Friday, October 23, 2026

Teaching Proofs through Critical Reading by Dana Rowland

 This is the second in a series of five blog posts from members of NE-COMMIT’s Summer 2026 Proofs Literature Reading Group, written in response to the call: For those of us who were in this summer’s group reading proof literature, write a bit about what you got from the group, such as articles you found interesting or plans for the fall. Our audience is NE-COMMIT and anyone else curious about improving teaching mathematics.

Teaching both Discrete Math and Real Analysis this fall, Dana turns the readings into a concrete list of changes, with special attention to equity and to helping students read proofs critically, including ones generated by AI.

The series: The posts move from how the group started, to changes in proof courses, to ideas for non-proof courses, and finally to the theory behind it all.

  1. Proofs Literature Reading Group, by Debbie Borkovitz 
  2. Teaching Proofs through Critical Reading, by Dana Rowland (this post)
  3. Rethinking Proof Teaching: Reading Proofs as Much as Writing Them, by Maria Fung
  4. Possible Uses of Comprehension Questions in Non-Proof Service Courses, by Michael Nathanson
  5. What “I Don’t Understand This Proof” Means: A Cross-Walk of Four Theories, by Ileana Vasu

This fall I am teaching proof in Discrete Mathematics and Real Analysis.  At my institution, many students encounter mathematical proof initially as part of the Discrete Mathematics course (required for math and computer science majors among others) and reinforce their proof writing in Linear Algebra, but really dive in when they reach Abstract Algebra or Real Analysis.  Recently we moved to offering these upper-level courses once every two years, so students now have a wider variation and on average less mathematical maturity when they take Real Analysis than in the past.  Also, the availability of AI encouraged me to rethink my out-of-class assignments and overall approaches to assessment in both courses.

I approached the readings from this context.  Though I appreciated reading the theory behind how students’ engage with learning proof, I was particularly interested in concrete interventions to apply in my classroom. Below, I note my key take-aways based on three of the articles we read, and outline some specific changes I plan to make in my fall classes as a result.

Our first discussion was based on "Lessons Learned about Incorporating High-Leverage Teaching Practices in the Undergraduate Proof Classroom to Promote Authentic and Equitable Participation," by Melhuish, Dawkins, Lew, and Strickland.  This was my favorite of the articles we read due to its focus on creating activities for student-centered approach to teaching proof, with an added emphasis on equity.  While the lessons used in the paper were developed for an Abstract Algebra course, the underlying themes apply to any mathematical course that emphasizes proof.  While reading, I recognized that my approach to teaching proof thus far had focused primarily on supporting students in constructing proof but hadn’t given as much attention to tasks related to proof validation and proof comprehension (which arguably need more attention in the age of AI as a way of getting students to critically examine a generated proof).  Furthermore, my structure for in-class proof-writing tasks didn’t always provide equitable access points for students with varying backgrounds.  As noted in the conclusion (p. 312), “structuring group work to distribute responsibility and providing opportunity beyond just formal proof production” are essential for more equitable participation.

Some implementation action items that I plan to incorporate into my teaching of proofs include:

  • Adjust “task launch” for proof writing (p. 297) to focus more attention on “unpacking key definitions and relationships in theorems” to be more explicit about to prior knowledge students need to engage with a particular task by creating a “public record” of necessary definitions and key ideas.

  •  Give more attention to students’ appropriate use of quantification (there exists an x, or for all x) and identification of mathematical objects and their referents when verbalizing (for example, ask students to “explain what type of object is being referenced by certain symbols” (p. 309).

  • Include activities that focus on analyzing, critiquing and comparing approaches to proof. Having students discuss proofs that are provided to them can provide a different access point (p. 303). The provided proofs might include different approaches from student work from prior semesters and/or AI generated results.

  • Assign content-specific group roles so that each group member is responsible for leading the discussion on a different prompt.

The next paper we considered was “An assessment model for proof comprehension in undergraduate mathematics,” by Mejia-Ramos, Fuller, Weber, Rhoads, and Samkoff, which as the title suggests, focused on the type of questions to ask to assess student understanding.  The authors note that “If comprehension tests only ask students to reproduce a proof by rote, students are likely to develop a superficial understanding of that proof and emphasize form over substance” (p. 4), and propose alternative types of questions to use to assess different dimensions of understanding.  These questions focus on aspects such as the meaning of terms and statements, the status of assertions (is a claim a definition, an axiom, a theorem, a hypothesis, or a deduction from prior steps?), justification of individual steps, identification of proof structure and method, summarizing high level ideas, generating examples, and transferring the method.  The related paper “Comprehension tests in mathematics” by Conradie and Frith provided additional examples of questions to be used for assessment.  Authors noted that a disadvantage to questions of the form “State and prove theorem _____” can lead to students who “try to commit to memory the entire proof, word by word, with possibly no understanding at all” (p. 225), and moreover, are difficult to evaluate. 

Recognizing that many of my proof questions on assessments from past semesters focus primarily on proof construction, my main takeaway is the need to develop exercises that also require critical reading of proofs with more detailed prompts for use in-class, on homework, and in assessments.  In particular, I aim to:

  • Develop explicit prompts to direct students’ comprehension of “local” statements (such as the meaning of terms and statements, the role of a statement as a definition, hypothesis, or justification of a particular claim) as well as “holistic” understanding (ability to summarize the key ideas, illustrate with examples/diagrams, or transfer methods to another context).

  • Add proof comprehension questions to weekly homework assignments (along the lines of the “comprehension tests on ‘unseen’ proofs”, Conradie and Frith p. 233)


A huge thanks to Debbie Borkovitz for launching this group to explore mathematical education research on teaching proofs. Summer is a great time to reflect on what is working in courses and what needs to be tweaked or improved. As always, I am grateful to colleagues in the NE-Commit community for sharing their ideas, their challenges, and their experience.


Proofs Literature Reading Group by debbie Borkovitz

  This is the first in a series of five blog posts from members of NE-COMMIT’s Summer 2026 Proofs Literature Reading Group, written in respo...