Learning Progressions As Evolving Tools in Joint Enterprises for Educational Improvement

摘要

In their article, "Using Learning Progressions to Design Vertical Scales that Support Coherent Inferences about Student Growth," Briggs and Peck (this issue) argue that an important goal of assessment should be "to support coherent and actionable inferences of growth." They suggest that current approaches to test design rely on domain sampling, as well as statistical methods for vertically equating tests, fall far short of this aim. Current test-equating methods fall short, Briggs and Peck argue, in part because these approaches fall to communicate assessment information about student growth. Parents and teachers both are critical partners in using assessment information, but they need to be able to interpret and make use of test score information to take action to support their children's learning. Briggs and Peck (this issue) argue that one of the strengths of learning progressions is that they specify hypothetical, but empirically grounded, pathways of student learning in a domain. A learning progression provides "a map of a territory" of possible ways students might develop in a domain, and each location on the map is linked conceptually to subsequent locations explicitly, as a possible next step in learning. Of particular importance is that those "locations" are described qualitatively, and they present an opportunity to represent growth in terms of those locations rather than as a number, a kind of percentile rank (as in the Colorado example described by Briggs and Peck, this issue), or set of ordered levels that characterize student understanding as "basic," "proficient," or "advanced."