What is Feedback?
Feedback is a central concept in language teaching. Our students’ language development occurs through communication, and all communication is predicated on the concept of feedback. The Open University in the UK has a page about how communication works available through a Creative Commons license (learn more about how those licenses work in Part 1 of this series). Let’s start with this image as a definition of communication.
If the Sender sends the message, but the Receiver doesn’t produce any feedback, then the Sender has no idea whether the message has been understood as intended. For example, let’s say I ask my partner to pick up dinner on the way home, and they respond, “Sure. Maybe from that new pizza place?” That feedback tells me that they have understood my message and have ideas to further communication. However, if they respond with, “Sure. Let’s meet at that new pizza place.”, that feedback tells me that my partner has not understood my message in the way I have intended it, and I need to clarify or try to figure out where the message got lost.
I would venture that being aware of whether or not I, meaning the teacher, am communicating effectively with students is more essential in the language classroom than in any other subject area because of the limited language proficiency of students. I typically teach first year language courses and my students usually come in as new beginners learning their first words in this new language with me. We use words, images, gestures, and anything at our disposal to communicate effectively! As students progress through the program, proficiency levels tend to grow at different rates, meaning that higher level courses will tend to have a broader range of language levels in one classroom leading to some communication mismatches. Communicating well within the constraints of language proficiencies is a hard skill to develop, and language teachers have to learn that skill on top of all the other teaching skills.
So, how do we know if students understand us when we speak in the target language? We look and listen for feedback from students. Anything that students do to show us that they understand the message is feedback that they give us. If I ask a question, I am sending a message. If students are able to respond “yes” or “no” appropriately, then they are sending feedback that the message has been understood. When I don’t get any feedback or perhaps I get feedback that indicates a misalignment between what I intended and what they understood, I have the opportunity to revise my message to make it more comprehensible. (By the way, that realignment process is called negotiating meaning.) Student feedback drives my next message, which then becomes feedback for my students indicating that I have understood their message, and on and on it goes in a communicative spiral.
How Is Communicative Feedback Related to Corrective Feedback?
Keep this definition of feedback in mind for the next conversation as we lean into the concept of feedback as a part of a teacher’s assessment practices. In communication, feedback indicates to the Sender that the message was received and moves the conversation further along. Now, imagine a classroom setting in which a student sends a message in the target language, perhaps trying to express that they enjoy pizza for dinner, and the instructor responds, “The subject and verb of your sentence do not agree.” Is that response “feedback”? It does not respond in any way to the meaning that the student intended to convey. It only responds to the form of the message. Therefore, while the student might have been trying to initiate true communication, the instructor in this scenario brought the communication to a halt. Often, when we think of the role of the teacher, we think of corrective feedback, pointing out errors or catching mistakes, rather than communicative feedback that negotiates meaning or moves communication along.
This is a good time to pause and reflect: when your students are trying to communicate in the target language, is your feedback for them usually related to the meaning of their message, i.e. communicative feedback? Or to the form of their message, i.e. corrective feedback? (I am going to insert a lightbulb image as a cue to reflect, because I really want you to take a minute and think on this question before you keep reading!)
When we start to think of feedback as a way to point out mistakes rather than a natural part of the communicative process, then we are no longer communicating in any meaningful sense. When we attach the concept of feedback to punitive measures that draw students’ attention to errors, then errors become more important than communication.
Let’s think this through. Students are, as a group, smart and efficient. They know they are being graded. They know that their future acceptance into med school or into their first-choice internship depends on their grades.
So, if students know:
- They will be penalized for errors.
- Their grades depend on their ability to avoid errors.
- The more creative they are with language and the more risks they take in communicating, the more errors they are likely to make.
Then, it follows that students whose grades are based on errors are going to be motivated to minimize risks and creativity and maximize error-free communication.
This is literally the OPPOSITE of what I am trying to cultivate in my classroom. I want my students to engage, connect, ask, answer, debate, and find joy and creativity in the language they are learning. I do not want them to see language-use as a minefield where they must avoid getting zapped with every misstep. So, how can I provide both communicative and corrective feedback in ways that help promote engagement, confidence, and community?
Error Correction
The question about when to provide feedback only *in the communicative sense* and when to correct errors in form has been the source of much debate among teachers and researchers in language education. In 1978, James Hendrickson battled against the prevailing beliefs in the field, namely that communication was like virtue, and errors were like sin to be weeded out immediately lest they grow and spoil the entire utterance. One of my favorite dialogs in language education journals was kicked off in 1996 by the publication of the article entitled “The Case Against Grammar Correction in L2 Writing Classes” by John Truscott, Professor in the Department of Foreign Languages and Literature at National Tsing Hua University in Taiwan. As is customary in academic journal articles, Professor Truscott made a strong claim in his article that grammar-oriented feedback is not useful and, in fact, can be harmful in L2 writing instruction, and then he presented evidence to support his claim. At the time of publication, I think it is safe to say that this claim was not widely accepted by either researchers or teachers, although there were research findings to support it. You can imagine that this (intentionally) provocative article ruffled many feathers. In fact, it spawned many of my very favorite kind of journal article: response articles that directly challenged or supported Truscott’s claims. Perhaps the most cited of the Truscott response articles is the 1999 response by Dana Ferris, a Professor in the University Writing Program at the University of California, Davis called “The case for grammar correction in L2 writing classes: A response to Truscott (1996).”
(This feels like a very good moment to pause and delight in academic publishing and journal articles because they are an academic discipline’s primary medium to hash out ideas over time. One author makes a Big Claim and provides evidence cherry-picked specifically to reinforce their ideas. Then, another author comes along and absolutely eviscerates the first author’s ideas with a whole slew of additional, conflicting evidence. Because, y’all, there is ALWAYS conflicting evidence. This stuff is too complicated for any one Big Claim. Then, more researchers catch inspiration and do more research, then point out in future articles how one or both of the original authors missed nuance or made flawed assumptions or whatever. The process of coming to a consensus is long and arduous and happening all the time in every journal. I greatly enjoy reading, synthesizing, and interpreting the academic body of knowledge for teachers who might have less time and interest for academic beefs, diss tracks, and rivalries than I do: a role I call the teaching methodologist.)
Another wonderful thing to come out of so many decades of research on error correction is the detailed and nuanced terminology to help us make sense of small differences in how we provide feedback. Some of these small differences turn out to have big effects in practice and are worth knowing, no matter which side of the debate you fall on.
*Focused versus unfocused feedback. If you are a teacher and have ever given grammatical feedback on a piece of student writing, you have definitely found yourself wondering exactly how much feedback to give. Should I point out every single error? Even if it is something we have not learned yet? Even if the student was really trying to be creative and express themselves? Even if they mostly use that structure correctly but get it wrong only one time out of ten? This conundrum is the question of focused versus unfocused feedback. In unfocused feedback, you correct everything you see. This level of feedback can be exhausting for the teacher, overwhelming for the students, and, fun fact, some research on this kind of feedback shows that teachers often give unnecessary or even incorrect feedback when they try to correct everything. On the other hand, with focused feedback, you decide prior to reading student work what kinds of errors matter enough to point out, then you only give feedback on those errors.
*Implicit versus explicit feedback. Some feedback on form happens in all communicative contexts. For example, if my colleague calls a conference room by the wrong name (a vocabulary error), I might ask a clarifying question about which room they are referring to. “Sorry, Jane, do you mean conference room A or conference room B?” If my young child tells me they go-ed to the store (a grammatical error), I might recast the sentence back to them with the correct verb. “Yes, we both went to the store.” This kind of form-focused feedback couched in the communication cycle is called implicit feedback. This is feedback that we give as part of the conversation, to clarify misunderstandings or address errors while moving communication forward. On the other hand, explicit feedback involves letting the speaker know that they have committed an error, often followed by a metalinguistic explanation of the error and/or how to fix it.
*Performance versus proficiency. Proficiency is a students’ ability to make meaning with language in any circumstance with no preparation. So, proficiency is a student’s real-world language ability. Performance, on the other hand, is what a student is able to do based on what they have learned. With preparation, a student can perform on specific tasks in controlled settings even when those tasks might not be within their grasp in a naturalistic setting with no preparation. Classroom assessments typically measure performance, not proficiency. Research settings tend to be even more controlled and performance-based than classrooms. Overall, this means that almost all the research is telling us what students’ level of performance is after feedback, not whether feedback improves proficiency, which is a much more complex question.
The Consensus on Error Correction
So, you might be asking yourself at this point, was Truscott right? Is error correction harmful to my students? Should I focus only on communicative feedback? Or was Ferris right and there is evidence to support error correction as a useful part of language learning?
As it turns out, they both have good points, and almost 30 years later, the question is way too complicated for an easy answer. Here are a few things that seem to be true. I would even go so far as to say that the below conclusions are consensus-ish opinions in the field of language education, meaning that almost all researchers would agree on these conclusions.
1) Error correction helps reduce near-future errors in performance.
There are studies that demonstrate that focused, explicit error correction related to specific forms helps prevent students from making those same errors in near-future assignments under the same testing conditions (checkout Shaofeng Li’s 2010 meta-analysis for examples). So, if your goal is for your students to reinforce specific grammar structures and produce fewer errors when using those structures in their next writing activity, error correction seems to help with that!
It is not at all clear that error correction helps to produce deep, subconscious language acquisition or writing development that persists over time. Most of the studies on written corrective feedback, in fact, only measured short-term gains in highly controlled conditions, and what we know about long-term acquisition isn’t promising. Frankly, the question is so complicated that research will likely never give us a simple answer, and the high versus low versus no interface debates rage on. (The interface debates are another long-running SLA war about whether declarative knowledge can become automatized and procedural under any conditions. In my view of the literature, the answer seems to be sometimes in some ways, but not usually, so don’t count on it.) Based on findings that the effects of explicit correction seem to diminish over time, it seems unlikely that correcting grammar forms contributes to meaningful changes in language proficiency, only in immediate performance.
Where we can get into trouble with this literature on error correction is when we imagine that reducing errors is comparable to improving overall performance. Having fewer errors is NOT the same thing as improving writing/speaking or developing more complex language skills. If you want to develop better writers, then your feedback to students should center their ideas, not their grammar erros. If you want to build underlying linguistic competence, then you have to engage classroom strategies that promote level-appropriate communication. To foster more communication, your teacher feedback should be more communicative than corrective.
2) Some kinds of error correction may also have unintended effects.
Another thing to keep in mind is the effect error correction can have on communication, such as in the example I gave earlier where the teacher halted a communicative event to give corrective feedback on verb conjugations. What effect does it have on students who are trying to engage in authentic communication if the teacher response is primarily corrective rather than communicative? Language teachers often discuss amongst ourselves how pointing out errors can make students feel less confident or less willing to take risks, and in those cases we are usually talking about explicit oral corrective feedback.
I can still remember many years ago when I was having a lovely professional conversation in my second language with several colleagues. After I made a point, one colleague responded to me by noting, right then in front of the whole group, a grammatical error I made and how I should have phrased my point instead. I was mortified. I was an adult with a lot of language confidence and pretty thick skin, but my mind went totally blank, and I was no longer able to meaningfully participate in that conversation. I often think back on that moment when planning my own error-correction strategies. I don’t want my students to be steamrolled by error correction or feel belittled by me as the corrector. Many arguments against corrective feedback are specifically geared towards protecting students’ psychological safety and confidence to experiment with language.
3.) Error correction seems to be more beneficial for more advanced students.
Novice and intermediate speakers and writers get less from error correction than advanced students. At the earlier levels, students’ brains are busy building an entire new language system from the ground up. They need lots of opportunities for authentic communication and time to develop. If you are working with more proficient students, you might find that they are more able to make use of error correction, and it is a better use of your time. And speaking of time…
4) Error correction’s benefits must be weighed against the amount of instructor and student time and focus they require.
Think about the goals for your classroom. My goals for my novice students include getting lots of interesting input, engaging in authentic communication with me and with peers, and developing increased intercultural competence and a willingness to communicate across difference. For me and my learning goals, error correction doesn’t help me get there. A little bit here and there to improve accuracy has some benefit, but I don’t want error correction to monopolize my time or my students’ focus in or out of class. If you are like me, time is a scarce resource in your classroom. Based on the meh-at-best research findings on the benefits of corrective feedback, decide up front how much of your teacher time and energy you want to spend on feedback, and then only give feedback that gives students the best chance at meeting the learning goals.
5) Acting on error correction improves its effectiveness.
Written error correction is only useful if students are reading the teacher’s corrections and using them to correct/revise the work. Since students only tend to correct/revise when required to as part of the class, teachers might consider giving the error correction on an early draft of the assignment that students are required to correct and resubmit.
Correcting errors on the final version of an assignment that students will not revise and resubmit is unlikely to serve any developmental purpose. If teachers are correcting errors on final drafts, the most realistic purpose is to justify the final grade the assignment received, rather than to actually promote long-term learning of the forms in questions. (See Hyland & Hyland, 2006)
6) Students may feel better when they are getting error correction, whether or not it helps.
One interesting finding in research across a range of disciplines, not just in second language acquisition, is that students often report preferring a strong teacher presence, error correction, and a clear path to success. In L2 research, learners have expressed a preference for teacher feedback over peer or self-correction and have even told researchers that they wished they were getting more error correction. This is an interesting phenomenon. I can easily imagine a situation in which the teacher believes that error correction will not produce deep, long-lasting language learning, but students want error correction because it helps them feel more control. How should the teacher respond? Is there any harm in focusing on forms when it gives students some sense that they can consciously improve their performance and do not have to rely exclusively on mysterious, subconscious processes to learn a new language? Or perhaps the teacher should educate students about how language proficiency works and convince them to focus on higher-yield learning strategies? (Such as with “Path to Proficiency” posters like these from Memphis-Shelby County Schools.)
If you want to work with students to deliver feedback that meets their perceived needs, a few approaches to finding out what sorts of feedback students what suggested by Hyland & Hyland are cover sheets, author autobiographies, and questionnaires. However, by far the coolest example of this I have ever seen is Meredith White’s (@techMEWithyou ) student questionnaire where they get to choose what kind of feedback they want.
So, What Kind of Feedback Should I Give?
In theory, I tend to fall on the side of letting language develop naturally without extensive error correction. Errors are just signs that our students’ language development is in progress. Seeing students take risks and try to say interesting new things in their new language should be thrilling, not be punished with points off their grade. I admire teachers like Lance P who prioritizes proficiency over performance at every opportunity.
This is where I will be vulnerable and tell you that despite my beliefs, my teaching reality is complicated. As a teaching assistant in my graduate program and later as an early-career instructor in the classroom, I was taught to see errors as impediments to learning rather than as a natural part of learning. The clear message I internalized was that errors should be immediately and firmly dealt with before they could “fossilize” and become permanent features of my students’ speech. I spent years giving detailed, unfocused, explicit feedback to students on every writing assignment, and then giving students grades based on how well they reproduced memorized structures with no errors. Over time, as my own understanding of proficiency versus performance grew, I realized that my students’ overall proficiency mattered to me more than their performance on any given task. However, as recently as two years ago teaching Spanish 101 in a college program, I was not allowed to create my own tests or determine how to assign final grades to students. This is so common in language programs! We often have to follow coordinated syllabuses created by people with different understandings and priorities.
So, my own practices related to feedback and error correction have always been in progress and always subject to external constraints. Knowing that students will need to produce error-free work at various points in order to advance, for the last fifteen years or so I have maintained what I think is a balance of my proficiency goals (making meaning without worry about forms) and my performance realities (fine-tuning grammatical accuracy to ensure it meets departmental standards). I try to choose concrete, low stress, and efficient ways to give students a limited amount of actionable feedback, often through rubrics on their writing or speaking performance (more on rubrics in the next installment of this blog series) or in-class retrieval practice activities. And, I also have to admit, I find that some error correction is helpful to keep us comprehensible to each other, especially when it is focused or implicit. So, I try to keep corrective feedback short, communicative, and confidence-building.
So, now I’m curious! How are you giving feedback? Do you have guidelines that help you manage feedback? The comments are open below or we can chat on Twitter @staceymargarita.
If you want to read other installments in the “What’s the Big Idea” blog series, click here for the list.
__________________________________
This work is licensed under Attribution-NonCommercial-NoDerivatives 4.0 International
Stacey Margarita Johnson 
Leave a comment