Bias and Exclusion in Assessment 

How can bias and exclusion in assessment be addressed in our course design and classroom activities? Jesse Stommel, a proponent of “ungrading,” suggests we start by reflecting on what we might call our grading philosophies.

Stommel recommends asking yourself the following questions:

  • Why do we grade? How does it feel to be graded? What do we want grading to do (or not do) in our classes (whether as students or teachers)?
  • What do letter grades mean? Do they have any intrinsic meaning, or is the value purely extrinsic? Does assessment mean differently when it is formative rather than summative?
  • How do written comments function as (or in relation to) grades? To what extent should teachers be readers of student work (as opposed to evaluators)?
  • What is the role of self-assessment and peer-assessment?
  • What would happen if we didn’t grade? What would be the benefits? What issues would this raise for students and/or teachers? Would we be forced to rethink our systems for evaluation?

Obviously these questions get at the heart of higher education practices. Stommel’s approach goes further than most conceptualizations of alternative assessment. Stommel, who defines “ungrading” as “intentional, critical work to dismantle traditional and standardized approaches to assessment,” is far from the only advocate for dismantling current grading culture and practices in the name of inclusivity. Asao Inoue points out that the goals of antiracist pedagogy cannot be achieved without changing the way we assess students. Alfie Kohn and Peter Elbow argued that traditional grading diminishes the learning enterprise for everyone. Many have noted the central role of flawed assessment in the persistence of gender bias in higher education, from student retention to tenure and promotion. Anne Meyer and David Rose propose fundamental revision of course design, including assessment, in their concept of Universal Design for Learning.

Here are some specific things you can try

  • Ask students to assess themselves.
  • Encourage meta-cognition.
  • Teach peer review and require it.
  • Use contract grading, specifications grading, or other types of criterion grading.
  • Use rubrics judiciously
  • Use labor-based grading
  • Do minimal grading
  • Provide extended time on exams.
  • Allow use of aids such as calculators during the exam
  • Allow audiotaped or oral versions of the exams
  • Design alternative testing methods (i.e., demonstrating mastery of course objectives using a research paper, oral presentation, etc.)
  • Design more frequent tests, quizzes, or exams to obtain additional feedback

        (The last five items on the list above are from the Heath Resource Center.)

        Can ungrading work in STEM classes? Stommel says yes. Standardized approaches to assessment in STEM include exams, problem solving homework, and lab reports. What if we could ask students to use a discussion board to show and discuss their strategies for solving problems, like Cross and Palese suggest in the documented problem solution method? What if instructors implemented specifications grading, contract grading, or labor-based grading contracts? What if students in STEM courses could be tasked with self-assessment? To what extent are STEM courses missing an opportunity to engage students’ meta-cognition about learning?

        It is well known that standardized testing such as the SAT has for decades struggled to formulate questions that don’t skew in favor of white, middle-class students. It is not only in testing that diversity and inclusion need attention. Written assignments and assessments often favor a thin slice of language users. The required modes, too, (and timing) of submitting work reflect an accessibility bias.

        Assessment instruments and grading criteria are entrenched in most disciplines. Norms of assessment suffuse entire institutions, accreditation schemes, and teaching evaluation instruments. In this context it may seem unnecessary to change things. It can even feel risky to change one’s grading practices. But what if our approach to assessment could be better? It can be presumed that no instructor wants to exclude students from learning. Just as implicit bias is manageable through conscious effort, reflective practice, and a commitment to action, so, too, can assessment strategies be revised for inclusivity by asking ourselves: why do we grade students?