Evaluating instructional programs: How much is enough?

Instructional developers need to determine the degree of economically justifiable rigor for evaluating instructional products, with consumers aware of the risks involved.

Abstract

Abstract Instructional developers need to determine the degree of economically justifiable rigor for evaluating instructional products. Rigorous evaluation, involving outside evaluators, random assignment of masses of individuals from well‐defined populations, and massive amounts of data, is often beyond their means. At the opposite extreme, representing no evaluation at all desite the size of consumer market, are most standard educational materials. A successive approximations scheme to empirical determination of effectiveness is needed, with consumers aware of the risks involved. The fear to “teaching to the test” (recently termed “wholly indefensible”) conflics with the basic assumptions of the teachonology of instruction and with the achievement of accountability of the instructional system to its consumers.