Peer Review Techniques

Sharing and Responding, by Peter Elbow and Pat Belanoff, remains the most important and influential guide to respond to peers’ writing. Although designed for the humanities, the techniques found in that guide can be adapted to many disciplines, helping reviewers generate and receive substantive, meaningful responses to drafts. Compiled by the GW University Writing Program's Phillip Troutman, this resource offers adaptations of Elbow and Belanoff’s terms and ideas, arranged roughly by stages or phases in the writing process. Try these with roommates, friends, classmates, peers. Mix, match and adapt at will.



For a designated time, just listen as the writer reads or talks out ideas. You can nod, or look puzzled, but you cannot talk. This is about letting the writer think out loud.


Restating in Question Form

Restate to the writer, in your own words, what you think the writer is getting at, phrased as a question. You may make inferences or guesses, but you should try to stay true to the writer’s ideas. The question format helps the writer reflect on and perhaps question what they’re doing in the draft, rather than only defend it.

Examples to try:

  • “So, are you arguing X? — but what about Y, which you also mention?”
  • “So, Chabanov defines the concept of Z as . . . — am I right?”


Descriptive Outline

This is restatement at the global, organizational level. Outline the entire paper, dividing it into sections and naming the function of each paragraph. For example:

  • Section A is intro: ¶1 provides a hook & defines the problem, ¶2 summarizes prior scholarship & its limitations, ¶3 says how you plan to address the problem and maps out the organization.
  • Section B lays out your methods: ¶4 gives key definitions, ¶5 defines source parameters.
  • Section C deals with the first major evidence: ¶6 quotes and summarizes the source, ¶7 analyzes and interprets it, ¶8 brings in a counterclaim and responds to it.

Don’t give advice here, only a summary of what you think they are trying to do in each section and paragraph.


Centers of Gravity

Identify what felt to you the weightiest, stickiest, most important, most interesting, most striking thing about the draft. It could be anything: the central claim, the use of a particular piece of evidence, a scholarly definition, a particular sentence or turn of phrase that struck you as beautiful. This can lead to insights about what the writer is doing most successfully or insightfully, things to develop further in revision.


What Is Almost But Not Quite Said

This is a great way to tease out larger claims and signal to the writer where you need more information, more explicit logic, or just further elaboration of the claim they are developing. These can be small matters of fact — “You say the dictator was ‘elected’; by whom and how?” — or larger issues of claim — “When you say ‘these images read like a text,’ do you mean they actually have a specific vocabulary and even a grammatical structure? What does that actually look like?”


Think Aloud

This works best with audio recording, but can also happen in person. Read a few sentences or a paragraph at a time, silently or out loud, then stop periodically to describe what you are thinking. These may be statements of understanding or summary, or they may be questions, ideas or other comments. Thinking aloud can also help with organization; if a question is answered later but the reader is asking it earlier, maybe the answer should move up in the paper. (Duke University’s Reader Project draws heavily on this method.)



Pretend to believe everything in the draft, but try to give the writer more ideas in this direction: better examples they could use, scholarly concepts or arguments that seem to align with their thinking and ideas on helping them convey this more effectively at the sentence level.

The point is to cheer them on and help them out, but also to highlight the incompleteness of this kind of positive-only feedback.



Pretend to doubt everything — major claims certainly, but even seemingly insignificant statements. Suggest alternate interpretations, counter-examples and different word choice. Demand verification and more detail. The point is to raise questions the writer may need to address explicitly in some way, but also to decide what level of response may be needed for each kind of question.

For example, some little-known facts may simply need a good citation, but perhaps the central claim needs to accommodate more contradictory evidence known to readers. The “pretending” part is important: you’re not attacking the writer, just playing devil’s advocate for them.


Criteria-Based Feedback

Here you are following a checklist or rubric provided by your professor, looking for these elements in your peer’s draft. For example, the components of the introduction may be assigned (as was done in the Descriptive Outline section). Or there may be specific steps to analyzing the evidence, evaluating options, making recommendations, etc. You may be simply identifying the presence or absence of these elements, or you may be asked to comment on them. For example, you may be asked to reply to the logic connecting evidence to claims.



Is the writer using style that seems appropriate to this discipline, topic, course and especially this particular genre of writing? Who is the audience? What is the level of formality of the prose? How much individuality or creativity is allowed, encouraged, or discouraged? Is specialized vocabulary (jargon) encouraged or discouraged? Does the writer follow the stylistic standards established by the professor for this piece of writing? E.g., is first person allowed, encouraged, prohibited? Is the passive voice preferred (as in some technical fields) or considered obfuscatory (as in history writing)?



Try to separate style choices — which may be determined by the discipline, genre or even your editor (or professor) — from grammar rules that apply everywhere, such as subject-verb agreement, article usage, commas, punctuation, spelling and citation format. You can mark grammar errors in early drafts, especially when they confuse meaning, but they are often best addressed separately in final polishing of the draft for submission. See Stanford University’s Top 20 Errors in Undergraduate Writing. To help separate actual rules from grammar “folklore,” see Joseph Williams’ Style: The Basics of Clarity and Grace, chapter two.

Pragmatics of Peer Review

There are many ways for instructors to assign and configure any of the lessons in Sharing and Responding, and many ways to support student work along the way. These may be disciplinary or they may play to your own idiosyncratic strengths and preferences as an instructor. Some tips and issues from the WID program that instructors may want to consider:

"Just in time and just the thing"

Decide what the most important moments are for peer response and which kinds of peer response would be most useful for each of those moments. What disciplinary knowledge can students pick up quickly and use to respond to peers’ work? What work can they do that allows you to focus on something else in your own feedback? How should they respond to each other’s brainstorming? To early drafts? To near-final drafts?

Assignment Sheets

Create a separate assignment sheet for each peer response, describing exactly what you want students to focus on and what format to use. If you have a checklist or rubric, include it. If you want them to mark the peer’s draft in certain ways, specify this. Also include format for the exchange and for your documentation. If graded, what are your criteria? If students evaluate their peer’s response, what format does that take?

Formats of Exchange
  • Oral feedback in class. This works best with short writing assignments or with brainstorming, topic development and proposal phases of larger projects. This also works with oral presentations, but respondents need specific instructions and you’ll need a way to ensure everyone participates.
  • Creative formats. Prompts might include paragraph shuffle, describe your peer’s evidence or brainstorm your peer’s introduction or conclusion.
  • Written feedback in class. Feedback should be targeted, focusing on one or two specific elements or aspects at a time. Criteria-based feedback can work well here, as does final editing.
  • Oral feedback outside class. This requires a reporting mechanism. Perhaps students submit a brief description or list of what they discussed and what questions their peer raised for them. Perhaps they have a checklist they complete during the session and submit to you. They can also record audio feedback and send it as an mp3 file.
  • Written feedback outside class. Can take place electronically, using comments tools in Word, Adobe Reader or Google Docs. You can use Blackboard’s discussion board for file exchange (PDFs work best) and even have Blackboard keep track of threads for you to check or grade. Some draft/feedback could simply be cut and pasted into the threads, with no attachments. Blogs, wikis and other tools can also work.
  • Peer/professor meetings and office hours. Assign written peer responses as preparation, then have both peers meet with you and use that written response to lead discussion together about the draft. Meetings could also be individual, with the writer bringing their peers’ responses.
Forming Groups

Trios give students two different perspectives, confirming some issues or generating genuine choices from diverse responses. Trios also build in a safety net in case one peer drops the ball. Pairs can work if both reviewers hold up their ends of the bargain.

You can assign peer groups or let students choose. You can make them commit for the term or require students to change groups for each response. Anonymous feedback can be valuable, but it can also generate unnecessarily harsh critique. In-person feedback likewise can be too light. Students need explicit instruction and limits in either case.


Be explicitly clear about what you want, and explain why. It can help to ban specific phrases (e.g., “nice flow,” unless they explain exactly what they mean by that) or certain types of response for specific moments of the process. For example:

Generally ban “advice,” except in criteria-based responses. Instead, keep readers in the mode of being readers, focused on what they perceive, how they are responding and how the writing is affecting them as a reader. Any advice should be offered only speculatively, in question form, and couched in terms of the readers’ perception. Some examples:

  • “I was confused by this term; in this context, does it mean X or Y?”
  • “I was uncomfortable with the word ‘slave’ here because it hides the person’s humanity, though I know you don’t mean to do that.”
  • “I was wondering about X early in your paper, and then you address it later. Do you think it would work better up front? Or is there a reason to save it for later? Maybe I need to know early on that it’s coming later?”

In early draft stages, ban focus on grammar. Peer readers may silently mark typos they spot along the way. The professor can do the same and also address grammar individually with each student as needed. If syntax creates confusion, those questions will arise with the techniques discussed previously. If style questions arise — first person, passive voice, etc. — address them in the draft, since those affect interpretation and communication. Later, students can focus only on correcting typos and grammar (and checking citation style), banning all other kinds of feedback.

Evaluating Peer Response

There are many possibilities: completion credit, criteria-based grades and grading that takes into account evaluation by the peer who received the feedback. It can be a separate grade or a required component of class participation. The key is to signal its value to students in specific ways.