The Smartest Kids in the World

by Amanda Ripley

  PISA attracted critics: For one critique of PISA, see Schneider, Education Next. My own conclusion is that these critics raised important points, particularly regarding the challenges of extrapolating causation from PISA data. Schleicher and his colleagues at the OECD have imperfect information, and their own biases, of course. Still, on balance, the data from PISA represents an important portal into a large, complex problem. It seems better to attempt to understand what differentiates education systems (with caution) than to abstain from the conversation altogether.

  “A TV reporter showed this graph:” OECD, Take the Test.

  Flu-shot notice: OECD, Take the Test.

  Vitamin C: OECD, PISA Released Items.

  “Good job!”: The PISA folks declined to translate my performance into a precise numerical score since a country’s mean score is normally derived from the aggregate score of all the kids who took the test. There are different versions of the test booklet given to different students to come up with a balanced sample. So, I can’t say with precision how I did compared to all kids in Finland or Korea. However, it seems safe to assume that we inhabit the same league, since I got all but one question right. I am of course, much older than PISA test-takers, so this does not mean much. But, anecdotally speaking, I can tell you that there was nothing on the test that I wouldn’t want my own child to know and be able to do by the age of fifteen. PISA is many things, but it is not rocket science.

  “The hallmark of American education:” Scott, “Testimony by Professor Joan Wallach Scott.”

  U.S. teenagers ranked twenty-sixth: OECD, PISA 2009 Results (Vol. I). Note that students from Shanghai, which is part of China, earned, on average, the highest score in the world on PISA in 2009. I did not include Shanghai in my rankings for this book because Shanghai is not a country and not representative of China as a whole. (Millions of children in China still lack access to a basic education, despite exaggerated media accounts of China’s educational dominance.) If I had included Shanghai (and Hong Kong), the United States would rank lower in every subject.

  Data from PISA can be most easily accessed using the PISA International Data Explorer, located at http://nces.ed.gov/surveys/pisa/idepisa/.

  Second in the world: OECD, PISA 2009 Results (Vol. IV), Table IV.3.21b. There are many ways to compare spending on education, all of them flawed. After looking at the options, it seemed most useful and fair to rely on OECD data for cumulative expenditures by educational institutions per student aged six to fifteen. The figures are in equivalent U.S. dollars, converted using purchasing power parity.

  One downside of this figure is that it does not include all of high school (or prekindergarten). Since the PISA test score data is based on fifteen-year-olds, these figures do correspond to the most relevant years for our purposes.

  A bigger downside is that these numbers do not include families’ private spending on tutoring and other educational supplements (although the figures do include private school expenditures in most countries, including the United States). As discussed in more detail in the portions of the book focused on Korea, that spending can be very high in Asian countries in particular. But, in all cases, most education spending flows through the school systems, which is where these numbers come from.

  One-to-one match: Robelen, “Study Links Rise in Test Scores to Nations’ Output,” and OECD, The High Cost of Low Educational Performance.

  One to two trillion dollars: McKinsey & Company, Economic Impact.

  A better predictor: The predictive power of PISA was analyzed in a longitudinal study of thirty thousand Canadian teenagers who took the test in 2000. OECD, Pathways to Success.

  chapter 2: leaving

  Pretty Boy Floyd: Ingram, “Family Plot.”

  Officially classified as poor: Poverty rates for Sallisaw School District from the Bureau of the Census, American Community Survey, 2005-9 Summary Tables, generated using American FactFinder.

  On the state test: In 2009, when Kim was finishing eighth grade, six out of ten of her Sallisaw classmates scored proficient or better on Oklahoma’s standardized test. Oklahoma State Department of Education, “Sallisaw Public School No Child Left Behind Act Annual Report Card 2008-2009.”

  But that test was notoriously easy: Peterson and Lastra-Anadón, “State Standards Rise in Reading, Fall in Math.”

  On a more serious test: I’m talking here about the National Assessment of Educational Progress (NAEP), which is the largest nationally representative test continually administered in the United States. In Oklahoma and other states, the sample does not include enough students to offer district-level results. But since Sallisaw’s scores did not differ dramatically from statewide averages on other tests, it is fairly safe to assume that Sallisaw’s NAEP results would be comparable to the statewide results (if such data existed).

  To put Kim’s experience in context, I referred here to the 2009 NAEP results. That year, 23 percent of Oklahoma eighth graders achieved proficient or advanced scores in math. In 2011, the number rose slightly to 27 percent, which was still below the national average of 34 percent. U.S. Department of Education, National Assessment of Educational Progress.

  If states were countries: Oklahoma’s world ranking comes from the 2011 report, Globally Challenged (Peterson et al.), which creates a statistical crosswalk between PISA and NAEP data in order to rank states’ performance relative to countries. The figure on pages 8-9 of the report shows that Oklahoma ranks eighty-first in the list of countries and states (not including territories).

  SAT scores: In critical reading, Kim did better than 40 percent of Oklahoma’s college-bound seniors. But she performed better than a whopping 69 percent of seniors nationwide. Why such a big difference? It turns out that only about 6 percent of Oklahoma graduates took the SAT (compared to 48 percent nationally). So, Oklahoma’s average SAT scores were higher than they were for the nation. Most Oklahoma students took the ACT instead.

  Meanwhile, Kim did much worse in math, as she’d expected. In math, she scored better than just 5 percent of Oklahoma SAT-takers and 15 percent of students nationwide. In writing, she did slightly better, scoring higher than 14 percent of the Oklahoma seniors and 34 percent of all U.S. seniors.

  At a high level, Kim’s strengths and weaknesses were not all that different from that of American students nationwide. She excelled in reading and tanked in math.

  More than doubled: The increase in Oklahoma education spending is in constant dollars and comes from U.S. Department of Education statistics on per pupil expenditures.

  Teachers’ aides: During the 1986 to 1987 school year (earliest available data), Oklahoma had 3,825 instructional aides; by the 2010 to 2011 school year, the state employed 8,362 aides. To be fair, the student population rose 11 percent during the same time period. But the ratio of students-to-aides still went from 155:1 to 79:1. Meanwhile, Oklahoma’s student-to-teacher ratio went from 17:1 to 16:1 over the same time period. Figures compiled through Common Core of Data “Build a Table” site, National Center for Education Statistics (http://nces.ed.gov/ccd/bat/).

  Lowered the student-to-teacher ratio: Working with the Oklahoma State Department of Education, the earliest data I could find was student-to-teacher ratios going back to the 1976-1977 school year. Since then, the ratio had gone from 20.21 students per teacher to a low of 15.01 in the 2000-2001 school year. Since then, the ratio had crept up slightly to 16.11 for the 2011-2012 school year.

  Over half the state budget: In Oklahoma’s FY11 budget, education rang in at $3.6 billion—out of a total $6.7 billion in spending.

  End-of-school test: Across the developed world, students in school systems that require standards-based external exams perform over sixteen points higher on average on the PISA test than those in schools that do not require such tests. These kinds of tests exist in Finland, Korea, and Poland, among other countries. They also exist in some U.S. states, but are not generally very rigorous.

  For the tortured history of this test in Oklahoma, see
Hinton, “Legislature Junks High School Grad Test Requirement”; Killackey and Hinton, “Outlook Uncertain for Literacy Passport”; Hinton, “Governor to Require ‘Literacy Passports’ ”; and Price and Hoberock, “Legislative Roundup.”

  For more on how end-of-school tests impact performance around the world, see OECD, Strong Performers and Successful Reformers in Education, pgs. 49-50 and 243. For me, the most memorable part of that section was this:

  “In the United States, high school students may be led to believe that the outcome is the same whether they take easy courses and get Ds in them or take tough courses and get As. Either way, they might think, they can get into the local community college and get on with their lives. Contrast this with a student of the same age in Toyota City, Japan, who wants to work on the line at a Toyota plant. That student knows that she must get good grades in tough subjects and earn the recommendation of her principal, so she takes those tough courses and works hard in school. . . . One of the most striking features of the American education system, in contrast with the education systems of the most successful countries, is its failure to provide strong incentives to the average student to work hard in school.”

  “Kids have a really good detector”: Killackey, “State Education Secretary Urges High School Graduation Test.”

  “Lost generation”: Archer, “Bill Would Lift Required Graduation Testing.”

  530 superintendents: By comparison, Finland has 399 superintendents for the entire country—a far larger land area with over 1 million more residents than Oklahoma. See Kanervio, “Challenges in Educational Leadership in Finnish Municipalities.”

  One of the top earners: Sallisaw’s median household income between 2006 and 2010 was $30,229, according to the Bureau of the Census. Superintendent salaries come from the Oklahoma State Department of Education.

  Hardly unusual: Usually, no one pays attention to the return on education spending—despite it being one of the largest items in any state budget. In a dramatic break with precedent, Ulrich Boser did a study in 2011 on the productivity of American school districts. This analysis showed huge variation from place to place, with the highest-spending districts being among the least efficient.

  Seen through this lens, Sallisaw’s education results, while unimpressive in absolute terms, were a relative bargain, given the small amount of money spent per student. For more, check out the interactive map that accompanied that report, available at http://www.americanprogress.org/.

  More involved parents: Americans’ views on parental involvement (and all things) vary depending on how you ask the question. But it seems fair to say that parental involvement is a widespread concern. A 2010 Time magazine poll of one thousand U.S. adults included the following question (percentage agreeing in parentheses):

  “What do you think would improve student achievement the most?”

  More involved parents (52 percent)

  More effective teachers (24 percent)

  Student rewards (6 percent)

  A longer school day (6 percent)

  More time on test prep (6 percent)

  No answer/don’t know (6 percent)

  They were in fact showing up: MetLife, The MetLife Survey of the American Teacher.

  Nine out of ten parents: Herrold and O’Donnell, Parent and Family Involvement in Education, 2006–07 School Year.

  Nearly one in four: Oklahoma State Department of Education, Sallisaw Public School No Child Left Behind Act Annual Report Card 2010-2011.

  Remedial classes: Oklahoma High School Indicators Project, Remediation Report, Fall 2010.

  For Sallisaw’s Class of 2010 alumni entering college as freshmen, the remediation rate was 55 percent. The statewide remediation rate for Oklahoma high school graduates attending state colleges and universities that fall was 38 percent.

  Nationwide data for the 2010 class was not yet available, and comparisons from one locality to the entire country are always complicated. But as one point of reference, about 36 percent of first-year undergraduates nationwide reported having taken a remedial course in 2007 to 2008. At public two-year institutions, about 42 percent said they had taken a remedial course. See Aud et al., The Condition of Education 2011, Indicator 22: Remedial Coursetaking.

  One out of two Oklahoma university students: Denhart and Matgouranis, Oklahoma Higher Education.

  One or two thousand American high school students: Poehlman, 2011-2012 International Youth Exchange Statistics.

  Teenagers in Finland did less homework: OECD, Mathematics Teaching and Learning Strategies in PISA.

  Not even tiny New Hampshire: The Class of 2011 in New Hampshire performed at about the same level of students in Hungary and France. New Hampshire’s teenagers were outperformed by their cohorts in eighteen other places including Canada, Japan, New Zealand, and Finland. See Peterson et al. Globally Challenged.

  Among the least challenging in the nation: Lerner et al., The State of State Science Standards: Oklahoma. The state’s science standards received an F in this report. The document to which this report refers (the one that does not mention evolution) is the Priority Academic Student Skills (PASS) standards for science, which were updated in 2011

  Door and window factory: Window & Door, “Therma-Tru to Close Oklahoma Manufacturing Facility.”

  Blue Ribbon Downs: Adcock, “Sallisaw: A Blue Town.”

  chapter 3: the pressure cooker

  Busan: Busan used to be known as Pusan. They are the same place. The spelling was officially changed in 2000.

  “Love stick”: The Korean government had outlawed corporal punishment shortly before Eric arrived. It was a controversial decision, one with which Eric’s principal and the local superintendent did not entirely agree. Teachers complained, worried they would have even less control of their sleepy students if they couldn’t hit them.

  But certain kinds of physical punishments were still allowed—like taps with the so-called love sticks or even forcing kids to stay in a push-up position for twenty minutes or to run laps around a field. Occasionally, more Draconian habits resurfaced.

  One afternoon, an older man that Eric hadn’t seen before came into his classroom and called up three boys who had been disruptive earlier in the day. He lined them up at the front of the classroom and told them to present their hands, palms down. Then he rapped their knuckles with a ruler, one boy at a time, said Eric, who saw the boys flinch. Then they slouched back to their seats.

  Worldwide, corporal punishment in schools is outlawed in about one hundred nations, including Afghanistan, China, Finland, Germany, Sweden, the United Kingdom, and Poland, according to the Global Initiative to End All Corporal Punishment of Children. So far, the United States is not one of those countries, although thirty-three states do have bans in place.

  More information on international comparisons can be found here, including interviews with children themselves about the humiliation caused by physical punishment at school: http://www.endcorporalpunishment.org/.

  Only 2 percent of seniors: Korean Education Ministry officials. This figure refers to the portion of all two- and four-year college students admitted in 2012 to Korea’s so-called SKY Universities: Seoul National University, Korea University, and Yonsei University.

  American tests were not high stakes for students: In a 2011 survey of 10,000 public-school teachers by Scholastic and the Gates Foundation, only 45 percent of teachers said their students took standardized tests seriously. See Scholastic and the Bill & Melinda Gates Foundation, Primary Sources: 2012 – America’s Teachers on the Teaching Profession.

  A small fraction of teachers: The stakes were higher for teachers in some places but, as of 2012, the vast majority of U.S. teachers were not evaluated based on test scores, despite widespread anxiety over such practices. In some places—like Washington, D.C., and Memphis—a minority of teachers had started to be reviewed based in part on the growth in their students’ test scores over time (compared to other students who had started the year at a similar level o
f performance). The rest of their evaluations were based on other things, including classroom observations. In 2011, about 6 percent of teachers in D.C. and fewer than 2 percent of teachers in Memphis lost their jobs after receiving exceptionally poor evaluations, according to my interviews with education officials in both places in 2012.

  “You Americans see a bright side”: Author interview with Korean Minister Lee Ju-ho on June 9, 2011, in Seoul.

  In tenth-century Korea: Lee, “The Best of Intentions,” 23.

  New words had to be coined: Sorensen, “Success and Education in South Korea.”

  Student-teacher ratio of 59:1:: Cavanagh, “Out-of-School Classes Provide Edge.” Korea’s current ratio is closer to 28:1.

  Only a third of Korean kids: Seth, Education Fever.

  Nobody got accepted because he was good at sports: Lee, “The Best of Intentions.”

  About 40,000 percent: GDP figures from Korean Culture and Information Service, Facts about Korea, 87.

  Education acted like an antipoverty vaccine in Korea: Kim, “Consequences of Higher Educational Expansion in Korea.”

  Dropout rates: The dropout rate for Minnetonka High School comes from the Minnesota Department of Education online Data Center, accessed in November 2012. The dropout rate for Namsan, Eric’s Korean high school, comes from my interview with the principal in June 2011.

  To be fair, Namsan only admits 70 percent of the students who apply, while Minnetonka must take all students in the zoned jurisdiction. However, even with its selectivity, Namsan has a more impoverished student body, with about 17 percent qualifying for a full tuition subsidy due to their parents’ low-income levels. (This formula is complex, but in general, qualifying families must earn less than $20,000 or so.) By contrast, at Minnetonka, only 8 percent of students qualify for free or reduced price lunch under the federal guidelines (which comes out to about $29,000 or less for a family of four). While these are two totally different measures, they give us some rough indication of the relative affluence of the Minnetonka student body.