What’s still wrong with rubrics: Focusing on the consistency of performance criteria across scale levels

Scoring rubrics are currently used by students and teachers in classrooms from kindergarten to college across North America. They are popular because they can be created for or adapted to a variety of subjects and situations. Scoring rubrics are especially useful in assessment for learning because they contain qualitative descriptions of performance criteria that work well within the process of formative evaluation. In recent years, many educational researchers have noted the instructional benefits of scoring rubrics (for example, Arter & McTighe, 2001; Goodrich Andrade, 2000). Popham noted their potential as “instructional illuminators” in a 1997 article entitled What’s Wrong and What’s Right with Rubrics, but he also cautioned that “many rubrics now available to educators are not instructionally beneficial” (p.72). Unfortunately, many rubrics are still not instructionally useful because of inconsistencies in the descriptions of performance criteria across their scale levels. The most accessible rubrics, particularly those available on the Internet, contain design flaws that not only affect their instructional usefulness, but also the validity of their results. For scoring rubrics to fulfill their educational ideal, they must first be designed or modified to reflect greater consistency in their performance criteria descriptors.