WIP: Identity-Based Bias in Undergraduate Peer Assessment

Peer evaluation is a commonly used practice for group work throughout higher education as it allows students to provide and receive insight as well as provides additional information to faculty. When such peer evaluations may be taken into consideration for grades, it is especially important to look closer at the potential bias that may be present. Students may unconsciously perceive members of their group in biased ways because of ingrained gender or racial biases, potentially resulting in skewed feedback. In this project, we look at how student feedback varies for students with historically underrepresented identities. To accomplish this, we will use survey data from a software tool that monitors team members throughout the course of a project to provide a link between students and instructors. Part of this software allows team members to evaluate one another and provide feedback, including items such as reliability, effort, quality of work, and idea generation. Using hierarchical linear regression, we will investigate patterns of ratings based on identity characteristics such as gender and race (of both evaluator and evaluated) to investigate how these factors are related to evaluations.