ABSTRACT
Measurement practices and research within the context of rater-mediated assessments tend to be dominated by two measurement theories: Generalizability Theory and Rasch Measurement Theory. This chapter provides a comparison of these two measurement theories viewed as component theories within two broader research paradigms—the test-score and scaling research traditions, respectively. It examines two key research traditions that have been found to be useful for organizing thinking about measurement theories during the 20th century: test-score and scaling traditions. The chapter evaluates the connections between test-score and scaling traditions, and considers the advantages and disadvantages of each of these traditions for building assessment systems in the human sciences based on the principles of invariance. It also considers two different research traditions in measurement within the social sciences in terms of their applications in the context of rater-mediated assessments. The chapter analyses invariant measurement for rater-mediated assessments grounded across research traditions.
