The 2015 PISA Results: What Do They Mean?

pisa2015

Cross-posted at Education Week.

The PISA 2015 results show no meaningful change in the performance of U.S. students since 2012 in science and reading and a decline in mathematics.  Yes, there is a story here about mathematics, but that is not all there is here.  This is what I make of the data the OECD shared in the report released on Tuesday morning.

A Sputnik Moment?  

Whereas mainland China was represented in the last PISA assessment only by Shanghai, it includes this time three additional Chinese provinces.  Many observers three years ago cautioned the American public and policy makers not to draw any conclusions about China as a whole from Shanghai’s top ranking in the last administration of the PISA survey, assuming that Shanghai was far ahead of the other provinces in elementary and secondary education. But that is not what the data show.  The scores for all four Chinese provinces are merged in the 2015 survey.  The average score for all four provinces places China near the very top of the global rankings in mathematics and science.  Their combined scores would have placed China in the top ten were it not for their scores in reading, which were about the same as the average U.S. score for reading. The U.S. scores higher in reading than in math and science. Twenty-four of the 70 participating jurisdictions had higher average scores than the U.S. in science, 38 had higher average scores in mathematics.

The four provinces in China that participated have a combined population of 226 million people, compared to the U.S. population of 319 million.  If they were a separate country, they would be the fifth largest country in the world.  But the high school students in these provinces are grade levels ahead of their American counterparts in the core STEM subjects, which gives the Chinese an enormous advantage in science, mathematics and engineering, which are widely expected to be the most important drivers of both economic and military prowess in the years ahead.

The average GDP per capita in these four Chinese provinces is $14,647; the average GDP per capita in the U.S. is $53,041. Thus, China is still a developing country, and, though wages have been rising quickly in recent years, still charges significantly less for its labor than the advanced industrial countries. What is striking about this PISA report is that it shows clearly that China has both a low-wage structure and, at the same time, one of the best education systems in the world, which makes it a formidable competitor in both the economic and military realms.

In 1957, the launch of Sputnik by the Soviet Union forced the United States to respond with an enormous national investment in science and mathematics education.  This challenge, from China, seems to this observer to be no less important.

pisa2015infographv2Immigration and Education   

Immigration may be the single most important issue in global politics today.  Note that Canada continues to perform among the top ten countries on the PISA league tables.  What’s the connection to education?  According to OECD data, second generation immigrants in Canada are the best educated in the OECD.  Second generation immigrants in the United States are the worst educated.  Opposition to immigration may have tipped the recent presidential election to Donald Trump in the United States, but, in Canada, the polls show that Canadians are very proud of their immigration policy and of their openness to immigrants.  It is very likely that the superior performance of second generation students in Canadian schools is due to two things: an immigration policy that is biased toward highly skilled immigrants, rather than extended family members, as in the United States, on the one hand, and, on the other, a very strong commitment to multicultural programming in their public schools.  It is worth considering the possibility that this combination of immigration policy and education policy might hold the key to the future of our democracies, at least in the short term.  My thanks to my good friend former U.S. Labor Secretary Ray Marshall for bringing these ideas to my attention.

What’s Responsible for the United States’ Poor Performance in Mathematics?

The shortest answer is: many things.  But there are two things that I think are most worth thinking hard about.

First, the United States has no comprehensive teacher recruitment strategy aimed at high-performing high school students and as a result gets most of its teachers from the bottom half of the distribution of high school graduates going to college.  Many go to colleges with very low entrance requirements; they will take virtually anyone who has a high school diploma. Most have to take the equivalent of College Math or College Algebra in college, but we know that is essentially a course with the content of Algebra I, a middle school course.  This is a very low bar.  In contrast, the top-performing countries are moving toward restricting teacher education programs to those in research universities, which means that young people going into teaching are recruited not from the bottom half of the young people going to college, but the top half.  So the future teachers in the top-performing countries are years ahead of our young people when they come out of high school and even further ahead when they come out of college and go into teaching.

Second, many of the top performers require their elementary school teachers to specialize either in math and science or in their native language and social studies.  When they are in teachers college, they are required to at least minor—and in some cases major—in the subjects they are going to teach.  This is crucially important.  Our elementary school teachers are expected to teach all the subjects in the curriculum.  Many did not like math when they were in school and took as little of it as possible.  They often learned the key math procedures, but do not understand why they work.  It is hardly surprising that many of their students end up not liking math and doing poorly in it.  If we were to require our elementary school teachers to specialize in the same way the top performers do and we were to ask our teachers colleges to require future elementary school teachers to at least minor in the subjects they were going to teach, we would fill our elementary schools with teachers of mathematics and science who loved math and science, were very good at it and good at teaching it.  Nothing, in my opinion, could make a bigger difference in student performance in these crucially important subjects.

Should Other American States be Copying Massachusetts?     

One piece of good news for the U.S. is the performance of students in Massachusetts. The state continued to perform well against our international competitors and improved considerably since the last PISA administration. Only students in Singapore performed significantly better than Massachusetts students in science. In reading, Singapore and Massachusetts tied for the top spot. The state is clearly among the world’s top performers in reading and science.   But there are three important caveats here.

First, Massachusetts is one of the wealthiest states in the United States, with GDP per capita in 2015 of almost $70,000.  The four provinces in China that participated in the 2015 PISA survey, which performed just as well as Massachusetts in science and better in mathematics, have, as we saw above, an average GDP per capita of less than $15,000.

Consider also that Massachusetts ranks number one in the country in the percentage of adults with a bachelor’s degree and number one in the percentage with an advanced degree.  The education of parents is the single most important predictor of student performance.  This combination of wealth and parental levels of education should lead us to expect that Massachusetts would be a world leader in student performance.

Third, though Massachusetts does very, very well in reading and science, it does not do so well in mathematics.  Eleven countries, including the four provinces in China, outperformed the state in that subject.

American states have a lot to learn from Massachusetts, which not only does very well on PISA but also outperforms all other states on NAEP by a wide margin.  But states that are not as wealthy and which do not have parents with so much education and that want to do better on mathematics and greatly improve outcomes for disadvantaged children should look elsewhere in the world as well, because the PISA league tables include many countries that have done as well or better for less money with a smaller supply of highly educated parents.