Which of the following statements best describes the benefit of using peer assessment in the classroom?

Peer Assessment

Peer assessment is a process by which learners rate their peers, and is, as such, of great relevance to teacher development. It signifies the joint collaboration by those involved in the learning process in the appraisal of their own learning. In a peer-assessment arrangement, the learners consider “the amount, level, value, worth, quality or success of learning of peers of similar status” (Topping, 1998). Peer assessment is not only a direct appraisal of what has been learned (outcomes) but also of the how of learning (process).The supposed beneficial effects of peer assessment are not only diverse, but also inconclusive (Falchikov, 2005). Peer assessment (and its related format: co-assessment, that is, mentor/assessee) is said to help learners develop meta-cognitive skills, for example, communication skills, self-evaluation skills, observation skills, and self-criticism (Havner and McDowell, 2007); and this may lead more readily to acceptance of feedback. However, supposed effects of peer assessment for learning vary considerably. The findings range from better attendance, learning gains, impact on the ability to self-assess, developing critical thinking, to no effects at all (Topping, 1998). Peer assessment in essence is a social appraisal process where feedback is given to and received by others, aimed at enhancing the performance of the learner'. Therefore, interpersonal and interactional processes play an important role, such as psychological safety, value diversity between peers, interdependence in social relations, and trust in the other as an assessor. Framing features in the arrangement of peer assessment might condition how peers step into the process of appraising each others' learning results. A first set of framing features has to do with specifying the contextual arrangement of the assessment, that is:

the why, that is, reasons for utilizing peer assessment;

the what, that is, objectives, teaching areas, and products/outcomes;

when, that is, time;

where, that is, place; and

how, that is, is it supplementary to grading or required; compulsory or voluntary?

A second set of framing features considers the interaction among peers in the appraisal; because of the interpersonal factors mentioned, the assessment might vary with respect to who assesses whom. This directionality in peer assessment can be one way (from assessor to assessed), reciprocal (peers assess each other, e.g., in pairs), and mutual (all peers assess all peers). In addition, peer assessment may differ in level of privacy (anonymous, confidential, and public) and nature of contact between assessor and assessee (at a distance or face to face).

A third set of framing features refers to the composition of the peer group that provides feedback – it can differ in ability or not; its constellation can vary or not.

In teacher education, student teachers often work and practice together during practice teaching/mentoring. Also in professional development programs, teachers work together as colleagues and share learning experiences. This setting provides a platform for peer assessment in which the ‘learners’ appraise each other as critical friends (Edwards et al., 2002).

Peer- and self-assessment are key skills because they enable students to self-monitor and self-regulate learning. These approaches help students make judgments with regard to their peer’s and their own standard of work. This helps students engage with expectations of quality for specific pieces of work and begin to understand the processes of assessment that teachers use. While peer- and self-assessment can save the teacher time in marking, considerable time is required to train students in these techniques. In some instances, this approach is used formatively to guide future learning, while in others it is used as an alternative to summative assessment.

Peer assessment

In peer assessment, colleagues or student teachers assess each other and provide each other with feedback (Boud et al., 1999). Training in using peer-assessment procedures and providing effective feedback is necessary (Dierick and Dochy, 2001; Sluijsmans et al., 2002). Usually, in peer assessment, groups of individuals are involved in defining criteria and procedures for their own assessments. In this way, they internalize assessment criteria and learn to provide useful feedback. Peer ratings tend to be fairer and more accurate in comparison to assessments by supervisors. However, peers often find it difficult to criticize their friends, and thus guidelines for evaluation need to be very clear.

Bureaucratic control and research assessment

Panel review models, and peer assessment generally, are systems of discipline, power and control, as noted. The inherent nature of bureaucracies to seek to control and rationalise behaviour also exists. This can be power used to ensure value for money or improve the performance of the research sector, but it might well have its own interests and other mixed reasons for exerting power. Power can be an end in itself, of course – and the means supposed to achieve the ends can instead become the ends themselves.

The problem of extending bureaucratic control is not simply one of power. It is about confirming and extending work and evaluation processes themselves, which can have tangible financial and other benefits to various parties. In the private sector, public research evaluation has created an accompanying industry that benefits from every new way to sell research evaluation products. The huge government bureaucracies that build up around research assessment, and the people that benefit from working with and associating with these bureaucracies, are also groups that benefit. The academics that staff these panels and receive the status, personal affirmation of their expert status and other benefits are, likewise, an interested group.

There is a large literature on the growth of evaluation and audit, bureaucratic growth and empire building, and while evidence can be mixed there is always a danger that bureaucracies justify their existence by continuing on paths they had set out on – whether they are useful or not; but also by acting to justify and extend their powers over more and more areas of control (Dollery, 2009; Goldfinch, 2009b; Power, 1997). Such mechanisms may suit and attract particular individuals: such as ‘bureaucratic personalities’ who have a strong commitment to rule-following and often have difficulties with ambiguity and uncertainty – with this often expressed in an attempt to control complexity with a potentially ever-increasing body of rules. These persons are not just found in central government bureaucracies, but also in the highly bureaucratised structures and career paths in universities that can reward an enthusiasm for the minutiae and tedium of committee attendance and administrative tasks. This can mean that senior positions at universities could be filled with those ready, and indeed keen, to undertake these tasks. There is a danger that these same individuals might be the ones selected for panel review. Combining the two groups can mean that the push from the centre, and the pull of seconded and panel sitting academics, contributes to the increasing measurement and bureaucratisation of research. The impact measurement, at least in its Funding Council variant, even sees an attempt to extend bureaucratic rule into the unknown future.

As such, the enthusiasm for extending control and rule-making and the self-perpetuating existence of bureaucracies and bureaucrats needs to be guarded against in the evaluation of research. Unfortunately, much of the history of panel assessment models sees the opposite, with an increasing layering of measurement and control. Measurement can be expanded from an actual focus on research output and performance, to a host of other input measures and indicators that have only a passing acquaintance with research, with some of doubtful usefulness, but of definite cost and burdensomeness. The measures of Peer Esteem and Contribution to Research Environment are conceded, even by supporters of the New Zealand PBRF, to include measures that are clearly not research. Impact measures proposed in the United Kingdom REF seem a particularly pernicious version of this.

Other measures, such as research teaching and so on, are largely input measures at best. In the case of some administration and teaching measures used, there seems to be little or no relationship with research. It might do to encourage some research funding and other input measures for other reasons, but these are conceptually separate to quality assessments of research. Research funding and some other measures might be a reflection of past performance of course – but then this perhaps becomes an esteem issue, underpinning the double counting and circularity of some measures. As such, if a lesson can be pulled from panel-based surveys it is that the importance of measures be limited in terms of administrative burdens, but also limited to those directly relating to research performance.

Stronger emphasis on Whole Brain® assessment

Assessment during the tooth morphology course included both formative and summative assessment. Formative assessment included tests performed during learning opportunity, self-assessment and peer assessment. The facts and details expected in the tests that students had to complete during learning opportunities required A- and B-quadrant thinking. Students also had to synthesise their knowledge into answers that indicated the relevance of tooth morphology in the clinical setting, which required D-quadrant thinking (Oosthuizen, 2001).

Summative assessment included a project and a final test at the end of the series of learning opportunities. The project required students to produce, manufacture or create something that would show what they had learned about tooth morphology in a Whole Brain® Thinking manner. They were free to do it in any way they chose. The freedom of choice, and the requirement to create an artefact to illustrate their knowledge, addressed D-quadrant thinking preferences. The content of these artefacts – the facts and application of these facts – required students to use A-quadrant abilities. The details of the content required B-quadrant competencies, while the physical activities necessary to complete the assignment involved C-quadrant skills.

The students created extremely creative artefacts that illustrated their mastery of the subject and ability to construct new knowledge in a meaningful way. The different products were evidence of the students’ diversity, commitment and willingness to be different. The artefacts they produced included games (Figures 4.43 and 4.44), an in-flight magazine (Figure 4.45), a clinical jacket (Figure 4.46) and a family album (Figure 4.47) (Oosthuizen, 2001).

The dialogue between the lecturer and students and among students improved considerably. Discussions were now taking place throughout the sessions and from the results of their projects it is clear that the students were highly motivated.

While the students were completing their projects, it was interesting to note that some preferred starting with the big picture first, and then breaking it down to detail. Others started with the detail and then constructed the big picture. Similar findings from student projects are documented by Lumsdaine and Lumsdaine (1995).

It was a positive experience for the students to become involved in their own learning process. When they were assessed and the final marks for the module compiled (including their projects), the students on averaged performed 30 per cent better in the module than during previous years, for an average grade of 92 per cent vs 62 per cent. Their experience of the Whole Brain® Thinking nature of the curriculum, the friendly and supportive learning environment and the option of creative Whole Brain® assessment were all features that motivated students and led to exceptional individual academic achievements.

Assessment and Instructional Planning in Practice

So what does it mean to plan for assessment to take place in a lesson? If we look at an English lesson, which is based on peer assessment, we can begin to see what Dewey meant when he said that the lesson must have high organization based upon ideas. Each of the activities builds on the one preceding it so that as the pupils come to each new activity, they can use what they have already learned through earlier tasks. There is a rationale to the lesson.

Above is an outline of the main events of the lessons.

In the lesson the teacher, Angela, asked the pupils to consider a dramatic rendition of a nineteenth-century poem that they had begun looking at on the previous occasion. The lesson has the potential for pupils to engage with the question of what makes for quality in a piece of work – an issue which is difficult in English and hard for pupils to grasp. Significantly, Angela adopts the procedures of formative assessment identified – sharing the criteria with the learner, and using peer and self-assessment as a means to this end. For this, two activities – modeling and peer assessment – are linked. The modeling activity at the start of the lesson appears to be designed to help pupils know what to do when they peer assess.

Angela began the lesson by asking the pupils to draw up a list of criteria for performing a poem. All suggestions came from the pupils while she probed, challenged, and polished their contributions. For example:

Pupil: You could speed it up and slow it down

Angela: Yes – pace, that's very important in reading [teacher then writes the word ‘pace’ on the board].

Interestingly, the Japanese have a useful term for describing such a process – neriage, which literally means polishing. In Japan, recapitulating the contributions made by pupils is an important part of teachers' classroom practice. It provides an opportunity for teachers to synthesize the contributions made by different pupils, to interject specific vocabulary, and also to refine or re-contextualize ideas.

The next activity built well on the previous task when Angela and the classroom assistant then performed the poem to the class and then invited pupils to critique their performance based on the criteria. A similar form of probing took place in these exchanges too:

Pupil: It [the performance] was boring.

Angela: What do you mean boring?

Pupil: There wasn't enough expression in your face when the poem was being read or in the reading.

Angela: So what could I have done to make it better?

Pupil: You could have looked and sounded more alarmed

Angela: Like this? [strikes a pose]

Pupil: Not quite

Angela: More like this? [strikes another pose]

Pupil: Yeah.

These three tasks in Angela's lesson the creation of the criteria, the performance of the poem, and the application of the criteria to Angela's and the learning-support assistant's performance all governed the pupils' thinking about what was needed when they acted out the poem themselves and the peer assessment of those performances. So we see again that each stage of the lesson built on the one that had preceded it. There is a logic to it. Angela's questioning of the pupils provided further opportunity for both her assessment of their performance and the pupils ability to improve on it, as Dweck suggests. Angela takes a more task/incremental view of learning believing that something can always be done or changed to make the pupils in her class understand better what they can do to improve.

Two crucial but subtle elements emerge – the potential scope of the tasks, the way they interlock, and the opportunities these afforded for current and future pupil independence. Although it is hard to separate out the various aspects of the lessons, as they overlap, it is possible to use James and Pedder's (2006) analysis, which has certain headings, as a way of organizing the analysis – making learning explicit, promoting learner autonomy, and performance orientation.

If we start with making learning explicit we find that pupils in Angela's lesson engaged both in technical considerations, such as clarity and accuracy, as well as the higher-order, interpretive concepts of meaning and effect. In addition, the modeling of what was required in Angela's lesson ensured that pupils went beyond an imitation of that model because it challenged them to think about the variety of ways they might enact their interpretation of the poem.

The sequence of activities, each building on the other, guided the pupils in Angela's lesson toward being independent or autonomous learners (the second of our factor headings) because the tasks, including encouraging the pupils to create their own criteria, helped them to think for themselves about what might be needed to capture the meaning of the poem in performance.

Pupils in Angela's lesson, therefore, also began to engage in the more complex issues of any performance be it verbal or written. That is the pupils were asked to explore the relationship between the meaning of a product and the way in which that meaning is expressed – between form and content. This leads us to the final element of performance orientation. Crucially, Angela always described the tasks as opportunities for the pupils to improve their performance. In this way, the activities had an open, fluid feel which corresponded with the notion of promoting pupil autonomy; it reinforced a sense of limitless progress whereby assessment is always seen as a tool for future rather than past performance. This was mainly done by creating tasks, all of which worked effectively together, designed to enable children to enter the subject community or guild (Sadler, 1989).More importantly, Angela has prepared a lesson which is high organization built upon ideas.

Group Assessment

Group assessment in higher education is a troublesome area on which experience and opinion are mixed, and on which the literature is large and growing. Often associated with group work and peer assessment, a common form of implementation involves dividing the students into teams, each team working on a complex group project. The single major piece of work produced is assessed by the teacher, or by members of other groups as peers, or by all involved. Various schemes have been proposed for awarding individual student marks that eventually contribute to course grades. Options include awarding the same mark to each group member, and creating a notional mark pool which is apportioned according to each individual's contribution to the finished work. If handled properly, quality, productivity, and student satisfaction with group assessment can be high. Poorly handled, group assessment gives rise to disputes about the influence of team composition and group dynamics on individual marks, and consequent fairness to individuals in marking and grading. However, group work that is directed toward creating highly productive collaborative sites for learning does not necessarily entail using group-assessment tasks as well. Individual assessments are often appropriate.

Peer- and Self-Assessment

Self-assessment by students is essential for their development as independent and responsible learners. To meet this need, practice must involve them in assessing one another's work so that they might thereby develop the skills both of peer assessment and, by seeing their work through the eyes of their peers, of self-assessment. To do this, students have to understand the aims of the work involved, and the criteria by which this work should be assessed. In certain topics in some subjects, the criteria might be well defined (e.g., calculating the value of a force in a physics problem), in others there might be many different ways in which work could achieve excellence (e.g., writing a critical appreciation of a poem). Overall, the aim was to move students away from dependence on the teacher toward independence in the power to guide their own learning. As one teacher expressed it:

The kids are not skilled in what I am trying to get them to do. I think the process is more effective long term. If you invest time in it, it will pay off big dividends, this process of getting the students to be more independent in the way that they learn and taking the responsibility themselves (Black et al., 2003: 52).

In helping students to develop their self-assessment, teachers must clearly specify aims so that students can both steer their work toward attaining them and understand the criteria by which that work can be judged.

Peer assessment is another powerful tool: helping pupils collaborate effectively in groups is essential to such work. However, while there is strong evidence that collaborative groups can improve attainment, a survey of group work in a large sample of UK classrooms (Blatchford et al., 2006) shows that the type of collaboration that can engage pupils in reasoned arguments about their own and one another's contributions is not often found. Intervention programs by Blatchford and colleagues and by Mercer et al. (2004) have demonstrated that students trained in such collaboration produce improved attainments, both in the quality of their arguments and in subsequent tests. The work of Dawes et al. (2004) is a good example of dissemination of the lessons learned about effective group work.

The formative use of summative tests extends this area of practice. Students learn by marking one another's test responses, particularly if they have to develop the marking schemes in the light of their understanding of the criteria. As for the new way of looking at written work generally, the point is that a test, while serving the summative function, might also be used, through formative feedback, as an opportunity for improving learning. One addition in this area was that teachers would strengthen preparation by asking students to compose questions which might well be suitable for the coming tests, for in so doing they would have to identify for themselves the main aims of the work. They would also learn about the need to frame the text of a question with care, and to exercise similar care in responding to questions framed by others.

Peer Assessment

Of course, self-assessments may be highly subjective – more able students might overestimate their capabilities, while weaker students might underestimate theirs. Practice with teacher feedback is likely to improve accuracy. However, teachers have too much to do. Peer assessment has the advantage of offering some degree of external objectivity, albeit not nearly as competent as that of the teacher, but much more readily available. For peer assessment, pairs should be matched which are not too disparate in reading ability. Activities should be suggested, and simple checklists provided for students to assess the performance of other students. The pair might begin by discussing each other's self-assessment, to become familiar with the other's perceptions, and give them some sort of reality check. Beyond this, peers can discuss the selection of best work for portfolios, give feedback on retellings, render evaluative comments in group discussions, and discuss the motivational and affective aspects of their book readings. Maybe, they could even do more complex things.

Designing group assessment

Group work is increasingly valued in higher education because it allows students to practise the kind of team-based tasks that they will encounter in the workplace; hence, it lends itself to authentic assessment. These activities provide immediate opportunities for formative peer assessment. This may be implicit in the design of the activity (through students negotiating), or you might explicitly request it. However, students often dislike summative assessment of group work when it does not recognise their individual contributions. Drawing on previous work, Falchikov (2005) discusses eight commonly used strategies for differentiating group and individual marks:

multiplying the group mark by an individual weighting factor;

distribution of a pool of marks;

use of a contribution mark;

separation of product and process;

equally sharing a mark but with exceptional tutor intervention;

splitting group tasks and individual tasks;

issuing yellow (warning) and red (zero grade) cards to individuals perceived as not ‘pulling their weight’; and

calculating individual grades in terms of deviations from the norm.

Peer assessment, where each member of the group ‘rates’ contributions of other members, has also been used to address the issue of equity but concerns are often expressed about assessing individual contributions fairly (James, McInnes & Devlin, 2002). The strategy of using peer assessment summatively remains unreliable despite procedures designed to reduce bias (Kennedy, 2005; Li, 2001Kennedy, 2005Li, 2001). Engaging students in discussion about the selected strategies provides an avenue to enhance peer assessment of group work with potential benefits for reliability and validity.