Developing a peer evaluation instrument that is simple, reliable, and valid

Abstract

A multi-university research team is working to design a peer evaluation instrument for cooperative learning teams that is simple, reliable, and valid. In this work, an overview of the process of developing behaviorally anchored rating scales (BARS) will be presented, including the establishment of a theoretical basis for the instrument and a description of the extensive classroom testing of the draft instrument conducted during fall 2004. Introducing the draft instrument to the engineering education community through exposure in the NSF grantees' poster session is expected both to improve the validity of the scale itself through the feedback we receive and to accelerate the dissemination of the instrument.