First of all, do we know why other countries are outperforming us? Especially countries with similar socioeconomic and cultural norms, like Canada? Second, wh

First of all, do we know why other countries are outperforming us? Especially countries with similar socioeconomic and cultural norms, like Canada? Second, why do we insist that a single number is going to usefully assess the many skills being learned? Or that any single metric, however informative, should be applied across all the many subgroups that make up a large population like the US? Why the hell don't we stratify students and policies and regions and then compare only across pairs across strata? Only in that way can you learn what works and what don't. We should be breaking down these numbers by student socioeconomic group, by state, by county, by school district, by education policy, by gender, so that we can compare results within the US to see how different policies on similar student groups turns out, and similar policies on different student groups. What works and what doesn't? After 70(?) years of standard tests, it seems we still have no idea. We're never going to learn anything useful from any kind of testing if we don't 1) partition the parameter space better, and 2) ask simple (actionable) questions of the results. Likewise, we need to break ourselves of the insistence that a single education policy is best for all students everywhere in the US. Geez. Let different states propose their own experimental design and then assess the results of their different strokes. Supposedly that's what governance at the state level was for, ever since 1789. Comparing the US to other countries using a single metric sounds like the least useful way to improve US students.