without taking the time to watch the video (these things always take 10 minutes to give you info you could read in 30 sec) i checked on how the sampling was done and in a few different articles /documents found that australia tested well over the required sample in order to track results in all the different states and territories and ensure a useful sample of indigeneous students. that austria allowed the different provinces to decide their own sampling and have an internal debate going on about that, and that france handled the problem by giving the job of sample selection to US firm Westat. what you can read almost anywhere, but nowhere better than in this 2007 (thus written three PISA's ago) paper is that only 15-year-olds going to school full-time (though not necessarily college-bound) are sampled. which doesn't really skew the results as the aim is to measure the efficiency of teaching. scholarization figures are a completely different (and more important) issue for which statistics have been kept for much longer than PISA has existed. just like teachers tell their kids before an exam, the countries that cheat are really cheating themselves. the goal is not to be able to say WE'RE NUMBER ONE! but to see if you need to make extra efforts in teaching your kids, where these efforts need to be applied and how. and perhaps to put into light who the "good students" among nations are whose methods you could study for benchmarking. maybe the finnish model that has been lauded here in europe for two decades isn't all it's washed up to be. on the other hand i doubt anyone is going to be rushing to follow the south korean model where some kids start studying at 6am, finish after midnight, and sleep with english language lessons in their earbuds. in fact that need to succeed more psychotic than simply neurotic which characterizes korea ,and to a lesser extent china and japan, makes one suspect they are the most likely to fudge their PISA scores, and section 2.5 of the paper i linked to suggest that very strongly. and some of the poor performers seem to be cheating too, so their case might even be worse than shown. though the entire paper is very interesting and gives a much better understanding of PISA than MSNBC can or even should provide, for those who wish to cut to the chase, i refer you to section 5 (sorry, it's unpastable).