Maybe the time has come to acknowledge that the changes we need can't be measured by the tools we've been using and that replicating by mandate in an "industry" like education has limits. Those who designed the instruments that measure us as failures admit they never intended them to be used for high-stakes purposes. There will be some schools and kids whose scores go up, some down, and some both or neither. But the whole demographic will remain unchanged unless we ignore cheating or what it means to be truly educated.
Since I began my crusade against this kind of testing in the late 1960s I've been told: (1) Something is better than nothing, and (2) we are developing better measurement tools. Bah, humbug. After 45 years of hearing this—literally—I don't believe it. We know there are both better ways of administering standardized tests that will do less damage: sampling. We know plenty of other ways that we can look at kids' work to assess their individual progress. I can name several-hundred public and private schools that have done so for decades. But the particulars of their solutions are not replicable because measures of success rest on "values" that need to be openly negotiated, not mandated. That's an advantage private schools have that could be made public