Abstract
Students need feedback from instructors to learn from their mistakes and improve their academic skills. Responding to student writing is a central and necessary component of the instructor–student feedback loop. There are many forms of feedback on written assignments, and not all of them have the same role in supporting student learning. For example, as for written feedback, one primary distinction is between comments made in the margins and comments at the end of the written work. Work has focused on practices of instructor feedback among instructors, and other studies have focused on practices of feedback within a single discipline. Although there have been many studies on the topic of instructor feedback, no studies were found that explored possible differences of feedback across disciplines. The purpose of the study described here was to explore what instructors consider to be the purpose(s)/value of writing feedback (whether it is in the margins or at the end) on written assignments, as well as how instructors from different disciplines conceive of using comments in the margins and end of essays/written work, that is, whether or not they would conceive of them differently, and therefore provide different feedback, according to discipline. Nearly half of the instructors valued the use of written feedback on written assignments as fostering future improvements. In contrast, when analyzing responses across disciplines, identifying specific errors seemed to be more important in making comments in the margins. Further analysis and recommendations are provided.
Feedback on written work
In written assignments, most students need feedback to learn from their mistakes and improve their academic skills. Responding to students’ written work is a central and necessary component of the instructor–student feedback loop. There are many forms of feedback on written assignments and not all of them have the same role in supporting student learning. For example, the feedback practice of writing comments in the margins may serve a different role for the instructor than providing end of essay/written work comments.
Broadly conceived, Carless (2006) defined instructor feedback as a response to student assignments. This could mean comments and annotations in the margins or oral discussion with students after assignment submission (Carless, 2006; Chalmers et al., 2018). Another way to view feedback is as a “consequence” of performance (Hattie and Timperley, 2007). In such a manner, comments include corrective information along with alternative strategies. Feedback can also take the form of encouragement or praise. However, there is some agreement that praise must be informative and provide information specifically on the assignment or learning objective that fills the gap between the learning objective and student comprehension (Hattie and Timperley, 2007).
There is a diversity of perspectives on scope and merits of feedback (Nicol, 2010). One way to define feedback was as input and response. From this perspective, the terms coming from the teacher are input messages for direct guidance, but these messages are “often unclear and deficient in quality” (Nicol, 2010: 502). Conversely, Carless et al. (2011) identified a broader definition of feedback as encompassing all dialogue necessary to support learning. An emphasis on dialogue is often necessary because comments by educators tend to be one direction. Carless et al. (2011) viewed interactive feedback as good teaching, since such outcomes may lead to students learning through the process of assessment.
In contrast, Ferris (2014) was of the view that the term “feedback” should mean or meant written commentary used by the instructor to correct the work of the student. The intent is to positively influence future writing outcomes. Likewise, when Boud and Molloy (2013) analyzed the term “feedback,” they analogized it to “telling” the students what to do or change (i.e. a one-way message). From this perspective, there was an assumption on the part of the instructor that the message would be interpreted the way it was given and then hopefully acted upon. Boud and Molloy (2013) saw the need to clarify the idea of feedback and rethink its place in assessment. “This will enable us to have a more robust view of feedback: one that focuses primarily on the needs of learning rather than the capacities of the teacher” (Boud and Molloy, 2013: 700).
Mulliner and Tucker (2017) use the term “assessment feedback” as the umbrella term to discuss all types of feedback. They also recognized that there is no generally agreed upon definition of assessment feedback, as they call it, and little has been studied about it. Mulliner and Tucker (2017) reviewed the literature, noting that the conception of feedback has evolved. It was characterized in the past as a one-way transmission to the student, but now there is an indication that it is moving toward more dialogic and student-centered conceptualizations. The written, reflective, and dialogic strategy is one such approach (Crimmins et al., 2014). Clearly, the student is now more involved in the assessment and feedback process (Mulliner and Tucker, 2017). Either way, it can still be questioned as to what makes feedback effective.
Effective feedback
The question of what makes feedback effective has led to a discussion in the literature about its scope (Arts et al., 2016; Hattie and Timperley, 2007; Yang and Carless, 2013). In other words, what works and what does not work for instructor feedback. For example, Hattie and Timperley (2007) analyzed 12 previous meta-analyses that included specific information on feedback and its effectiveness. These meta-analyses involved 196 studies. They identified the power and influence of feedback on student learning and achievement, focusing on effectiveness and outcomes of feedback.
For effectiveness, feedback needs to be clear, purposeful, meaningful, and compatible with students’ prior knowledge, to provide logical connections and that the timing must be immediate and relate to specific, clear goals, and provide little threat to the person at the level of self (Hattie and Timperley, 2007). Mulliner and Tucker (2017) supported this and reiterated the diversity of conceptions as to what effective feedback comprises. To be effective, it must be “interactive and dialogic; and also part of assessment in that much student learning is driven by the assessment tasks they undertake” (Carless et al., 2011: 396). For feedback to be effective, students must act on it to improve their future assignments (Giles et al., 2014). The student’s role in using feedback is crucial to the learning process (Jonsson, 2013).
Characteristics of effective feedback are that it should clearly explain the scope of good performance (Nicol and Macfarlane-Dick, 2006). Instructors need to deliver information to the student in such a way as to be easily understood by students and also allow for self-reflection. The feedback must encourage the exchange between teacher and student about the learning while inspiring motivation. Furthermore, it should also allow for opportunities to improve on how the students may have missed the mark. This forward pointing role of feedback intersects the second most important theme, namely, to foster self-assessment. Most instructors hope that students will self-reflect and grasp the basis for the assessment. Instructors must involve students in the identification of the standards that guide their work, which often comes in the form of a rubric or other explicit standards. This reflection will allow students to see if they met the mark on the rubric and then they can self-assess. Gibbs and Taylor (2016) stressed the importance of fostering this self-assessment as part of an important piece of feedback as well. For feedback to be effective, self-assessment is the first step in improving on future assignments (Giles et al., 2014). Finally, for feedback to be effective, it must help shape further teaching for the instructor (Nicol and Macfarlane-Dick, 2006). Others have supported the need for student involvement in the feedback process (Baker, 2016; Boud et al., 2013). For instance, Falchikov (1986) explored the role of group and self-assessments in fostering the teaching and learning process. Gibbs and Taylor (2016) looked at the importance of self-assessment as an integral part of effective feedback practices, especially for online learning. They posited that this method of feedback may allow for students to take more ownership of their learning and to be more self-directed. As can be seen, there are many, different, definitions of effective feedback.
Practice on providing feedback
Just as there is a variety of definitions on feedback, the literature indicates there are various ways instructors provide feedback. Some provide little more than grades, while others provide written feedback. This feedback may be in the margins or at the end of the essay/written work. Still others utilize exemplars as a way to align student assumptions with learning objectives and enhance student learning (Hendry and Anderson, 2013; Hendry et al., 2012, 2016). Spelling errors and other corrections on writing mechanics were at the top when it came to the comments that instructors provided as feedback at the time of writing (Connors and Lunsford, 1988). Later, Connors and Lunsford (1993) focused only on “general evaluative comments found at the end or the beginning of papers” (p. 206). Of the 3000 papers analyzed in their study, the end of essay/written work comments just explained the grade given. Stern and Solomon (2006) completed a follow-up to this study. They intended to replicate and extend Connors and Lunsford’s (1988, 1993) investigations of the types of feedback on student’s work. Stern and Solomon found that instructors were providing feedback on poor writing mechanics, but there was a dearth of comments on how to improve them.
Student perceptions of feedback are a common theme in the literature (Mulder et al., 2014; Pokorny and Pickford, 2010). For example, Carless (2006) explored feedback with surveys, specifically focusing on student perceptions. It was apparent that students were seeking guidance, and that they needed instructor feedback to make necessary changes. Ultimately, most students want to improve and are “interested in tutors’ responses to their work” (Carless, 2006: 225). The instructor’s responses to students’ work were helpful, but the students wanted specific help in making future improvements. This kind of feedback is what would have made the learning process more valuable for the students.
The merits of certain types of feedback from the student’s perspective have been explored (Ferguson, 2011). Jones and Gorra (2013) focused on instructors providing feedback to students willing to take an active role in the feedback process, as well as reasons particular students read and focused on particular feedback. Ferguson (2011) specifically focused on the students’ preference for specific types of feedback, finding that students are not interested in feedback on the more mechanical issues. Arts et al. (2016) support this finding. Students in Ferguson’s (2011) study valued comments on structure and content, specifically on “the key ideas and issues—both of which were considered of high importance right across the sample” (p. 56). This result contrasts with the instructors’ approaches in Connors and Lunsford (1988, 1993), but aligns with that of Li and De Luca’s (2014) review of research on feedback, which revealed that instructors’ practices of feedback showed a lack of information provided to students on what they had to do to improve their performance in subsequent written tasks.
Even though feedback on written assignments is instrumental to the academic growth of students, the literature does not provide a clear definition of what kind of feedback is effective (Boud and Molloy, 2013; Carless, 2006; Carless et al., 2011; Ferris, 2014; Giles et al., 2014; Hattie and Timperley, 2007; Mulliner and Tucker, 2017; Nicol, 2010; Williams and Smith, 2017). The practices of providing feedback via the margins or at the end of written assignments are but some of the many ways that may be utilized by instructors. Some focused on feedback within a single discipline (Carless, 2006; Glover and Brown, 2006; Nicol et al., 2014; Williams and Smith, 2017). No studies were found that explored possible differences in the practices of instructor feedback across disciplines.
Single-discipline research
Most of the research on feedback has been from the student’s perspective. This led to a study exploring the reasons for instructor behavior in providing feedback, one which built “on recent efforts to incorporate the teachers’ voices into conversations on response to student writing” (Ferris, 2014: 9). However, this study looked only at how instructors who taught first-year English courses provided feedback. Studies have explored feedback in other disciplines, however. For example, Glover and Brown (2006) specifically addressed effectiveness of written feedback to students in bioscience classes. Undergraduate education students were the subjects of Carless’ (2006) study. Education students were also the focus of Ferguson’s (2011) study. Nicol et al., (2014) performed a peer review of first-year engineering design students. In an attempt to improve the quality of feedback, Williams and Smith (2017) explored feedback on assessed work in political science and international relations departments.
One of the areas lacking in the research is studying instructors’ practices and perspectives across disciplines. There is, therefore, a need to further explore the value that instructors place on giving written feedback. There is also a need to find out how such instructors across disciplines conceive of using comments in the margins and end of essay/written work.
Three research questions guide the inquiry:
What do instructors consider to be the purpose(s)/value of writing feedback on written assignments?
How do instructors across disciplines vary or otherwise in their practices of giving comments in the margins of written assignments?
How do instructors across disciplines vary or otherwise in their practice of giving comments at the end of the written assignments?
Methods
Participants
Participants were 271 instructors at a large university in the southwestern United States. All instructors in three colleges (business, education, humanities and social sciences) received email invitation to participate. We followed the US National Center for Education Statistics (NCES) taxonomy of education programs to differentiate the disciplines (Morgan and Hunt, 2002). The 271 responses included 180 (66.4%) female and 91 (33.6%) male participants, with an average age of 50 years old (M = 50.77; SD = 11.639). The majority of participants (219) were adjunct instructors, which means they are part-time instructors (80.8%), and the mean teaching experience for the sample was 9.76 years, with a large standard deviation of 7.52 years. As for teaching modality, 200 (73.8%) of the instructors primarily taught online, while 71 (26.2%) of them primarily taught in a face-to-face setting. As for their formal training, 151 (55.7%) had completed a master’s degree and 120 (44.3%) had completed a doctorate degree. There were 65 instructors from the discipline of business, 92 from the discipline of education, and 114 from the discipline of humanities and social science.
Materials and procedures
After obtaining site authorization and Institutional Review Board (IRB) approval, approximately 950 university instructors in three colleges (business, education, and humanities and social sciences) received email invitations to participate. In all, 271 instructors responded, representing approximately a 28% response rate. The email invitation included a hyperlink, which linked to an online survey through SurveyMonkey. An implied consent form was included in the first page of the online survey.
Participants were asked a series of demographics questions, such as age, gender, years teaching, instructor rank, formal training (master’s or doctorate), and college affiliation. They were asked a series of 7-point Likert-type scale questions about their experiences with providing feedback and grading, and they answered a series of open-ended questions about their feedback and grading practices. The open-ended questions were as follows: “What value do you place on providing extensive feedback on written assignments?” “When grading a paper, what kind of comments do you give in the margins?” and “When grading a paper, what kind of comments do you give at the end of the essay?”
We used thematic analysis to analyze the data. A two-cycle coding schema was utilized (Saldana, 2016). In the first cycle, open codes were assigned to meaningful chunks of the text. Memoing was used to capture the coder’s thoughts during the coding process. In the second cycle, axial codes were created by collapsing the first cycle codes into broader categories. Finally, the axial codes were collapsed to create broader themes. Each researcher separately coded the data, and then codes and themes were compared to achieve inter-rater reliability (Morse, 2015). We obtained consensus on themes by analyzing differences in codes and achieving agreement on codes and themes for the final analysis.
After completion of the initial thematic analysis, a second analysis was performed to analyze the responses according to college/discipline. To accomplish this, the dataset was divided into three groups: business, education, and humanities and social science instructors. The originally generated themes were reviewed by college/discipline. A series of tables were generated showing themes for each corresponding open-ended question. We generated percentages to show the portion of data accounted for by each theme. To do this, we counted the chunks of coded data for each open-ended question, as well as the count for each theme for the corresponding question. We then divided the total chunks for each theme by the total chunks for the entire question, which produced the percentages for each theme. As chunks of data may vary in length from a word to multiple sentences, they merely provide an indication of the order of theme prevalence. The percentages were rounded up to the whole number.
Results
What do instructors consider to be the purpose(s)/value of writing feedback on written assignments?
As shown in Table 1, there were six prominent themes that emerged. The theme with the largest percentage was “foster future improvements” (43%). One instructor stated, “The students will keep making the same mistakes if we don’t provide feedback as suggestions for improvement.” This quote shows the instructor’s forward looking orientation. The second most common theme was “foster self-assessment” (22%). The third most common theme was “provide new insights” (17%). These three themes accounted for 82% of the data.
Themes for instructor value of written feedback.
How do instructors across disciplines vary or otherwise in their practices of giving comments in the margins on written assignments?
Tables 2 through 4 show themes for instructor conceptions of feedback in the margins. Specifically, as for business disciplines, Table 2 shows two prominent themes that emerged. The theme with the largest percentage was “identify specific errors” (52%). As one instructor stated, “I mostly mark grammar and APA.” In this case, the emphasis was on grammatical and formatting errors. The second most common theme was “offer corrective suggestions” (30%). For business instructors, these two themes accounted for 82% of the data.
Feedback in the margins: Themes for instructors in business disciplines.
APA: American Psychological Association.
Feedback in the margins: Themes for instructors in education disciplines.
APA: American Psychological Association.
Feedback in the margins: Themes for instructors in humanities and social sciences disciplines.
APA: American Psychological Association.
For education disciplines, Table 3 shows two prominent themes that emerged for instructor conceptions of feedback in the margins. Similar to business disciplines, the theme for education disciplines with the largest percentage was “identify specific errors” (62%). In contrast to the business responses, the second most common theme for education disciplines was “providing encouraging comments” (18%). One instructor stated, “I point out areas of strength.” For education instructors, these two themes accounted for 80% of the data. Whereas the theme of “offer corrective suggestions” was 30% in business disciplines, it only accounted for 11% of the data for business instructors.
For humanities and social science disciplines, Table 4 shows two prominent themes that emerged for instructor conceptions of feedback in the margins. Similar to business and education disciplines, the theme for education disciplines with the largest percentage was “identify specific errors” (64%). In contrast to education but in alignment with business, the second most common theme for humanities and social science disciplines was “offer corrective suggestions” (21%). For example, one instructor stated, “[I offer] comments on the students’ perspective, additional information that may be helpful, grammatical corrections, suggestions for writing.” This comment underscores the range of emphasis intended by corrective suggestions. For humanities and social science instructors, these two themes accounted for 85% of the data.
How do instructors across disciplines conceive of comments at the end of essays/written work?
Tables 5 through 7 show themes for instructor conceptions of this kind of feedback. Specifically, as for business disciplines, Table 5 shows two prominent themes. The theme with the largest percentage was “reinforce and summarize what was done well” (52%). For example, one instructor stated, “I might say thanks for your insights on the topic, provide areas where they did well.” The second most common theme was “reinforce and summarize areas for improvement” (43%). For business instructors, these two themes accounted for 95% of the data.
Feedback at the end of essay/written work: Themes for instructors in business disciplines.
APA: American Psychological Association.
Feedback at the end of essay/written work: Themes for instructors in education disciplines.
Feedback at the end of essay/written work: Themes for instructors in humanities and social sciences disciplines.
As for education disciplines, Table 6 shows two prominent themes. In contrast to business disciplines, the theme with the largest percentage was “reinforce and summarize areas for improvement” (47%). One instructor stated, “[I would provide] overall summary comments which help guide the student’s conceptual understanding of the assignment.” In this way, the instructor provided feedback through emphasizing conceptual understanding. The second most common theme was “reinforce and summarize what was done well” (27%). For education instructors, these two themes accounted for 74% of the data. In addition, a close third was “suggested actions” (26%).
As for humanities and social science disciplines, Table 7 shows two prominent themes. In contrast to business disciplines but in alignment with education, the theme with the largest percentage was “reinforce and summarize areas for improvement” (54%). One instructor stated, “I give an overall summary of paper: feedback on areas of improvement, and adherence to rubric/assignment instructions.” In this example, the instructor provided feedback that emphasized alignment to the learning objectives. The second most common theme was “reinforce and summarize what was done well” (25%). For humanities and social science instructors, these two themes accounted for 79% of the data. A close third was “suggested actions” (20%).
Discussion
Most instructors hope that students will self-reflect and grasp the basis for the assessment. As noted earlier, instructors must involve students in the identification of the standards that guide their work, which often comes in the form of a rubric or other explicit standards. This reflection will allow students to see if they met the mark on the rubric and then they can self-assess. The aim of this study was to find out what instructors consider to be the purpose(s)/value of writing feedback on written assignments. In addition, we compared instructor perspectives on feedback across disciplines. Nearly half of the instructors valued the use of written feedback on written assignments as fostering future improvements. However, as will be explained, when analyzing responses across disciplines, identifying specific errors seemed to be more important than making comments in the margins.
We also analyzed how instructors across disciplines conceived of using comments in the margins on written assignments. Our results indicate that the majority of the comments made on students’ papers in the margins were to “identify specific errors.” We observed this result as relatively consistent for instructors across business, education, and humanities and social science disciplines. Within the errors pointed out by the instructors, the key takeaway is that the comments in the margin were used to show precise errors: explaining a “track change” and asking for clarity on unclear points, as well as pointing out content issues, grammar and mechanical errors, and errors in formatting or citations. The goal of these comments was to show what was wrong but not necessarily what to do next. Stern and Solomon (2006) made the point that instructors tend to put on the hat of becoming judge and critic rather than coach. It is clear that the majority of our respondents were interested in using the margins to do just that.
Making suggestions to correct was what came next for business instructors and humanities and social sciences. Pointing out errors is one thing, but in providing suggestions for change, instructors made comments in the margin that were action-oriented steps in the hope that the students would make corrections in the next assignment. For the education instructors, instead of making suggestions, the second most prominent theme was providing encouragement. They would point out areas of strength, show how the students were making connections to the course content, make a compliment or congratulate, confirm what is correct, or point out an interesting idea.
We analyzed how instructors across disciplines valued using comments at the end of an essay/written assignment. There were mixed findings, as instructors across disciplines had some disagreement on the purpose of comments at the end of essay/written assignment. Education and humanities and social science instructors predominantly used comments at the end of essays/written assignments to reinforce and summarize areas the students could improve. Education and humanities and social science instructors used comments at end of essays/written assignments to a much lesser degree to point out areas for improvement and provide specific suggestions for next steps. In contrast, business instructors tended to use comments at the end of essay/written assignment to focus equally on reinforcing and giving a summary of what the students did well and areas for improvement, with a very small number providing specific suggestions for next steps.
As feedback is in part designed to be “feed forward,” that is, to help students to learn from the feedback that they get and to apply it to their next piece of written work, this pattern seemed to hold for the education and humanities and social science instructors. They used comments in the margin to identify specific errors, and comments at end of essays/written assignments to reinforce and summarize areas for improvement. However, even though the business instructors tended to identify specific errors in the margin, they did not have as much consensus on the purpose of comments at the end of essays/written assignments, slightly favoring reinforcing and summarizing what the students did well. It is not clear from the findings why the business instructors deviated so much from the other disciplines in this regard.
A contribution this study makes is that it provides a comparison across disciplines of instructor beliefs about feedback. First, all three groups of instructors used comments in the margins to identify specific errors; however, there were limited directives on how to correct the errors in such a way as to help students write better assignments in the future. This finding seems to contradict the belief that written feedback is for fostering future improvement. However, the second most common theme for comments in the margins was to offer suggestions as to how the students could correct their work in future or encouraging comments. Second, the findings on how instructors across disciplines conceived of using comments at the end of an essay/written assignments also had mixed results. While education and humanities and social science instructors predominantly used comments at the end of the essay/written assignment to reinforce and summarize areas of improvement, business instructors tended to use comments at the end of the essay/written assignment to focus equally on summarizing what the students did well and areas for improvement. There are many possible reasons for this difference, including disparities at the discipline level, differences in socialization of teachers in each discipline, and divergent motivations of students in each discipline.
There are several limitations for this study. First, the sample limits the generalizability of the results. The sample included predominantly adjunct/part-time, female instructors from just one university within one country. In addition, the sample was primarily online instructors, and there was a small subsample of business instructors in relation to the other two disciplines. Second, there could be many reasons for the apparent discipline differences, and the reason for these differences was not explored in this study. There are several implications for future research. Future research could replicate this study with a mixed-method approach to explore the variance we found for business instructors. In addition, future research could compare dialogic approaches across disciplines, which could help explain the variance we observed with educators providing encouraging comments in the margin. Future research could also explore why instructors prefer to use identifying specific errors in comments in the margin as part of a larger goal of fostering future improvement of students.
Differences uncovered in the study described here may perhaps be due to differences in the disciplines themselves, or the socialization of faculty into these disciplines, and so this needs further exploration. It may also be the case that their prior experiences or the motivations of their students may account for the differences, and so this needs to be explored. As data were gathered from only one university in one particular country, future work is needed about the use of feedback in different contexts/countries given that this is context specific. Work is also needed to look at the differences between males and females as, in the study, there were many more females than males, and also to look at the issue of tenure (in the study, the majority were adjunct/part-time faculty, not full-time) and to look more closely at those who teach not online but instead in the face-to-face classroom environment (nearly three-quarters of faculty surveyed taught online, only). These findings could be compared to those gathered from other disciplines, and work is needed to explore this in other contexts/cultures, including in other countries, given that feedback is context specific.
Instructors hope that the feedback they give leads to improvement and understanding on the part of the student. This in turn can be a teaching tool to further instruction. However, there seems to lurk a frustration in that that there is something missing: the student’s part. If the student does not read and then act upon the feedback, then it may be time wasted. How to encourage the students to read and act upon it is something that must be further understood if we are to best help our learners in making use of the feedback that we provide.
Footnotes
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.