In elementary and secondary education, formative assessment is in
vogue. It has become a common theme at educational conferences, a standard offering in test company catalogues, the subject of state requests-for-proposals, and
a focus for teacher-inservice training. A key reason for the popularity of
formative assessment is, undoubtedly, the claims that have been made for its
effectiveness. The most commonly cited quantitative claims are for effects
of between .4 and .7 standard deviations, although some individuals have
asserted effects of up to 2 standard deviations. This paper reviews the
evidentiary sources cited for these claims, and summarizes how the
effectiveness of formative assessment might be responsibly represented.