For better or worse, testing—formative,
evaluative, and diagnostic—is entrenched in our K-12 education system. But
policies around assessment haven’t all been evidence-based. In this book,
Howard Wainer, a long-time research scientist at ETS and statistics professor
at Wharton, illustrates this point through a series of rebuttals to opinions he
sees as ubiquitous in today’s conversations around testing. In one chapter, he explains
why the SAT should not be made optional. In another, he gauges the practicality
of using value-added models (VAM) as components of teacher
evaluations—concluding that available data render VAM implementation premature.
While many of Wainer’s arguments dive far into the weeds of testing, his
overall message rings clear and true for much more than assessment: Policy that
is formed without full analysis of the breadth of data available on a topic is
policy that will fail.