Skip to content Skip to navigation

Giving Feedback

Tomorrow's Research

Message Number: 

Before you say anything at all, remember that your job is to help the author, not to make yourself look good.  Your ultimate measure of success is the degree to which the author walks away knowing what to do next, not the degree to which you have made your expertise apparent. 


The posting below, a bit longer than most, looks at various ways to give effective feedback on written work.  It is from Chapter Eight, Holding Up the Mirror: Giving and Receiving Feedback, in the book, Writing for Scholars: A Practical Guide to Making Sense & Being Heard, by Lynn P. Nygaard. End Edition. Published by SAGE Publications Ltd. 1 Oliver’s Yard, 55 City Road, London EC1Y 1SP Copyright© Lynn P. Nygaard, 2015. All rights reserved. Reprinted with permission.

Rick Reis
UP NEXT: Does Putting the Students in Charge of Their Learning Enhance Transfer?
Tomorrow’s Research
---------- 3,270 words ----------
Giving Feedback
Wielding the red pencil is not necessarily any easier than reading its marks on your work.  Because you have been asked to give feedback, you (hopefully) possess substantial knowledge of a subject area.  Expertise in a subject area, however, doesn’t guarantee that a person can provide useful guidance to another hopeful writer.  That depends at least as much on understanding the writer and the writing process as it does on understanding the content. 
Situations in which we give feedback vary in their formality, the extent to which the participants know each other, the power relationship between the participants, whether the feedback is given orally or on paper and what stage the draft is at.  Acting as a peer reviewer for a journal is an example of formal peer review, where participants are anonymous, the reviewer is in a slight position of power, feedback is given in written form and the work should be complete.  Acting as a supervisor for a student is also a formal relationship, although the encounter may be somewhat informal and often face to face, but here the participants know each other, there is a clear power imbalance and the supervisor works mostly with early drafts.  Reading through a colleague’s work is clearly more informal, the participants know each other and are presumably on relatively equal footing, and feedback may be given either in spoken or in written form, and at any stage in the writing process. 
Step 1: Get an idea of what the author needs
The beauty of a mirror as a feedback tool is that it reflects you as you are – your strengths and your weaknesses.  As a reviewer, you should do the same.  But you cannot begin to give the author a picture of her strengths and weaknesses until you understand not only what she is trying to accomplish, but also how you can be of help. 
Thus, the first thing you need to figure out is what is expected from you, both in your general relationship with the author and in the specific feedback session.  Sometimes, this means doing some groundwork before you even meet with the author or pick up the manuscript.  If you are acting as a supervisor, for example, find out exactly what you are meant to be doing and make sure you and your student share the same expectations.  For example, are you expected to provide an actual research question, or is it the student’s job to formulate one?  Are you expected to provide the student with references?  Or is it the student’s job to track them down?  How many drafts are you expected to comment on?  Some of this is likely to be written down somewhere (perhaps even in a formal contract), but many of these expectations may well be a product of departmental tradition, and thus asking your colleagues is the only way to find out.  Finding out about these expectations will not only protect supervisors from students who expect their supervisors to do a lot of the work for them (such as finding a good research question and all the relevant literature), but will also protect students from supervisors who may either force their students to carry out research that will be of little benefit to anyone other than the supervisor, or refuse to provide comments on anything other than a finished draft.  If you are acting as a formal peer reviewer, ask the journal whether they have guidelines for reviewers, and if they do, pay attention to them.  And if you are reading a paper for a colleague, listen carefully to what they want from you before you start reading.  If they don’t tell you, ask!  Two things to ask about specifically are: “What is this I am looking at?” and “How far along are you?”  There is a big difference between a half-finished book chapter and a ready-to-submit journal article, and you need to know what it is you have in front of you before you can begin to give useful feedback.  If the author wants your feedback on a manuscript they are revising to resubmit to a journal, then ask for the reviewer comments as well, just to get an idea of what the author needed to take into account. 
The second thing to figure out is what the author ultimately wants to accomplish.  One of the hardest things we ever do as scholars is to let go of our vision of how research should be enough to accept that others might have a different vision. You should not be asking, “What kind of paper would I have written?” or “What kind of method would I have chosen?” or “What kind of research question should I have asked?” but rather “What is the author trying to say?” and “Does the research question reflect the aims of the author?” and “Is the method appropriate for the research question?” You cannot criticize an apple for not being a bicycle.  If the author has decided on a case-study approach, you cannot judge her work on the basis of it not being a large-n study.  If you are acting as a formal peer reviewer, you thus need to look for the core argument: find something that looks like the research question, and something that looks like the answer to that question.  (See the text box ‘Active reading’.) [Note: The various text boxes referred to here and below are not included in this excerpt – RR] The rest of your feedback can then be based on the extent to which the author did justice to their own core argument.  Questions you can ask yourself include the following: 
-       What is the author’s stated research question? 
-       How well does the author answer the question? 
-       What is the identified knowledge gap?
-       Is the relationship between empirical data and theory clear? 
-       How clearly is the method or analytical framework explained? 
-       Are terms defined or operationalized as necessary? 
Make sure all your feedback responds to your understanding of the author’s aims – and does not merely reflect your own preferences.  
If you are giving oral feedback as a supervisor or peer, you also have an opportunity to learn a lot by listening to what the author tells you (see the section on the ‘Good conversation’ further on in this chapter).  It is always a good idea to begin each feedback session by letting the writer talk.  One advantage of letting the writer talk first is that it allows him to get all his defensiveness and excuses out of the way so he can better hear what you are telling him.  We are always acutely aware of the weaknesses in our own early drafts; when the person whose paper you have reviewed walks through your door, he is probably feeling apprehensive and vulnerable.  Launching into ‘constructive criticism’ right away will only make this worse.  But if you give him a chance to purge himself – ‘I haven’t really worked on the wording yet’, ‘The introduction still isn’t quite right’, ‘I’m having trouble describing my method’ – he will be more receptive to your comments.  Allowing him to point out flaws in his work before you do will make him feel more capable as a writer.  This is especially true if you can acknowledge his own evaluation as a preface to yours: ‘As you say yourself, your research question isn’t quite there yet.  I, too, thought it was too broad.  Let’s talk about how you can tighten your focus.’ 
Note that there is a big difference between having the other person define problem areas in the paper and letting him inundate you with explanations for why this is so.  ‘I’m not sure whether I should use the theory to analyze the case, or the other way around’ is a helpful point of departure; ‘My dog ate my first draft’ is a distraction.  If the person you are trying to help is pummeling you with excuses, gently but firmly redirect the focus: ‘Yes, I understand that finding good childcare is difficult.  But let’s focus on how to make the best of it.  In the limited time that you have, what do you think you should be working on the most?’ 
Another advantage to letting the writer talk first is that allowing the writer to identify her problem areas for herself gives you a clearer idea of how you can be most helpful. Even if the writer thinks that her biggest problem is her references and you think her biggest problem is a lack of correspondence between her research question and thesis statement, if all she wants to talk about is references, you are better off addressing references first – and perhaps branching out to other subjects afterwards. 
Listening actively to the author can also help you address the problems that arise when authors are unable to successfully explain in a written format what they mean.  (See the text box ‘Active listening’.) If the author can explain to you face to face what he is trying to accomplish, you will be in a better position to judge whether or not he has accomplished it in his manuscript.  Do not underestimate the importance of asking questions: ‘When I hear you talking it sounds like you are most interested in looking at the effect of news broadcasts on voter behavior, but your research question makes it sound like you are interested in television in general.  What is your main interest here?”  While you might not hesitate to ask a student of yours what she is really trying to say, you might feel more uncomfortable asking a colleague – either because you do not want to make your colleague feel bad, or because you are afraid of revealing your own ignorance about something you feel you should know more about.  If there were ever a place to sell new clothes to the emperor, it’s in academia.  Nowhere is the fear of looking ignorant more pronounced – or more crippling.  But if you deny the other person a chance to explain, you rob them of an opportunity to achieve greater clarity for themselves – because there is no better way to understand something than to explain it to someone else.  So it is better to ask too many questions than too few. 
And, finally, sometimes just listening is enough.  Consider the following scenario: a colleague bangs on your door, desperate to seek your advice on something.  She starts explaining the problem to you, and every time you’re about to say something helpful, she cuts you off and keeps talking.  You eventually give up and just stare at her.  You are actually thinking about where to eat lunch, but you nod every now and then just to make her think you are paying attention.  Finally, she gets up and says something like, ‘Thanks so much.  You’ve really helped! I know what to do now.’  You may feel that you have done nothing, but sometimes just putting a particular problem into words makes the solution apparent.  And since most people find talking to inanimate objects overly eccentric, they need a live human being who will allow them to think out loud. 
It’s not always easy to get authors to talk: sometimes, they have so much to say that they do not know where to start.  You can help them get started by asking ‘trigger questions’, which are simply general questions that you can ask, without needing to read the manuscript first, to trigger conversation.  Some good trigger questions include: 
-       What is the most interesting thing you have found so far? 
-       What are you finding most difficult to write about? 
-       What is it you want people to remember when they are finished reading this?
-       What interested you in this topic to begin with? 
No matter how the author answers these questions, you will get some insight into what they are trying to do, and what might be blocking them.  The question, ‘What is the most interesting thing you have found so far?’ in particular is helpful: sometimes academics are so worried about being objective that they do not allow themselves to dwell on their own feelings about their work.  If you ask them what they find interesting, they can express enthusiasm without having to feel like they have to be academic.  You can then check the draft afterwards to see if you can find any vestiges of this exciting find: often it is disguised so thoroughly that you would have had no clue that the author found that part interesting if they hadn’t told you.  Asking authors what they have trouble writing about forces them to put into words what they are struggling with, and making this effort often gives them an idea of how to solve the problem.  The last two questions – what they want their readers to remember and what interested them in the first place – ask the author to take a step back and think about why they are doing what they are doing. 
Step 2: Respond 
Once you know what the author wants from you and have done your best to understand the aim of the paper, you can start giving more active feedback.  Before you say anything at all, remember that your job is to help the author, not to make yourself look good.  Your ultimate measure of success is the degree to which the author walks away knowing what to do next, not the degree to which you have made your expertise apparent.  With this in mind, there are three main areas to which you can address your comments: the content; the core argument or structure; and the general writing (including grammar, style and sentence flow).  
If you have been chosen because you are an expert in the field, all comments you have on content are relevant (unless, of course, they are about how you would have approached the research). Comments on content can include pointing out omissions of important references or theory, questioning an interpretation of sources or data, or otherwise drawing attention to the scholarly or scientific strengths and weaknesses of the work. 
Comments about the core argument or structure are also useful – even if you are not an expert in the field.  Here, instead of drawing on your expertise on the content, you base your comments on your understanding of the writer’s aims: ‘Given that your research question narrows your scope of inquiry to the postwar period, I’m not sure I understand the function of your section on prewar economics.’  Your goal here should be to point out places where it’s not clear what the author is trying to say. 
Sometimes, it’s hard to understand what the author is doing because the author himself is unsure.  In such cases, reading actively and trying to answer the basic questions about knowledge gap, core argument and structure will help you pinpoint the most important problem areas.  Ironically, it is often more difficult to pinpoint problems in well-written papers than in mediocre or outright terrible ones.  A good writer will keep you riding the wave of her prose, making it harder for you to stop and ask the critical questions.  With poor writers, you have no choice but to stop and try to figure out what’s going on. 
Commenting on general writing style is where reviewers should exercise the most caution.  First, at an early stage of writing, comments about grammar and punctuation are seldom welcome – or even useful.  Second, you may simply not be qualified.  Presumably, you have been asked to review a piece of work because of your scholarly expertise in the field, not because of your expertise in grammar or house style.  The deep end of the grammar and style pool is no place for a novice.  Even experts will disagree about whether to use ‘which’ or ‘that’, or whether splitting an infinitive is excusable.  Moreover, grammar, usage and style all vary according to geographical region (e.g., US v UK English) and, more important, according to journal preferences.  You may have learned to always write out ‘per cent’, but the journal for which you are acting as a reviewer may use ‘%’; if the author has diligently read the instructions to authors and replaced every ‘per cent’ with ‘%’, he will not take kindly to a snide comment from you about writing out ‘per cent’.  What makes this especially important is that once an author dismisses one comment as irrelevant or erroneous, he will tend to view the rest of the reviewer’s comments with greater skepticism.  So, if you have told an author to spell out ‘per cent’, when he knows he did it right the first time, he may be inclined to sneer at your valid point that the figures in his tables do not add up.  This does not mean that you have to completely overlook the author’s use of language.  Comments like the following can be useful – and welcome: ‘This sentence is awkward and unclear’, ‘Who is the agent here?’, ‘The referent in this sentence is ambiguous’ and so on.  Even at the sentence and word level, your comments should go toward improving the clarity of the writing, not toward airing your own linguistic bugbears.  It is well within your prerogative to suggest that an article might benefit from professional copy-editing, but leave comments on specific grammatical choices and house style issues to the journal editors. 
The main thing to remember here is that, regardless of what you are commenting on, if you want to move beyond making queries or pointing our problems and into the realm of giving advice, the author will be most receptive to your advice if she feels you have understood her aims.  If the author does not feel understood, then everything you say – even the useful advice – is likely to be disregarded. 
Step 3: Plan the next step 
The final and most overlooked step of every review session is to reach agreement about where to go next.  A long face-to-face session of receiving feedback can be very overwhelming for the author, particularly a student receiving feedback from a professor.  It is very likely that when the student leaves the office, he will no longer remember everything that was discussed.  It is even possible (probably) that at some point he stopped listening to what you were saying because he got hung up on something you said earlier.  So, before the student gets up to leave, have him repeat back to you the main points of the discussion.  Agree on what steps need to be taken and in which order – and have the student write them down.  This also helps them start thinking about what they will do, not just about all the things that might be wrong with their work. 
The same applies to written comments: the comments are likely to vary widely in their importance, but the sheer number of them may make it difficult for the author to sort out which are the most important.  For this reason, most journals request that reviewers not only make specific comments about particular elements in the text, but also write a brief summary indicating the reviewer’s overall impression and assessment of the most pressing problems.
Even if you are meeting informally with a colleague, try to end the session by asking, ‘So, what is your next step?’ At the very least, the author will have to focus attention forward, so that he leaves the session thinking productively about what to do next instead of obsessing about how his current draft isn’t good enough.  The mirror works best as a tool for feedback when it shows you not only where you are now, but gives you a glimpse of where you want to be.