People who work in the area of learning outcomes assessment are frequently asked why course grades can't be used as measures of student learning.  After all, aren't grades a measure of student performance?

Unfortunately, the answer is more complicated than a simple "yes" or "no".  Most instructors know that students engage in a wide range of processes as they produce material that is graded in the context of a course, including, but not limited to: interpreting the assignment, developing an idea to be submitted in response to it, laying a foundation for implementing that idea (e.g., conducting research, practicing a skill, working at an internship or service site), producing the actual product (e.g., writing up the experiment or research paper, making an oral presentation, gathering materials for a portfolio), presenting that product for evaluation by conforming to instructor's criteria for format and deadlines.  Any of these processes may be graded separately, or the whole might be graded in aggregate. Which grade best reflects achievement of a program level learning outcome?  Or, different instructors may have different philosophies of grading, differing in opinion as to whether a grade reflects an objective degree of mastery in a discipline, or an individual's progressively better mastery within a course, or of a task at a level appropriate to the course level, etc.  Unless instructors know what is being evaluated, and share an understanding of what that evaluation means, it will be difficult to use grades to assess program-level learning outcomes.

It is tempting to say that grades cannot be used for assessment; however, doing so could be a lost opportunity.  After all, instructors are very familiar with the concept of grades and grading.  The work that is graded tends to be assigned by instructors so they have authentic evidence of what students are expected to know or do as a result of what they've learned.  And grades provide students with information that helps them gauge performance against the instructor's criteria. And, importantly, final grades that evaluate student performance in a course signal student readiness to move to the next course. Grades, in general, would seem to be an efficient way to directly measure student learning.

So how to make grades work for assessment of student learning, moving evaluation beyond the boundaries of an assignment or course, and into the program level?  For purposes of program-level assessment, using grades may work if the faculty agree to be as clear about what they mean, and as systematic as possible in applying them. It may be reasonable to use grades for assessment if certain conditions apply:

1. The grading criteria are developed with excellent practice. This means that all of the faculty in a department, program, or discipline collectively develop a common understanding of what each grade means, work together to connect those standards to the knowledge in the discipline, work together to explicitly and clearly connect course assignments and/or courses to those standards, and use these common grading criteria to assign grades.  By doing this, every professor and instructor who is grading student work is evaluating that work in light of what the student has demonstrated about the learning outcomes. Grades can then be aggregated to reveal, at a program level, how well the program is doing for all students.
   
   What does this mean, as a practical matter?  It will reduces variation in grading by implementing common standards and reducing practices like adjusting grades based on matters that don't relate to the identified learning outcomes (e.g., if a paper was submitted late, or if a student improved a second draft far beyond the promise of the first, or if a student "almost got it").  Of course, since many instructors use grades to evaluate performance on complex tasks that include personal responsibility, working on teams, improvement over time, integration with other work and projects, etc., this may seem very restrictive.  Furthermore, establishing common grading criteria means having an open and explicit conversation about whether an "A" in this course is the same thing as an "A" in that course. To use grades for assessment, instructors must have that conversation...which isn't always easy.
   
   
2. Once grading criteria are standardized, the criteria should be well written and shared. It is good practice to make these standards public, so students are aware of what grades mean, too.  (Just as instructors' understanding of grades vary, so too, do students'.) Sharing the criteria for grading will help students understand that the work they do in one class is related to and evaluated by the same criteria as the work they do in another.
   
   
3. Finally, the information obtained from grades should be used to understand and improve learning.  For all of this effort to be worthwhile, it's important to ensure that the information obtained from this process is gathered, analyzed, and that changes are made to improve student learning in the program.

Using grades for program-level assessment requires a lot of work, and will likely engage faculty and instructors in challenging conversations (which will likely need to be repeated periodically, as new members join the community of learners in the department/program). It is difficult - but it can be done.

For additional information about using grades, I recommend consulting Barbara Walvoord and Virginia Johnson Anderson's book, Effective Grading: A Tool for Learning and Assessment (Jossey-Bass Higher and Adult Education, 1998).

Elaine M. Klein
Associate Dean for Academic Planning, L&S