Grading & Commenting

Words of Wisdom

Human beings love poetry. They don't even know it sometimes... whether they're the songs of Bono, or the songs of Justin Bieber... they're listening to poetry.

— Maya Angelou

Grading is a form of summative feedback provided to justify an evaluation (generally a grade). Although it should also be used to help improve performance, it is provided at the end of the process of creation.

In contrast, formative feedback is provided before a grade is given, and it is meant to help the writer revise. It may look much like summative feedback, in that it can consist of marginal and end comments or use a rubric, but it should have as its major aim helping the writer or speaker know how to take a draft or a performance to the next stage.


Methods of Grading

Most commonly, papers or presentations are graded in one of two ways: analytically or holistically.

Analytical grading breaks the final grade into specific criteria, often weighted or tied to a numerical or other value (such as High, Middle, Low). Most often, analytical grading uses a rubric, which is, quite simply, a descriptor of each criterion and its weight.

Holistic grading is based upon a general impression of the whole document or performance, often relying on an evaluator's expert knowledge; for example, a holistically graded paper may use a rubric but one that does not assign weights, since all traits are seen as equally important to the whole effect.

Another distinction that can be made in types of grading is between norm-referenced grading (comparing performance across a range, or grading on a so-called curve) and criterion-referenced grading. The latter compares a given performance to a predetermined list of criteria, or traits, which characterize an optimal performance.



Whether teachers grade holistically or analytically, they generally provide comments to guide future learning and revision (formative feedback), not to mention to justify grades (summative feedback). Comments can be:

  • pre-prepared (a rubric is most useful in this regard)

  • marginal (written in the margins)

  • in-text (written between the lines)

  • summary (written at the end)

Before you begin commenting, attend to your pedagogical aims. Focus the most on issues of importance to you or on issues you stressed in your class. Perhaps you want students to make a strong argument or use technical jargon properly, and that might be something you want to make note of for each student.

Also, remember that you are not an editor. If you mark every error or deal with every problem in an attempt to produce the perfect product, you are doing the work for the students, but you are not encouraging active, independent learning and critical thought. Comments need not be extensive to be useful. In fact, marking every error can be detrimental to learning.

Although many students don't realize it, one reason instructors like double-spaced papers (and editors like double-spaced manuscripts) is that the ample spacing leaves room for marginal and in-text comments. They are used to call attention to specific text features, to query the author, and to signal possible revision. Marginal comments can be effective, but they can also be intrusive.

Summary comments, found at the beginning or at the end of a document or, in the case of a presentation, as a note at the end of a rubric can be used to point toward further improvement or revision. Summary comments can also develop procedural knowledge, further critical thought (especially about the larger issues that require a bit more elaboration than a margin affords), or provide commentary on issues not apparent in one specific area of the text, for example, how well an audience is addressed or how effectively an author appeals to logic or uses evidence.

A Few Tips on Commenting

  • Be tactful.

  • Use questions to provoke thought ("What other evidence might be cited to support this?" "Might this word choice offend some readers?" "Does this example really fit your thesis?").

  • Be constructive; help the student see possibilities.

  • Define any and all grading or editing symbols.

  • Tie your comments to a handbook or website, or suggest students come to the University Writing Center.

  • Comment on strengths instead of weaknesses.

  • Be specific ("Strong evidence" is more helpful than "Good" and "Unclear word choice" more helpful than "What?").

  • Make your comments point ahead to further development. Suggest ways to improve.

Remember to use the principle of independent and active learning in formulating comments. Your student needs to find and identify problems in the text and to consider options for revising. Minimal marking is a concept developed by Rich Haswell to describe an approach that puts responsibility on the student. [See "Minimal Marking" College English 45 (October 1983): 600-604.] In this approach, you may explicity mark one or two errors, but usually, you place a check mark in the margin on a line where an error occurs and let students find and correct it.



Rubrics are timesaving and less problematic than other forms of commenting, though they may also be somewhat impersonal and "canned." To design a rubric, decide on the traits to be assessed and their relative weight. These traits are what Richard Lloyd-Jones (1977) called primary traits, and deciding on what they should be is primary trait analysis. For a quick and simple discussion of primary trait analysis and criterion-referenced grading, see "Establishing Criteria and Standards for Grading" in Effective Grading by Barbara Walvrood and Virginia Johnson-Anderson. They suggest that primary trait analysis use a 2-5 point scale; a two-point scale is basically "Pass/Fail." However, you might use this same technique with a different numerical scale, weighting some traits more than others, or tying them to letter grades or to a 0-100 scale, for example.

Keep in mind that some traits may be "gateway" traits: in other words, if they are not included or satisfactory, the paper isn't even graded. Spelling, grammar, or format issues are often best dealt with as gateway traits.

Rubrics can work very well to keep your grading even and fair. The less variation possible in the way an assignment can be approached, the more useful a rubric will be. If there is a wide range of possible "A" papers, make sure your rubric can encompass that variation. In selecting a style of rubric, consider that while points may seem more neat and less arbitrary than a designation such as High, Middle, Low, they can also be problematic in assessing writing, which is, after all, partly a matter of judgment. Grammar and punctuation rules can vary widely: what you regard as an error may not be so regarded by even very good editors. So whatever system you use, leave some latitude for judgment and flexibility. The rubric below assigns points to specific traits; it is an abbreviated example of a rubric for evaluating a business memo. Many more traits could be included, but these are the traits the instructor deems most important for the specific assignment.

  Excellent Good Fair Poor
Clearly defined audience & purpose 4 3 2 1
Appropriate tone 4 3 2 1
Important information in 1st ¶ 4 3 2 1
Spelling 4 3 2 1
Punctuation 4 3 2 1

If you feel that these traits should be weighted equally, potentially a paper with inadequate mechanics could be rated as a "B." Therefore, some instructors might prefer to weight the traits, as, for example, in the following rubric:
  Excellent Good Fair Poor
Clearly defined audience & purpose 7 6 5 4 3 2 1 0
Appropriate tone 7 6 5 4 3 2 1 0
Important information in 1st ¶ 7 6 5 4 3 2 1 0
Spelling 4 3 2 1
Punctuation 4 3 2 1
Format 4 3 2 1
Total Points        
  A (27-33) B (20-26) C (13-19) D (6-12)

Another form of rubric, below, simply uses High, Middle, or Low for each trait; the grade might be based on a holistic sense of how many of each constitutes a letter grade, or the letter grades can be defined by the number of each level (i.e., an "A" requires at least 4 Highs).
  High Middle Low
Development of content      
Audience Accommodation      

Following is a frequently-used rubric created by Diedrich in 1974 for large-scale assessment of writing. This rubric works especially well with multiple evaluators, since the total possible points add up to 50. Two readers can score separately, and add together their scores, or one reader can score and multiply the result by two. If two evaluators are more than 5 points apart, it signals they need to have a third reader render a judgment or provide a new score. (Cited in Cooper, Charles R. "Holistic Evaluation of Writing." Evaluating Writing: Describing, Measuring, Judging. Eds. Charles R. Cooper and Lee Odell. Urbana, IL: NCTE, 1977. 7.)
  Low           Middle           High
Ideas 2 4 6 8 10
Organization 2 4 6 8 10
Wording 1 2 3 4 5
Flavor 1 2 3 4 5
Sub Total: ______
Usage 1 2 3 4 5
Punctuation 1 2 3 4 5
Spelling 1 2 3 4 5
*Format 1 2 3 4 5
Sub Total: ______

Total: ______

*The original rubric uses "handwriting" here, a rather outdated criteria, except perhaps for in-class essays. Following is a rubric used 

Additional Resources

"Running a Grade-Norming Session" by Pamela Flash (University of Minnesota). Helps if you want an assistant to grade.

Sample rubrics from Holt, Rinehart and Winston.

Rubistar, "a tool to help teacher make quality rubrics."  

Model Rubrics and Descriptors

Video on Minimal Marking by Richard Hswell