem impossible to reach national consensus on anything, at the moment. But it most assuredly will never happen unless we begin to have these conversations. And we could make a start on the process at the state level, meanwhile. Suppose we subsidized higher education for teachers, to include the master’s level required for all Finnish teachers, in exchange for their teaching in Michigan’s schools for a certain period of time? We could make the payback period shorter for work in our needier schools and communities. This subsidy should be offered on a competitive basis, to attract our best and brightest to this important profession. And suppose we got back to reducing the inequity in per-child funding in our public schools from one district to the next? Suppose our state mandates for days or hours in the school year made allowance for time for teachers to collaborate and to develop their skills on an ongoing basis? Suppose we lobby our political representatives to allow us to divert some of the absolute fortune we spend on mandated testing of nearly all students in most subjects every year to targeted testing with results that are available to teachers in time to be used to actually guide instruction?

Are these ideas really so crazy? Does it make more sense to keep doing what is not working?