Letting Students Correct Their Mistakes to Improve Their Grades

What if our students could improve their grades by correcting the mistakes they made on an exam? This is what we tried in a Special Education Techniques course, and we are very pleased with the results. The first 2 exams became learning opportunities, and students demonstrated a higher level of competency on their final evaluation.

First observation: students don’t read feedback

For several years, Jessie has taught Intervention Planning (Planification de l’intervention), a 4^th- semester course in the Special Education Techniques program. She noticed a recurring problem: students were not taking the time to read the feedback on their work. They looked at their grades but didn’t read the comments. As a result, students repeated the same mistakes on the final evaluation that they had made on the first 2 exams. This was particularly concerning because the final evaluation was open book. Students were allowed to consult their corrected copies of previous exams during the final. Although the structure of the evaluations was the same (in each case, students had to write an intervention plan for a patient), the students did not apply the corrections suggested in the feedback they had already received.

In the winter of 2025, we taught the course together for the 1^st time. We would have liked to co-teach, but for logistical reasons, we taught our groups separately. However, we planned all of our class activities together. As the saying goes, 2 heads are better than 1! Together, we looked for a solution to the apparent lack of student interest in the feedback they received.

Our strategy to encourage self-correction

For the first 2 evaluations, we encourage students to correct their own mistakes rather than doing it for them.

Instead of providing very detailed feedback (for example, “You forgot to define the timeline, which is one of the 4 components of a good objective. Here, you should have written [blah blah blah],” or “This information would fit better in section [X] because [blah blah blah]. Instead, write [blah blah blah]”), we provide what we call deliberately vague feedback. (For example, “Check that all 4 components of the objective are included.” “Is this information in the right section?” or, in the case of a syntax error, “Read this sentence again.”)

We then give students the opportunity to increase their exam grade slightly (by up to 5%) by applying our suggestions to their work. Their grade can’t go down; it can only increase or stay the same.

Each of the first 2 exams lasts 2.5 hours and includes a theoretical section with traditional questions, followed by a section in which students have to write an intervention plan.

• For the mid-term exam, students only have to write the beginning of an intervention plan: identifying the patient’s needs.
• For the following exam, in April, they complete the next part of the intervention: setting objectives and outlining strategies.
• For the final evaluation, there are no theoretical questions. The exam consists of writing a complete intervention plan.

Exams are completed in class, on a computer, to allow students to write the intervention plan in Word.

Each intervention plan is unique. Even though we study several cases in class, new reflection is required to analyze each individual situation. To develop their competency in this, students need opportunities to practise. By understanding their mistakes, they can improve. Simply providing an answer key is not enough; students need to understand why a particular element was inadequate or why a particular decision was a mistake.

When we grade an intervention plan, we annotate the Word document by adding comments in the margins. Some comments are positive (for example, “This section is excellent.”) to help students know what they did well and be able to use that as a model in the future. Other comments are constructive but deliberately vague. We also complete a descriptive evaluation grid.

Part 2 of Exam 1, along with the evaluation grids [in French]. This is the Word document that students must complete for the part of the 1st exam that involves writing an intervention plan. We had a hard time finding the right mathematical formula to ensure the corrections students make afterward could actually increase their grade by up to 5%, but we eventually figured it out (with help from our pedagogical counsellor, Marie-El Domingue!) See the result in the “Récupération” section in the exam header and consult the “Récupération” evaluation grid at the end of the document.

During the class session following each of the first 2 exams, students have 1 hour to improve their intervention plan by completing the “Correction(s)” sections (in red in the exam document), based on the completed evaluation grid and, of course, the comments we have left in the margins. If a student makes the appropriate corrections to their work, their original grade is increased by up to 5%. They can earn the full 5% if all of their corrections meet the requirements. Time is not really a constraint for this: students with fewer corrections can finish in less than an hour, while others may need 45 to 60 minutes. Either way, everyone has enough time to complete the work.

Results

We soon realized that several students didn’t know how to consult comments in a Word document, even though many teachers in the program use Word for digital grading as of the 1^st semester. This was somewhat discouraging, especially when thinking about how much time we have all spent annotating students’ work (often feeling as though we were working harder than the students themselves …).

In any event, this technological limitation is certainly not the only reason students don’t read feedback. In fact, even when we correct students’ work on paper in other contexts, we often feel that our comments go unread and our suggestions go unheeded. Some students even say things like, “If I get over 70%, I’m happy with my grade.” They simply don’t feel the need to look at their feedback.

With our new strategy, we were pleased to see that all our students attempted to correct their work. The activity wasn’t mandatory, since their grade couldn’t go down. Students could have left the classroom without trying to improve their work, with no direct penalties, but they didn’t. They all saw the value of the activity. Some students were unsure how to proceed and were afraid of making their work worse by introducing new mistakes, but we encouraged them to try anyway. If a proposed correction is incorrect, there is no risk. Our feedback on students’ corrections helps them to see where they stand.

Students are glad to have the opportunity to correct their mistakes! We’ve achieved our goal: we’ve enabled our students to reach full potential. We wanted to put the ball in their court, enabling them to learn more effectively and improve.

Without conducting a formal statistical analysis, we’re convinced that this activity has positive outcomes. Final exam results are higher, and by the end of the semester, students demonstrate stronger skills in writing an intervention plan.

Last year, Marie-Soleil taught the course on her own and repeated the same activity, with equally positive results. With the support of Marie-El, she added a descriptive evaluation grid that gives students even more feedback on their corrections. (The 1^st year, we wrote comments in response to the students’ corrections and assigned a grade without providing a grid.)

Is it worth trying in other courses?

The problem of student disengagement from feedback is not specific to our course. For example, we notice that students do not naturally refer back to work from their 1^st semester to understand and correct their mistakes.

We think this self-correction approach could be very useful in other courses and contexts. It is particularly well suited to courses that require students to think critically and apply their knowledge, rather than simply memorize theoretical concepts. It is also more realistic for high-stakes evaluations. In courses with many low-stakes assessments, time constraints can make this approach more difficult to implement.

What about you? Can you see a way that self-correction could be integrated into your own teaching context?