COLI Guide: Providing Feedback on Student Work

This article is a companion piece to Episode 11 of the Pedagogy Primer Podcast.  You can listen to that episode below, or here. 

On a wide variety of assessments - papers, projects, lab reports, exams, quizzes, and so on - professors commonly provide commentary alongside number or letter grades.  Ostensibly this is to assist students in improving their work, either within a course or along their academic and professional career, generally. 

However, professors frequently regard feedback as a part of grading, which is an unpleasant chore imposed by current expectations in higher education.  In those circumstances, it is tempting to view our marginalia on student work as justification for any grade below 100%, or worse, an effort to discourage grade complaints by students. 

But we can instead think of feedback as a learning situation, or a major part of the course where students learn knowledge, skills, abilities and good habits.  Rather than a distraction from teaching, it is teaching, and perhaps some of the most effective pedagogy in our courses.

More than a Grade

A letter or number grade is a form of feedback, but in most cases students need more.  

  • Grades are a crude measure of performance, but provide little indication of process - did the student use good methods, properly, to arrive at an outcome?
  • Grades can be a distraction.  Especially with assignments or exams that are a big portion of the overall course grade, students may be more interested in grade calculation than reading or hearing the professor's directions for improvement. 
  • On the other hand, grades may be an incentive, especially with frequent, low-stakes assignments.  In many cases, students are unlikely to complete ungraded assignments or activities just for the benefit of your feedback.  (Although especially motivated students might.)

Feedback Prior to a Grade?

Students occasionally request feedback on their work prior to a grade.  This can be beneficial but it is also additional work for the professor.  In a worst case, a student may be (repeatedly) testing a level of effort against your grading practices, or seeking a guarantee of a higher grade along with inappropriately specific instructions.  Here are some tips for handling student requests for feedback prior to graded assessment:

  • set a deadline before your final assessment deadline, so that you have time to read, assess, and return their work with comment, with suitable time for them to employ your comments effectively. 
  • indicate that students may only seek pre-grade feedback once on an assignment, and perhaps once in a semester.
  • set some ground rules for what you need to see, so that you aren't faced with something so incomplete or fragmented that it's impossible to comment effectively.  (see checklists below.)
  • be explicit to students, that these rules reflect practical efforts to reserve time for more productive teaching tasks.  Plus, they mirror real-world circumstances where creators must exercise judgment in soliciting feedback, since they cannot constantly pester supervisors or editors.

Course and Assignment Design

Design assignments that work in tandem with feedback as part of a bigger learning and assessment process.  

Clarity is Crucial  

When designing assignments

  • describe as exactly as possible what you need to see, for an A grade or for various grade levels.
  • indicate clearly the scale and scope of the assignment.
    • How long must their written submission or video recording be? 
    • On or within which topics or areas should it focus? 
    • What sources are required or admissible?  
    • What types of evidence, calculation, writing structures, or other information must they include?
    • How might they structure their efforts, for efficient completion of the requirements?
    • What might be a basic requirement for an accepted submission - see "Go/No Go" checklists below.

Vague assignment instructions can confuse or demoralize students and make the assignment and subsequent feedback seem in bad faith.  It may be tempting to have vague instructions with the thought that you'll grade with a light hand, and are flexible as to what you will accept as good work.  But students don't necessarily know or believe that.  

You may be reasonably irritated if research colleagues or academic or college leadership issue vague expectations for tasks they expect you to complete.  Bear those experiences in mind when designing assignments. 

Be sure your assignments serve or relate to learning goals and objectives in your course.  These need to be published to students anyway.  Tie your feedback back to those learning objectives as well.  This helps students understand an assignment's purpose.

Repetitive Assignments

A good instructional design practice is to have more frequent, lower-stakes assessments.  This has several advantages, but can particularly empower your feedback to students.  

Offer repetitious assignments that allow students to practice the same skills and abilities, albeit with different content.  For example, perhaps students write eight short analysis essays throughout the semester, on different cases, scenarios or data. But each assignment requires students to use the same professional or disciplinary research, interpretation, analysis, and writing skills.  With each assignment, students have the opportunity to improve upon those skills, directed by your feedback on their previous assignment submissions. 

This makes it easier for you to track individual student progress, too.  If assignments are smaller, and focused on a short set of objectives or criteria, you may track individual student weaknesses, efforts and progress across the set of assignments.

This is what Johann Koehler calls the "Feedback Loop."

Repetition in Asynchronous Discussion

Online asynchronous or "message-board" discussions present one of the best places for repetitive assignments where students can improve skills.  For example, you may ask students to share and compare analyses of various sources.  Both the analysis skills, as well as written communication skills can be developed over a series of weekly discussion topics and posts.  A common rubric can be used to assess each student contribution.  At the end of the semester, a student can be asked to choose what they regard as their best discussion contribution, and briefly describe why it is so.

Scaffolded Assignments

Similar to repetitive assignments, scaffolding means having a set of assignments that repeat and gradually combine skills as they progress.  A traditional example is the "term paper" or research-writing project, beginning with a proposal and ending with a final paper that's akin to a published article.  But these can also be a business proposal project, a series of mathematical assignments, or an artistic creation.   

Early and Often

Assignments and feedback can be more effective if:

  • Some assignments are completed early in the semester.  This gives students ways to improve their work during the course.  It also allows you to /wiki/spaces/AcadAffairs/pages/34964116 early on, so advisors, associate deans, and the Griff Center can then assist struggling students in various ways.
  • Assignments are frequent, and as graded assessments, low-stakes.  This incentivizes students to complete regular, reinforcing exercises, while knowing that they have plenty of opportunity to recover from a single bad experience.  This also cuts down on counterproductive stress and preoccupation with overall course grades.  Grading short assignments can be easier for the professor that grading a giant stack of large papers or exams, with more complicated rubrics and requirements.  
Early, frequent, and repetitive or scaffolded assignments within a course offer the possibility that students may act on feedback they received from their professor within the same course.  At the same time, the professor may help students, in several assignment-feedback cycles, to improve their work.  This avoids the familiar problem where feedback is (as Barbara Walvoord and Virginia Johnson Anderson put it) basically an autopsy that just explains a grade.

Feedback Structures

Checklist and the "No-Go" Principle

Do you find yourself having to write the same comments about mechanics, or even higher-order problems in student work?  Are these skills or abilities that you think are well within students' capabilities, and so are basic requirements for any passing grade?  You can develop a checklist as part of an assignment prompt, that lists a priori requirements for any assignment to achieve a grade above failing, or for any assignment to be graded at all.  This can be a Go/No Go List, that determines whether you will (continue to) read and grade a student submission, or regard it as submitted.  A Go/No Go policy can avoid you having to spend precious time wrestling with or commenting on work that is distinctly short of the assignment's requirements and your students' abilities.  Instead you have more time to provide higher quality feedback where students can best benefit by it.  

Obviously you need to be careful in drafting a list: what items do you think are baseline prerequisite knowledge, skills and abilities your students must have if they have reached your course and this assignment?  On the other hand, what are challenging standards that you cannot assume all of your students will achieve to the extent that they belong on a checklist?  If your assignment closely adheres to learning objectives for the course, these are probably the targets of the assignment.  In any case, you want be sure that your checklist is not turning away work that a sincere effort to meet the assignment's requirements. 

Go/No-Go Lists for a student research paper or lab report might include basics such as a title, page numbers, a bibliography, appropriate citations, obligatory data visualizations, calculations, structural elements, and maybe even some composition elements such as topic sentences.  


Rubrics are a common form of structured feedback that potentially saves a professor time, and provides a more consistent assessment that is tied to a specific set of criteria.  Students in American higher education today have probably seen rubrics in K-12 schooling and are accustomed to their use.  Rubrics are typically understood as a grid structure featuring criteria in rows, with levels of achievement for each criteria (and, perhaps, all criteria collectively) represented in columns.  A professor "grades" by choosing a level for each criteria.  In D2L, this is a few quick clicks.  While rubrics might make providing feedback easier, they typically take time to develop, perhaps several semesters' use to fine-tune, and occasional adjustment thereafter as new students represent different experiences in prior education.  

A few tips on rubrics:

  • Rubrics are fairly easy to construct and use in D2L, to create a push-button tool for quick feedback.
  • Before writing a rubric from scratch, ask colleagues for examples or advice.  Google searches will bring up many examples, and conversations in social media groups or other professional groups often include links to rubrics.   
  • Writing rubrics helps professors develop assignment prompts.  When writing a rubric you hone your priorities, rank them, and eliminate unnecessary elements.  
  • When writing rubrics, try to associate criteria directly or as closely as possible with course learning goals or objectives.  
  • You may write additional, specific comments to each student, around your rubric.  The rubric can still help you focus your thoughts on what each student needs to do next.
  • Instead of simply duplicating language between your rubric and your assignment prompt, provide the rubric as a part of your assignment prompt.  
  • If you use lots of similar or repetitive low-stakes assignments, a rubric to assess them is even more efficient.
  • Be patient with yourself and your rubrics: a great rubric often requires several semesters' time to hone into a very effective instrument.  You may use a rubric for decades thereafter, with occasional modifications to reflect changes in students' strengths and weaknesses, or your assignment design.

COLI offers an example of a rubric for asynchronous discussion topics.  While you will likely need to customize it for your course, it includes common language that can get you started.

Automated Quizzes

Short, low-stakes quizzes are a common teaching tactic.  Completed and collected at the beginning of class, they can assess the degree to which students are reading effectively beforehand, or at least enforce reading requirements.  But traditional paper quizzes can be time-consuming to print, pass out, collect, and grade.  

D2L's quiz engine can move quizzes outside the class period.  While students then have access to their readings (and each other), the quizzes become a sort of low-stakes worksheet exercise that helps them quickly determine their level of reading comprehension.  Features such as question and answer randomization incentivize students to make a sincere attempt at the assignment.  If questions are limited to multiple choice, D2L can do the grading for you, and you can quickly review student performance on these quizzes by scanning your D2L gradebook.  

It is even possible to produce take-home exams in D2L.  These open-book assessments can, like a higher-stakes comprehensive worksheet exercise, help you and your students understand how they are bringing together several weeks' worth of coursework into more comprehensive problems, essays, or multiple-choice reviews.  

Actionable Feedback

In or outside of classrooms we often break up learning into manageable chunks; courses and textbooks are organized into lessons, and golf or swimming instructors encourage students to focus on a particular action or aspect of good form on a particular day on the course or in the pool.  This makes it easier for those teaching to organize what they teach into a meaningful or beneficial sequence, and for students to manage learning large bodies of content, methods, or practices.  We can do something similar in assignments and feedback.

Whether on a daily math homework assignment, or hefty lab report, a team business report, or a semester-level research paper, junior faculty often respond with every criticism or suggestion we can think of, or become exhausted in the attempt.  For student work that merits improvement, this can be an avalanche of red ink down the margins.  This contributes to the sense that grading is an exhausting chore that distracts from more useful professorial work, and more experienced faculty often devise ways to avoid making so many comments on student work. 

Brutally extensive commentary can also be discouraging for students, who hardly know where to begin, and adopt unhealthy attitudes toward feedback.  Some students ignore it altogether.  If faced with revising a draft, students move as quickly as they can through feedback commentary, making perfunctory or inadequate changes, while making no distinctions between major concerns and minor errors or omissions.  Students might reasonably assume that they only need fix what you specifically identified at each point in the paper, rather than consider other places where your criticism is relevant, or their work as a whole.  

Instead of providing over-comprehensive feedback consider how you can respond to student work with a few important action-items:  between now and a similar situation, what two, three, or four things should they work on improving? 

  • Address the items you feel are most consequential for their learning or future work.  Table (for now) items that are easier to fix, or less crucial to the exercise.
  • Be explicit: in your course and assignment instructions, and interactions with individual students, explain that you are using this action-items approach, and that there may be other issues with their work you are not addressing on each assignment.
  • Action-item feedback works best with repetitive and scaffolded assignments: students can employ your feedback within the same course, overall grade calculation, and your mentorship.
    • Refer back to your comments on previous assignment submissions and consider weighting your grade toward their progress on those items.  Collecting digital (instead of paper) submissions can make this practical and efficient.
    • Ideally, as students achieve proficiency or overcome challenges listed in one assignment's feedback action-items, you can address other issues in subsequent assignments' action items.
  • Put lesser and easier-to-avoid problems into a Go/No-Go policy.  For example, excessive spelling or grammar mistakes will not be an action item because those assignments must be revised and re-submitted for a grade and feedback.
  • Try to link action-items back to course learning goals and objectives.  Students then see that working effectively with your feedback means real learning, and not just grade-driven compliance.  Moreover, your comments are sincere attempts to help them, rather than justification for points taken off the grade.
  • Although you might officially determine that some concerns are not action-items for a student, you can still mention them in general terms, or suggest that they consult the writing center or tutoring center. 
  • However, avoid providing minor or secondary items that might distract students from the action items that should govern their next steps.  Make clear what are their priorities for the next assignment.
  • Even if it's the last assignment of the semester, action-item feedback may be more effective than more comprehensive feedback, if you feel students may read it and take it to heart.

Action-Item feedback can make your grading process more efficient.  But it isn't simply letting students get by with less work or attentiveness.  You are saying "your work can be improved in these ways," rather than saying "here's what's wrong with your work."  Action items help students organize their own learning effort, form healthy creative or analytical habits, and acquire self-awareness and confidence about their learning process.

Be Specific

If you choose a particular concern as an action item, you should be clear in your comments as to what is wrong, by way of explanation or example.  Terse words or phrases, such as "awkward" or "confusing" are not enough.  For example, a professor might point out to a student: 

  • “When you wrote that ‘Frida did not like Aldo’s brother, but she respected him,’ I’m uncertain if you mean she respected Aldo, or Aldo’s brother.”  
  • “You have a great set of graphs, but each uses a different longitudinal scale, so for a reader they aren’t directly comparable.  Why not use the same scale?”  

If your action-item concerns multiple parts, places, or instances in their work, you need not address every one, but indicate to them that there are more circumstances they need to revise, than just the one you picked out as an example.

Don't Do the Work for Them

When providing feedback, offering an example calculation or suggesting a phrase isn't bad practice.  But resist the urge to extensively rewrite their work, or perform steps they should properly be doing themselves.  You can always point to one example of a concern, and emphasize that there are similar issues elsewhere in their work that they need to find and fix.  Rather than repeat the textbook, lecture, or discussions at length, indicate in general terms where they need to go: 

  • "See Chapter 7 for a discussion of regulatory capture"
  • "Revisit the lecture from two weeks ago, where I discussed the NotPetya Maersk disaster."
  • "When we discussed CLR James in class, what did we conclude concerning his understanding of the revolution in Metropolitan France?"
  • "In this case, how did we plot the Gravity Turn?"

Interactive Feedback

In several ways, you can interact with students to improve the effectiveness of your feedback:

  • Invite questions. Emphasize that you will gladly meet with individual students to field questions concerning your feedback.  This may feel like provoking grade complaints, but many students may do better if they can go over the matter with you.  
  • Ask students to describe their work.  Before offering feedback to a struggling student, have them describe to you what they think they are doing or did.  This can help you diagnose troubles and prescribe more productive approaches.
  • Response writing.  Assign a short writing piece where students summarize your feedback to them.  This can be especially appropriate for large drafts requiring extensive revision, or assignments toward the end of the semester, where revision isn't possible.
  • Self-assessment.  Toward the end of a semester have students write a reflection piece on how their work has changed, as a result of your feedback, and where they think they still need or desire improvement.

Be Positive, Constructive, and Properly Focused

In several ways, professors can support student motivation in feedback tone and style:

  • Complement or point out good work. Where students did a good job, tell them.  They may not be entirely aware of their success, and it helps to confirm it.  They acquire confidence, but also clarity toward other areas where they need to improve.  
  • Avoid snark or mean-spiritedness.  Harsh commentary is unlikely to encourage student effort.  We are rarely as witty as we think we are, especially when issuing criticism.  Skip mean-spirited comments and focus on professional direction.  If you feel yourself becoming terse or angry due to exhaustion, take a break from assessing student work.
  • Express your confidence and expectations that they will improve.  Make clear that you believe they will consider your feedback and act on it.  This both supports their self-confidence and sets high expectations that they can subsequently see in action, if faced with similar work in the same class.
  • Comment on the work, not the student.  Obviously, personal put-downs are inappropriate, but even compliments should be directed at student work.  "This is excellent" sufficiently reinforces their sense of accomplishment, while "You are very smart!" can leave a student wondering, on several levels, what exactly you think of them and their work. 
  • Consider a meeting or a video.  If you are dealing with complicated or even controversial topics, consider whether recording a short video, or meeting with a student, is a more appropriate feedback situation, so students can hear nuance and tone in your voice, and perhaps see your face.  

Feedback as an Investment

If your feedback methods are closely integrated with course design, and honed to support continual improvement in student work, it can be a more powerful teaching tactic than any lectures or demonstrations.  The extent to which your students understand and act on your feedback may mean it is a major learning situation in your course, worthy of your additional time and focus.

Further Reading and Resources

Episode 11 of the COLI Pedagogy Primer Podcast

Brookhart, Susan.  How to Give Effective Feedback to Your Students.  ASCD, 2017.  

Koehler, Johann.  The Feedback Loop: When Less is More, and When More is Less.  Graduate Student Instructor Teaching & Resource Center, University of California at Berkeley. 2016.

Walvoord, Barbara and Virginia Johnson Anderson.  Effective Grading: A Tool for Learning and Assessment in College. Wiley, 2011.

COLI.  Lesson 4 of the COLI Guide for Teaching Online.

D2L Tutorials