In the April issue of the Journal of Assessment and Analytics there was an article that very positively presented an approach to grading that a professor at Tulane University was using to give her students “more control over their own learning.” Her grading plan included seven items each with a four-percentage point range within which students could choose the weight of the item in their grade. (For example, a student could choose the weight of the final exam to be between 20 and 24 percent.) The seven items were Clicker points, Journal entries, My Econolab, In-class projects, Mid-Term #1, Mid-Term #2 and a Final Exam.
This approach to grading is very problematic in a number of ways.
The most serious problem with a grading plan like this is that, although it has been the norm for university (and high school) grading, it says the course is about grade grubbing for points and percentages, not learning. Grading plans should be organized by learning goals/standards, not methods of assessment or activities. A more appropriate grading plan for an economics course would identify 4-10 big ideas or learning goals and each assessment would provide information about the students’ proficiency on whichever learning goals were included in that assessment.
The suggestion that this approach “encourages students to more carefully consider course expectations and reflect on their own abilities” is a deception because it assumes that students will make good choices and the small range of choice of points is basically meaningless. For example, if a student wanted to maximize the impact of formal assessments and minimize class work and participation and received 90% on the formal assessments and 80% on class work their grade would be 86.4%. If the reverse was the case and a student received 80% on the formal assessments and 90% on the class work their grade would be 82.6%, a difference of only 3.8%. If there had been no choice and the professor had set the percentages to the midpoint of the range for five of the items and the high end for the last two formal assessments, the first student’s grade would be lower by 0.2% and the second student’s grade would be higher by 1.2%
It appears that only the last three items are summative assessments and these are the only items that should be include in the determination of a subject grade. The first four items are learning activities so they are formative assessments and ideally receive descriptive feedback not scores. If they are scored, the scores shouldn’t be used to determine grades.
It is way past time to eliminate percentages and use 2-7 levels of proficiency. (see article about Guskey in the June newsletter)
If a professor really wants to “give students more control of their own learning” simply playing with the percentages in the grading plan doesn’t work. To give students more control of their learning, grades need to be based on learning goals and levels of proficiency, not methods of assessment and points and percentages. Students also need to have at least some real choices in the types of evidence (products, observations, conversations) that are used to determine their grades and they must be encouraged to be reflective, self-directed learners.
About the Author: Ken O'Connor is a former Curriculum Coordinator with the Scarborough Board of Education in Ontario, Canada. He is an expert on grading and reporting with a particular emphasis on using these techniques to improve student achievement through student involvement.