Just as the U.S. is about to move away from over testing its students, PISA’s Andreas Schleicher says American students are not really over-tested: “The U.S. is not a country of heavy testing,” said Schleicher in a column published in the Hechingner Report.

Schleicher drew the conclusion based on PISA 2009 student survey data, which was not released publicly. Schleicher claims to “trust the reports of students on what actually happens in the classroom more than the claims of many experts” in his blog post that argues that U.S. is not a country of heavy testing. One wonders why he has not released that data.

However the publically available PISA report contains standardized testing data reported by school principals. Given the lack of access to the student data, reports by school principals are the best source we have. I’d like to think that school principals know as well as students if standardized tests are given in schools. Moreover, comparing the few data points Schleicher reveals in his blog suggests that the perception of students is not far off from that of principals. Based on the principal reports about standardized testing, I found Schleicher’s statement misleading, to say the least.

Is U.S. a country of heavy testing?

First, what is considered heavy testing? Schleicher seems to think at least once a month is not heavy enough: “In many countries there is a test going on every month,” while in the U.S., “only 2% of students said they took standardized tests at least once a month.”

Is 3 to 5 times a year heavy? That is about one standardized test every 2 months of the school year. About 40% of American students take standardized tests 3 to 5 times a year. How about 1 to 2 a year? That’s almost every American student (97%).

Schleicher’s conclusion is based on international comparisons. He highlighted two nations that have more standardized tests than the United States but neglected to mention that there many countries that have fewer standardized tests. For example, according to the 2009 PISA principal survey, 76.4% of students in Slovenia, 73.2% in Belgium, 71.1% in Spain, 67.6% in Austria, and 60% in Germany “never” had standardized testing. Japan (34.6%), U. K. (32.5%), Australia (29.9%), and Ireland (35.0%) also have more students never given standardized tests. Only 2.5% of students in the U.S. never had a standardized test. Only three OECD nations—Korea (2.1%), Luxembourg (1.0%), and Finland (1.5%) – reported lower percentage of students who never take standardized tests.

While the U.S. does not have the largest percentage of students given standardized testing “at least once a month,” it is one of the countries with the largest proportion of students experiencing standardized testing “1 to 5 times a year.” With over 95% of students who attend schools whose principals reported giving standardized testing 1 to 5 times a year, the U.S. is only after four  (Korea:96.5%, Finland:96.3%, Luxembourg: 96.4%, Hong Kong: 98.4%) out of the nearly 70 countries participated in PISA had a slightly larger percentage of students experiencing standardized testing 1 to 5 times a year. Even Shanghai, Singapore, and Chinese Taipai reported fewer students taking standardized tests with this frequency.

More importantly, does standardized testing help improve education quality?

PISA scores do not necessarily reflect the quality of education, but even use PISA scores as a measure and according to PISA’s own report: “there is no measurable relationship between the prevalence of standardised tests and the performance of school systems, and that also holds for most countries at the school level” (p. 46). The two countries Schleicher highlighted for testing students much more than the U.S., namely the Netherlands and Israel, ranked 10^th and 37^th in 2009 Reading scores. Israel is much lower than the U.S. and the Netherlands was below nations that test less.

Moreover, according to PISA’s own report, the use of standardized testing has negative relationship with PISA’s reading scores in a score of countries, including the United States (see figure below).

(Captured from PISA 2009 Results: What Makes a School Successful? RESOURCES, POLICIES AND PRACTICES VOLUME IV, p. 48).

I hope Schleicher is aware of this finding. If he knows that standardized testing is negatively correlated with reading performance in the U.S., why does he encourage more standardized testing?