Providing sandwiches: optimising feedback at the education picnic


The National Student Survey (NSS) has indicated over several years that not all students in UK higher education are satisfied with the feedback they receive on their assessments. However, the Open University consistently receives high scores for student satisfaction in the area of feedback and assessment. This article provides an overview of the Open University’s model of providing written feedback on student assignments, which may be useful for other institutions. The basic model is to provide students with a sandwich in which constructive criticism is presented embedded within praise for attainment. There is also a focus on general feedback that can be applied to future assessment situations, rather than detailed corrections on assignment tasks that are not necessarily repeated.

Linda Robson

The Open University
Linda Robson has been a Lecturer at the Open University for over 15 years. She has particular expertise in online delivery and support for students with additional requirements.


For many years, there has been much debate within the higher education sector regarding the function of assessment. Is it to test student achievement against the desired learning outcomes, or is it part of the learning and teaching process? Most people would probably argue that assessment fulfils both roles, but to ensure students are able to learn from the experience, it is essential that they are given effective feedback. Mory (2004) argues that the primary role of feedback is communication, and that assessment has a critical role in knowledge acquisition.

Since its inception, the Open University has placed particular emphasis on providing extensive written feedback on assignments. To some extent, this is driven by the fact that part-time distance learning frequently results in tutors not having opportunities to meet students face-to-face. Although other channels are available, assignment feedback may be the only dialogue that a student engages in. As other institutions are increasingly adopting online learning practices and electronic assessments, it is likely that opportunities for face-to-face feedback are reducing, and consequently that there is a need to develop more effective written feedback in order to support student learning. Some institutions are developing a distance learning offering, whilst others are adopting a more blended approach. Giving detailed written feedback to students allows them the opportunity to reflect on their work, and revisit comments. Having a written record of the feedback, and reminding students to revisit this allows for a much deeper engagement with the feedback over a period of time.

The National Student Survey frequently shows that quality of feedback is an area where students are least satisfied (Beaumont et al., 2011; Higher Education Funding Council for England, 2011). As the NSS only surveys students who complete a qualification, the situation may be worse than reported. Feedback is one of a number of factors within measures of quality of learning experience which students who withdraw frequently cite as a contributing factor (Poulos and Mahony, 2008).  Blair et al(2013) suggest that dissatisfaction stems from an over reliance on written feedback. However, the Open University, which has a higher reliance on written feedback than any other UK institution, receives NSS scores for assessment and feedback which are exceptionally high (Gibbs, 2010).

This article highlights features of the Open University feedback model which may be beneficial to aid other institutions in addressing the gap in student satisfaction, and more importantly the potentially missed opportunity for enhancing student learning. Although some individual tutors are utilising recorded oral or video feedback, this article focuses on the use of written feedback only.

Overview of the Open University model

Although there are variations from module to module, the basic model for assessment at the Open University requires students to complete one substantial piece of assessment for each 10 credits of study (Quality Assurance Agency, 2008), followed by either an examination or an examined project. This model for assessment has been maintained throughout the Open University’s 40-year history (Gibbs, 2010). The vast majority of assessment at the Open University is summative, and students are not allowed to ask their tutor to review any of their assignments prior to submission.

Prior to high levels of internet access, students posted hardcopies of their assignments to their tutors, which were then annotated by hand and returned to students through the postal system. This process took up to three weeks. With the adoption of electronic submission, students upload their assignments using a web interface and tutors are able to annotate electronically before returning the marked assignments to their students within two weeks. The adoption of electronic assignments and marking allows tutors more flexibility in both the positioning and length of their comments.  In the days of hardcopy handwritten assignments, tutors would often be restricted to making comments in the margins or on a separate sheet paper. This restricted the level of detail which could be inserted at key points in the student script.

In addition to on-script comments, students also receive a cover sheet that shows a grade and some summary comments from their tutor. Although the vast majority of assignments are summative, they also play a formative role as assessment is designed within each module to ensure that the assignments build upon each other. Students are encouraged to go back and read the feedback on previous assignments as they work through the module. Tutors are able to see if students have downloaded their marked assignments, but are unable to tell if students open the files to access the feedback comments.

The sandwich

The basic format for giving feedback to students on the summary sheet is to provide a sandwich. First, praise the student for something they have done well, next, give them the harder news about where they have not achieved the intended outcomes, and finally finish with more praise. The purpose of this sandwich is to try to achieve a balance between motivating students and encouraging them to stay with the course, whilst at the same time being able to give constructive criticism to highlight any misunderstandings or inaccuracies. It is hoped that after reading the summary sheets, students will go to look at the on-script comments for more detailed feedback on specific points within their assignment.

What is in the sandwich?

There has been much research demonstrating that students generally want and value quality feedback, but frequently there is a ‘significant mismatch between students’ and tutors’ perceptions of the feedback experience’ (Beaumont et al., 2011). Often tutors feel that they put a significant amount of time into feedback that is not engaged with by learners, whilst students sometimes feel that the feedback they receive is not useful. There appears to be a fundamental communication problem between tutors and students, but it is not clear whether the issue is that tutors send the wrong message, or whether the message is not being interpreted well. Chanock (2000) suggests that a key factor is student misinterpretation of comments, possibly due to tutors’ overuse of dense academic language (Blair et al., 2013).

In order to improve the situation, tutors need to consider what messages will be useful for their students, and assess whether the way they communicate their feedback is clear, meaningful and effective, which is particularly important in online teaching. Although remote students may have questions about the feedback they receive, they are less likely than face-to-face students to engage their tutor in discussion about it. In a face-to-face environment there are informal opportunities for students and tutors to discuss attainment and progress, and check if feedback comments have been interpreted fully.  Students who are working remotely need to be more proactive about contacting their tutors, and may be reluctant to do so for fear of ‘bothering’ them. Face-to-face students also have the advantage of being able to discuss feedback with their peers, whilst remote students are less likely to do so. At the Open University, students are actively discouraged from sharing feedback with their peers, even though such conversations and reading of peer papers would be seen as legitimate activity at other institutions.

The ‘praise, constructive criticism, praise’ sandwich model of feedback is designed to encourage students to engage at several levels. The initial hurdle is to get students to read all the feedback that is provided. Only then is there a possibility that students will engage deeply with the feedback, move on to taking positive action, and demonstrate an improvement in performance. The initial praise attempts to put students into a positive frame of mind in preparation for the constructive criticism that is to follow. The final slice of praise serves to remind learners of their positive achievement and boost self-esteem. A student with high self-esteem is more likely to be motivated, and more open to working with feedback, in order to develop their skills and knowledge for the future.

Optimising the sandwich filling

Just applying the sandwich model does not ensure quality feedback. Attention also needs to be given to the content of the three sections for the feedback to be sincere, meaningful and useable. Walker (2009) carried out a detailed analysis of feedback comments provided on Open University assignments. Looking for evidence of feedback that promoted post-assignment learning, she reported that ‘statements of what is wrong, or even of what is expected, are considerably less likely to achieve this reconstruction than comments that include an element of explanation about why students’ answers are incorrect, incomplete or inappropriate, and of why what the tutor is suggesting is more acceptable’ (Walker, 2009, p.3).  This mirrors the findings of Poulos and Mahony, who quote a student from their study of effectiveness of feedback, saying that ‘feedback needs to be provided to you so you can actually make a change…If you can’t make a change from what provided then it is useless’ (2008, p.153).

However, both these statements miss the corresponding and equally important point that students also need to be clear on what has been done well, in order that they can repeat that achievement in subsequent tasks. Merely informing learners that a particular response was good or excellent misses an opportunity to communicate specifically what was good or excellent about their answer. This is particularly the case where a student’s ‘good’ answer has not attracted full marks and there is no explanation as to why marks were lost. Learners need feedback that clearly indicates which elements were good, and how to close the gap in order to achieve excellence.

Whilst grading against marking criteria is required in summative assessment, to give a measure of performance relative to the desired learning outcomes, student feedback can be ipsative, that is relative to previous performance (Hughes, 2011).  High achieving students may be motivated by grades, but lower achievers may find their grades demotivating. The use of ipsative feedback and highlighting improvement from one assignment to the next can motivate weaker students, as they see the progress they are making. The ipsative approach facilitates the praising of students who incorporate previous feedback, whilst giving room to promoting further improvement. For example, a student who has progressed from not referencing to including an attempt to reference can be praised, whilst still receiving low marks for presentation as the formatting is still incorrect.

Ipsative feedback is easier to deliver in an electronic environment than it has previously been in paper based assessment systems. Most electronic assessment systems act as a repository for marked assignments. Consequently, it is easy to refer back to a student’s previous assignment to be reminded of the previous attainment and feedback.  This facilitates the impression of a personal connection with a student, even where class sizes are large.

This approach allows sincere praise that is still compatible with awarding the full range of possible grades. The individual nature of this type of feedback also gives students an impression that they have a personal relationship with the tutor, which is motivating. Tutors should be encouraged to consider which of the desired improvements are most important. Weaker learners are likely to respond better to feedback aimed at supporting them to achieve a moderate level of attainment, more than if they are overloaded with extensive feedback on how to attain full marks. Setting challenging but realistic goals, which a student may surpass, is more motivating than pushing a learner to seek a level of excellence which they will almost certainly fail to achieve. Equally, a high performing student can be encouraged to strive for attainment beyond the level of the required learning outcomes, and given a clear message that there is always potential to develop further.

A student receiving a grade of fifty per cent may consider this to be a poor, average or good performance, depending on their previous attainment. It is important that feedback is in the context of a student’s previous attainment; where there is a big discrepancy between a tutor’s and a student’s view of the attainment, the associated feedback becomes less sincere and it is less likely that a student will engage with it at a deep level.

The focus of feedback comments needs to be carefully considered. Ivanic et al.(2000) found that tutors tend to provide more negative than positive comments. Whilst it is important to highlight errors in order that a student is aware of problems, Knight and Yorke suggest that ‘although many teachers give a lot of feedback on specifics, it is general feedback that has the greater power to stimulate learning’ (2003, p.33). The term general here refers to individual but not task-specific feedback, rather than general whole class feedback which is reported to be disliked compared to individual feedback (Poulos and Mahony, 2008). Balancing positive comments with negative increases student motivation and highlights achievement. Using the sandwich method ensures that feedback is not dominated by what needs improvement because it forces a marker to look for positive aspects of the work to comment upon.

Hughes (2011) has identified that ‘task specific feed-forward is only useful if the task is to be repeated in the short term’ (2011, p.359). Part of a tutor’s role in producing feedback is to identify the most important aspects that they hope a student will engage with. Students are looking for advice on where to focus their efforts in order to have the maximum impact on improving performance. Therefore, it may be prudent to direct learners to improving aspects of their work which may be considered general study skills, and hence valuable for future assessments, rather than specific to a particular task. For example, investment of time in developing academic writing skills is likely to have a greater impact on students’ attainment over their entire learning journey than using time to focus on the specific details of a single assessment task.

Having identified which elements of a student’s performance are to be focused on in the feedback, it needs to be delivered in a style that is accessible to the student. Canock (2000) reports that fifty per cent of students did not understand what a tutor was looking for when they were told that there needed to be less description and more analysis. Such advice needs to be backed up with clear description, or demonstration of what is required. In later assignments, if a student has achieved the objective of including more analysis, this needs to be praised to reinforce learning.

The provision of this level of detailed, personalised feedback can be time-consuming. However, this investment helps to support students to be more motivated, leading to better retention and attainment. Blair et al. (2013) suggest that there is a danger in giving electronic feedback because word processed comments can easily be copied and replicated on many assignments, thus resulting in formulaic and impersonal communication. Whilst this can be a concern, the reuse of selected and appropriate comments that are relevant to an individual learner can be a useful tactic to speed up the feedback process. Using a bank of comments embedded into and supplemented by individual feedback saves time, whilst maintaining a personal connection with students. It can also be useful to embed hyperlinks into feedback in order to direct students to supportive resources. This may be module materials, institutional generic resources, or links to other institutions or the wider Internet. Using hyperlinks allows tutors to be specific regarding supporting materials that they feel students would benefit from engaging with. This is more powerful, and the advice is more likely to be followed, than asking students to find resources for themselves.

The first slice of praise given to students should highlight something that has been done well. Moving on to the constructive criticism section, this should highlight areas which are important for students to addresses in preparation for future assessment tasks. Specific points can be made on the script, leaving the summary sheet available for the type of general feedback Knight and York (2003) highlight as having the power to stimulate greater learning. Selecting a small number of key points for students to work on helps them to prioritise their development and makes the task appear more manageable.

Finally, close with a reminder of what learners have done well in order that they can put the constructive criticism comments into the context of overall achievement. It is important to highlight the aspects of assignments which should be repeated in future tasks, and specifically signpost similarities with other assessment tasks. Throughout the feedback, care needs to be taken regarding the language used in order to ensure that the comments are accessible to students. Terms such as analysis, argument and discussion may not be distinct, and hence will need further explanation or demonstration. It should be made clear to students why the feedback is useful. For example, a tutor might point out to students that learning to reference correctly will result in them achieving higher marks on all subsequent assignments, assuming that academic presentation is rewarded in the marking schemes.

Providing feedback that is accessible and useful to students raises learners’ opinions of the effectiveness and credibility of their tutors (Poulos and Mahony, 2008). This model is not without its critics, who describe it as ‘candy coating’ (Marcus, 2013; Adey, 2004). Obviously, there are situations when a feedback sandwich is simply not appropriate. A student who has performed significantly worse than expected would find it patronising and unhelpful to be praised for an insignificant aspect of their work. When a student has significantly under-achieved, this should be directly addressed.


The basic Open University model is to provide students with a sandwich of feedback, wherein constructive criticism of general points to be applied to future assessment tasks are presented between statements of praise. Positive feedback is offered to all students, regardless of their level of achievement against the learning outcomes, as it is recognised that praise is motivating. Highlighting achievements helps individuals to engage with constructive criticism within the feedback in order to achieve more highly in future assessment tasks.


Adey, P. (2004) The Professional Development of Teachers: Practice and Theory. Dordrecht, The Netherlands: Kluwer.

Beaumont, C., O’Doherty, M. and Shannon, L. (2011) ‘Reconceptualising assessment feedback: a key to improving student learning?’ Studies in Higher Education. 36 (6): 671-87.

Blair, A., Curtis, S., Goodwin, M. and Shields, S. (2013) ‘What feedback do students want?’ Politics. 33 (1): 66-79.

Chanock, K. (2000) ‘Comments on essays: do students understand what tutors write?’ Teaching in Higher Education. 5 (1): 95 – 105.

Gibbs, G. (2010) Dimensions of quality. York: Higher Education Academy.

Higher Education Funding Council for England (2011) National Student Survey Findings and trends 2006 – 2010. Bristol: Higher Education Funding Council for England.

Hughes, G. (2011) ‘Towards a personal best: a case for intoriducing ipsative assessment in higher education.’ Studies in Higher Education. 36 (3): 353-67.

Ivanic, R., Clark, R. and Rimmershaw, R. (2000) ‘What am I supposed to make of this? the messages conveyed to students by tutor’s written comments.’ In: Lea, I.M.R. and Stierer, B. (Eds.) Student Writing in Higher Education: New contexts. Buckingham, UK: SRHE & Open University Press: 47-65.

Knight, P. and Yorke, M. (2003) Assessment, learning and employability. Maidenhead: McGraw-Hill Education/ Open University Press.

Marcus, D. (2013) A Thought on Teaching and Feedback in the ED. EM IM Doc. Available from:

Mory, E.H. (2004) ‘Feedback research revisited.’ In: Jonassen, D.H. (Ed.) Handbook of Research on Educational Communications and Technology. 2nd ed. Mahwah, New Jersey: Taylor and Francis: 745 – 784.

Poulos, A. and Mahony, M.J. (2008) ‘Effectiveness of feedback: the students perspective.’ Assessment and Evaluation in Higher Education. 33 (to), pp.143-54.

Quality Assurance Agency (2008) The Framework for higher education qualifications in England, Wales and Northern Ireland. Mansfield: Quality Assurance Agency for Higher Education.

Walker, M. (2009) ‘An investigation into written comments on assignments: do students find them usable?’ Assessment and Evaluation in Higher Education. 34 (1): pp.67-78.

Creative Commons License
This paper is licensed under a Creative Commons Attribution-ShareAlike 3.0 Unported License. How to cite this paper.
Tagged with: , , ,
Posted in Research Paper

Leave a Reply

Your email address will not be published. Required fields are marked *


You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Subscribe to BeJLT

Get email alerts when there is a new issue.
* = required field

Send this page to Kindle

your kindle user name:
(, without
Approved E-mail:
(Approved E-mail that kindle will accept)
Kindle base email |
(Use to download on wispernet or wifi, use for wifi only.)
using may incur charges)