On the Achievement Gap

This is likely to be one of my more heretical posts. I think heretical well chosen here since the notion that all demographic groups of significant size should ideally produce equal achievement test results is so broadly accepted and so little debated in American education as to be the equivalent of canon law. But I will open that debate here with three related theses: (1) I'm not sure the gap can be eliminated, (2) I'm pretty sure it can't by any other than unjust means, and (3) I don't understand why anyone thinks this would be a good thing to achieve in the first place.

The achievement gap is not some new discovery; its investigation goes back to the Coleman Report of 1966, and I have read historians' estimates of it having been far more pronounced, at least between Americans of African ancestry compared with Americans of European ancestry, in 1900 than it was in 2000; and indeed, if you want to see a gap, visit those two continents consecutively, and then reflect on how far the gap has narrowed in this country. But pretty intense efforts to close it have been taking place in American schools for 40 years, with discouragingly little evidence of recent success. I am not here asserting that it cannot be eliminated, merely observing that it doesn't appear that it will happen soon.

Achievement gaps are apparent even before children enter compulsory schooling (first grade), and unless we are willing to have some theoretical totalitarian government control all human reproduction and seize all infants at birth for placement in orphanages, we are unlikely to eliminate those gaps prior to first grade. So the only way to eliminate a pre-established gap would be to (1) eliminate all private education, including homeschooling and tutoring, and then (2) make unequal distributions of public educational resources (such as time allowed for and quality of instruction) so as to favor those who enter school behind, to enable them to catch up. Now look at the next bright six-year-old you see from an education-loving family, and then look at that child's mother or father, who has perhaps been diligently reading in bed to the child and playing educational games with her or him since she or he was a toddler; do you really want to discriminate against such young human beings so early in life, for the sin of growing up in a family that values education highly? Or are you going to wait until later to start handicapping that child so as to achieve your social vision?   

My parents valued education highly; evidence of such was the subscription to National Geographic which appeared on the glass coffee table in our middle class living room, and I used to read it periodically as I was growing up. I remember reading about travelling on the Siberian Express when the Soviet Union was showing periodic glimmers of opening up to the rest of the world. During the Glasnost period of the 1980s, this journey sounded at first like one of the most romantic in the world; but through the account of the Western journalist undertaking the trip, we learned that it was one of the dreariest. Town after town, village after village, station after station, for 8000 miles (13,000 kilometres), through what in a non-totalitarian state should have been nine time zones, came and went, all exactly alike, the achievement of a Soviet state that prized equality, not just of opportunity but also of condition and outcome, before all else, at least in theory (less so in fact, we later learned).

Isn't the idea that there should be no achievement gap; that people of all races and ethnicities, of all backgrounds, linguistic, religious and otherwise, that all their family cultures should be rendered irrelevant and counteracted, and that all students should achieve exactly alike in every subject, a Soviet theory of education?  

The Best Schools in the World

A couple of years ago, when I was busy with refining the plans for One World School (an earlier conceptualization of One World Secondary School), I would have answered this question with model independent schools that were inspiring and still inspire me (the International School of Geneva, the United World College of South East Asia, others). Now, however, like others, I have shifted my focus away from individual great schools (one of which I am still hoping to start) to great school systems.

Professor Linda Darling-Hammond, in her The Flat World and Education, nominates Finland, Korea, and Singapore as having systems that America can learn from, and many other commentators (including, influentially, McKinsey & Company) have named these same three as having perhaps the best systems; but none of them would make my top three.

Any ranking (I don't believe in precise numerical rankings, but do believe we can learn from others) of educational systems will depend upon its criteria to generate output. What criteria should be used in ranking school systems?

Most commentators on education depend upon test scores (I promised in an earlier post to return to this theme), the most influential internationally being those of PISA (the OECD's Programme for International Student Assessment), but that's not where I would start. In the final analysis, I don't care as much about how many multiple choice questions on reading, mathematics and science that a student got right at the age of 15 as I do about how well educated that person will be as an adult, after having left school, with fifty or more years of life left to live. There is no precise way to measure how well educated someone is (a point to be borne in mind by those obsessed with measurement in a field where much of importance is exceedingly difficult to measure); the best statistical proxy is, I think, attainment, which generally refers to the number of years of schooling successfully completed. (For example, in my case, having studied [in school--I hope it's clear I've done a lot of studying since then] one year past a bachelor's degree, that number is 17.)

There are two statistics for attainment we should care about, one for students in school today and another for those who have already finished school. I care more about the future, and so give a bit more weight to the statistic for those youngsters currently in school than for those slightly older who are currently out of it, but both are important, since the former numbers are, if well reasoned, nonetheless speculative, and the latter, real numbers for recent graduates will continue to be important for a long time, with many decades worth of elections upcoming for people to vote well in or badly (or perhaps, even, to vote badly twice in a row!). I don't give weight to the attainment of older people, which hurts any U.S. ranking, since we were for a long time number one in the world in educational attainment but have slipped back in recent years as other countries have been improving their educational systems faster than we've been improving ours.

Adding together these figures for expected years of schooling successfully completed (easily available from UNESCO's Institute for Statistics) yields a set of figures useful for estimating educational attainment. But we also care about student achievement--how much a student learns in any given year of schooling--as a measure of the productivity of the school system. We might actually be able to multiply this productivity by the total anticipated attainment to give us a better measure of how well educated today's young people are going to be as they go through their lives, are productive (or not), and make important decisions. This is where the PISA scores become useful. I have multiplied the most recent median PISA score for each country available by the above estimate of anticipated attainment to estimate how well educated the average future citizen of each of these countries will be, not taking the numbers too seriously but becoming convinced that I was on to some useful estimates and methodologies. The top countries were New Zealand, Australia, and Korea. Only the latter is commonly mentioned as having a really top-ranking educational system, since most of these assessments depend more upon test scores than upon attainment. I will return to this theme later, including taking the time to debunk some myths about which top-scoring systems we should emulate.

But these figures only estimate how well educated the average future citizen will be. How about those at the extremes?

I believe the basic humanity of a country can be judged, in part, by how well it treats its least fortunate and neediest citizens. In addition, a law of classical economics, that of comparative advantage, teaches us to help our weakest citizens to become productive first if we want to maximize the gain for society as a whole.

The students most in need of special protection, it seems to me, are children with special needs, which every country will always have. A clear sign of treating such students well is making sure that they are gainfully (and, it is hoped, happily) employed when they become adults. I looked for employment statistics for people who had been students with special needs; these were very difficult to find, but I found some, along with other information on youth unemployment and the reputations of various countries' laws and programmes for those with special needs, and made some adjustments to my rough rankings, having at this point to give up any pretence of a formula.

Finally, those who are best off, educationally speaking, are those bright students commonly called gifted. They are inherently important, as all human beings are, but especially because of their impact in generating original ideas that often lead to jobs for everyone else. They also have special needs, and are often well served by programmes designed with their needs in mind, which allow them the opportunity to progress at their own pace; and they often desire to end up in the kind of prestigious, highly resourced universities that dominate the international ranking systems. The three most widely used international university ranking systems are those of the Times Higher Education Supplement, the Academic Ranking of World Universities, and the Top Universities ranking of Quacquarelli Symonds (QS). American universities dominate the top of all three. I made adjustments to my rough rankings, and pushed up the placement of the United States many steps on this basis. Other countries with relatively highly ranked universities also moved up, including England, Canada, Japan, and Switzerland.

This, I feel, is a pretty good way to approach assessing the overall quality of educational systems. Of course, others with more time or better resources could probably elaborate on it and improve it; but using this approach is useful, I believe, in helping to debunk various myths or shallow platitudes regarding which educational systems we all ought to be looking at as models (I expect to return to that topic at some point in the future).

The results? Of the three oft-cited countries mentioned earlier in this article, Korea did the best, but I found three other countries I would recommend before it. I have already mentioned two, Australia and New Zealand, which UNESCO says lead the world in expected years of schooling for those currently students; but I have one placed above them, and I will elaborate upon this country's system and various other implications of this study in another post, since this one has already gone on long enough. Therefore, the (surprise) winner, in this little competition to determine the world's best school system?

Norway.