Grades Versus Comments: Research on Student Feedback

The most recent Marshall Memo has this short piece about grading and learning.  I thought you’d find it interesting.  I thought that the focus on WHY we would give grades and/or comments was refreshing!

Thomas Guskey on Grades and Comments

            “Are comments on student work superior to grades?” asks assessment guru Thomas Guskey (University of Louisville/University of Kentucky) in this article in Phi Delta Kappan. “It depends… The research on this issue is far more complicated and more highly nuanced than most writers acknowledge.” Guskey cites several studies that provide helpful guidance for K-12 educators.

            • A 1958 study by psychologist Ellis Page – Secondary-school teachers gave numerical scores on their students’ assessments and then converted the scores into A, B, C, D, F grades. Three randomly-selected groups of students then got their papers back with:

-    Numerical and letter grades only;

-    Numerical, letter grades, and standard comments for each grade:  A: Excellent! Keep it up.  B: Good work. Keep at it.  C: Perhaps try to do still better?  D: Let’s bring this up. and F: Let’s raise this grade!

-    Numerical score, letter grade, and individual comments based on each teacher’s personal reactions and instructional priorities.

Page compared the impact of these three approaches by looking at how students did on their very next assessment. Here’s what he found: students in the first group did no better; students in the second group did significantly better than those in the first; and students in the third group did better still. The conclusion (which has been confirmed by subsequent studies): grades are helpful only if they’re accompanied by teachers’ comments.

            What’s striking about this study is that the standard, boilerplate comments given to the second group of students had such a positive impact. The comments involved very little work for teachers, but made almost as much difference as the much more time-consuming individualized comments given to the third group of students. Guskey believes a little-recognized insight from Page’s study is the nature of the standard comments. First, each of these seemingly robotic comments communicated the teacher’s high expectations and the importance of students’ continued effort. Second, all the comments made clear that the teacher was on students’ side and willing to partner with them to improve. Instead of saying You must raise this grade, the comment was Let’s raise this grade! – conveying, I’m with you in this, we can do it! In other words, says Guskey, “The message teachers communicate in their comments may be what matters most.”

            • Benjamin Bloom’s mastery learning – In the late 1960s and 1970s, Bloom promoted the idea that on formative assessments, students should receive a grade of Mastery or Not Mastery. Bloom defined Mastery as the clearly described level of performance that teachers believe would deserve an A, which then becomes the standard of mastery for all students. Students scoring below Mastery on formative assessments are in a temporary state, not there yet, and should receive diagnostic and prescriptive instruction from the teacher and additional chances to demonstrate mastery. Bloom believed that with sufficient time and skillful corrective instruction, 95 percent of students can achieve Mastery. In short, Bloom believed in comments to guide under-par performance to mastery grades, guided by clear expectations up front.

            • Ruth Butler’s 1988 study – Fifth and sixth graders took a test and were then divided into three groups, each receiving a different type of feedback:

-    Grades from 40 to 99 based on students’ relative standing in the class (norm-referenced or competitive grades);

-    Individual comments on students’ performance on the objective (criterion-referenced or task-focused);

-    Both competitive grades and task-oriented individual comments.

The study found that students in the second group did best, indicating that competitive grading is not an effective practice, and task-focused comments can boost learning by giving students specific information on their performance and suggestions for improvement. What’s interesting is that the competitive-grades approach benefited high-performing students, maintaining their interest and motivation, while undermining the interest and motivation of low-performing students.

            Guskey adds that the nature of the comments is the key factor. In Butler’s study, they were task-oriented and instructionally helpful. Additional research by John Hattie and Helen Timperley reinforces the idea that it’s the quality, nature, and content of teachers’ comments that make a difference.

            • Guskey’s conclusions – First, he says, grades – whether they are letters, numbers, symbols, words, or phrases – are not inherently good or bad: “They are simply labels attached to different levels of student performance that describe in an abbreviated fashion how well students performed.”

            Second, grades should always be based on learning criteria that the teacher has clearly spelled out. Grades that compare students to their peers do not move learning forward. In fact, says Guskey, “Such competition is detrimental to relationships between students and has profound negative effects on the motivation of low-ranked students, as the results from the Butler (1988) study clearly show.”

            Third, assessments must be well-designed, meaningful, and authentic, and grades should reliably and accurately measure the learning goals and provide useful information to guide teachers and students to improve learning.

            Fourth, grades by themselves are not helpful. “Grades help enhance achievement and foster learning progress,” says Guskey, “only when they are paired with individualized comments that offer guidance and direction for improvement.” And of course those comments must be followed up with time and support for students to improve their work.

            Fifth, students and families must understand that grades don’t reflect who students are, but their temporary location on the learning journey. “Knowing where you are is essential to understanding where you need to go in order to improve,” says Guskey. This metacognitive awareness also makes students better judges of their own work and increasingly self-sufficient as learners.

            Finally, Guskey sums up the collective wisdom of researchers, especially Benjamin Bloom and his colleagues, on effective comments on students’ tests, essays, products, performances, or demonstrations:

-    Always begin with what the student did well, recognizing accomplishments or progress.

-    Identify the areas that need improvement.

-    Offer specific guidance on steps the student needs to take to meet the learning criteria.

-    Communicate confidence in the student’s ability to achieve at the highest level.

“Grades Versus Comments: Research on Student Feedback” by Thomas Guskey in Phi Delta Kappan, November 2019 (Vol. 101, #3, pp. 42-47), available at https://bit.ly/2P3DSnW for PDK members, or for purchase; Guskey can be reached at guskey@uky.edu.