The critique is fair. I suspect the original 'test kids in Kindergarten and fast track the best performing for the next 12 years' has its roots in the Sputnik m

The critique is fair. I suspect the original 'test kids in Kindergarten and fast track the best performing for the next 12 years' has its roots in the Sputnik moment and the need to rapidly create a high performing STEM population. Accuracy wasn't necessarily a primary goal. That being said, not sure what the alternative is. On one hand, challenging students to the limit of their abilities and bump them higher as soon as they exhaust a level is a laudable goal. Alas, as long as the educational pedigree is used in competitive scenarios (grant, internships, jobs), this will simply create an inflationary pressure to bump everyone to the top level and degrade the top level quality to the lowest common denominator. We are missing a objective bar, for example a hard standardized exam to gate college admissions. Such an exam becomes an incentive for developing strong programs. Time to take a clue from China's gaokao. Alas this requires that we come to terms that not all schools, not all programs, not all demographics, not all kids are equal, which doesn't seem to be something XXI century America is willing to publicly accept.