Wayne Bishop recently made me aware of the unfortunately completely one-sided discussion of US mathematics education at the recent Aspen Ideas Festival.
David Leonhardt is Washington Bureau Chief for the New York Times, won a Pulitzer Prize for his reporting on economic issues, and majored in applied mathematics as an undergraduate at Yale. Mr. Leonhardt chaired the panel, “Deep Dive: Is Math Important?” an “event” in the program track “The Beauty of Mathematics”. Other program track events included individual lectures from each of the panelists.
Mathematicians might consider the panel composition rather odd, and ideologically one-sided. Three panelists are not mathematicians, but are wholehearted believers in constructivist approaches to math education, often derided as “fuzzy math”. Two of them claim, ludicrously, that high-achieving East Asian countries teach math their way. The aforementioned panelists are: journalist Elizabeth Green, education professor Jo Boaler, and College Board’s David Coleman, with a degree in English lit and classical philosophy. When only one side is allowed to talk, of course, it can make any claims it likes.
Watch for yourself.
Observe David Coleman from minute 25:40 on, starting while Elizabeth Green is talking.
Then listen, from minute 26:55 on, as he asserts a “kind of dirty little secret” that:
“the worst math problems of all are test prep problems” these are problems manufactured to prepare for an exam and they are typically done, utterly… … if any good science or craft goes into making a really reliable assessment problem for an exam, none of that goes into test prep material. The test developers have hidden from… …because they are trying to hide the exam from the test prep people, to try to keep it, right?”
Test developers, including the College Board, at least until Coleman arrived, have made available for free complete retired operational exams for test prep. These are not “manufactured to prepare for an exam”. They are the highest quality test items that have survived the lengthy gauntlet of editorial review, item review, bias review, more editing, field trials, more editing, comprehensive statistical analyses, more editing if needed, and still more statistical analysis.
Coleman has been at College Board for over three years, plenty enough time to learn the trade. That, even now, he says “if any good science or craft goes into making a really reliable assessment problem” should frighten us all.