Designing and Implementing Peer Review
Section Headings below:
- Design Decisions
- Frequently Asked Questions
There are many ways to design programmes of study so that students engage in peer review with feedback. There are also many pitfalls to successful implementation of peer review. In order to make it easier for those wishing to design peer review the PEER project is developing a set of guiding principles for good peer review practice. Also, on this website are examples of the implementation of peer review across different disciplines.
Topping (1998) has produced a typology of the different clusters of variables that might be considered in peer assessment design. These clusters focus on the purpose of peer assessment, the types of peer interaction that might be involved, the composition of feedback groups and the incentives to participate. Recently, Gielen, Dochy and Onghena (2010) have updated Topping's framework and provided a comprehensive inventory of peer assessment variables. This paper might benefit researchers and those wishing to design peer review activities. Examples of peer review are also described in a very useful paper from the University of Melbourne (see right panel).
Peer review in this website is about students making evaluative judgements about other students work and providing written feedback. It is not about students marking each other's work or even rating it, as rating is often perceived as marking by students. In designing peer review one must first decide on the target task upon which reviews will be focused. This should normally be a complex task, that is challenging and that might generate a range of responses. Typically this might be an essay, a report, a design, a case example (e.g. law or sociology). In general, if the task is too simple students will get bored doing the reviews and especially will feel that the feedback they receive is not of much value.
Next there is the review activity. Time should be spent considering why this is being done and what you think students will learn from it. You will also have to consider how you present this to students to inform them why it is part of the course or programme. In the pilots at the University of Glasgow computing science the teacher first explained how peer review worked in the computing industry when writing a new program and the importance of different stakeholder perspectives and then he explained what he described as 'good', 'bad' and 'ugly' feedback.
There are many design decisions around the peer review activity such as:
- Who should carry out the target task - individual students, pairs, groups?
- How many reviews should students carry out? One review is normally not enough, two reviews is perhaps minimum but more reviews might often be better because then students are likely to get at least one good review. However, there is a trade-off in other areas. The more reviews students have to do them time it takes. This can be compensated by a smaller review task.
- How should the review rubric be framed?
- How should students respond to reviews?
- Unfinished section..
Why should I introduce peer review in my course or programme?
Peer review is a powerful way to develop expertise in the discipline. It starts from the premise that making evaluative judgements about work against standards is an important skill. This skill is required in professional practice and for life beyond university but it is not usually explicitly taught in university. Peer review can also help underpin the development of critical thinking and enquiry-led learning. From the literature, different reasons are discernable for introducing peer review. Those most common reason is to enhance the feedback that students receive, to develop their writing skills, to put them in the role of the assessor so they know what the teacher has to do and to develop judgement. [see, Gielen, Dochy, Onghena, Struyven and Smeets, 2011)
Students don't have the skills to mark each other's work
In these web pages we are not promoting peer marking often called peer assessment. Not only do staff have some problems with this but so do students. In two recent implementations reported in this website(Engineering Design and Sociology) the majority of students in an online survey expressed concern about being asked to mark other student's work. Also, if students do mark or even rate the work of others and then the teacher provides marks that are quite different from those of the students then this undermine the value of peer review in the minds of students. The sitution is quite different if students are required to produce written feedback of some kind - a suggestion for improvement, a reaction to what is written etc. Most students respond positively to this and to a process where they see the range of work produced by their peers.
Students don't like to criticise each other's work
Most people think about peer review as criticism or as finding weaknesses in the work of peers but this need not be the case. In peer review students might be asked to summarise the work of peers in a few sentences (e.g. an essay), to make a suggestion for improvement, to suggest another solution pathway, to suggest something that could usefully be added. The list is endless and depends on the discipline, the skills you wish to invoke in the peer review task and the overall purpose.
Aren't there plagiarism issues in the use of peer review?
Peer review is a developmental process - it is a means of helping students to learn. They learn the skills of producing feedback required in professional practice and they learn from the feedback they receive. One could ask the same question about the provision of teacher feedback - if students use this is that not plagiarism? A lot rests on how peer review is integrated into a course and the peer review tasks.For example, if students are writing essays but different groups are writing different essays the same essay them and each student reviews essays that are not on their essay topic this would minimise plagiarism but still leave room for learning transfer. There are numerous ways of dealing with this issue but what is important is not to discourage collaborative learning and undermine student development. If plagiarism avoidance is an over-riding concern the simplest approach is to make sure that the summative assessment (e.g. final work carried out for marks) is different from the peer review activity, although by necessity it will always draw on the skills developed through that activity.
To be continued...