My interest in putting peer assessment into practice in my own teaching environment has been increasing as a result of being exposed to these influences. In 1993 I incorporated some peer evaluation into a law unit taught at Murdoch University and then again in a different form in the same unit in 1995. This paper is a brief account of those experiences, focussing on some of the dilemmas I have encountered, and beginning with the theoretical structure upon which my practice has been based.
There is a growing awareness that students should be encouraged to think of their university degree as merely a milepost in a life of learning rather than as a terminal destination in their intellectual development. In order to become effective life-long learners students must acquire some proficiency in answering such questions as: "What do I know?" What do I need to know?" What must I do to gain that knowledge?" It is thought that this sort of insight can be improved by practising both self and peer assessment. As one writer puts it:
... life-long learning requires that individuals are able not only to work independently but also to assess their own performance and progress. Involvement in the assessment process would hopefully heighten our awareness and knowledge of the student approach to learning and enable students to make rational and objective judgements about their own strengths, weaknesses and range of skills [5].In this view peer assessment acts as an exercise in which students can both practise assessment and observe how others evaluate the results of learning. Becoming familiar with the process of assessment is as important as the conclusions reached. Here we see a conception of learning not as the accumulation of information but as a dynamic ability to organise and modify ideas.
One school of thought has focussed on the importance of "metacognition" in higher learning, particularly in relation to problem solving skills. Metacognition has been described as involving "knowledge and control of one's own thinking" [6]. The manifestation of metacognition lies in the ability to plan, monitor and evaluate in the course of problem solving which entails a kind of dialogue with oneself. In the words of two mathematics teachers, students "need to develop such metacognitive skills of 'inner speech' in order to become effective [problem solving] modellers" [7]. One way of stimulating this ability is to allow students to experience dialogue with their peers concerning the evaluation of their work as a model of what an individual student may also do within herself. Here again it seems there is a conception of learning as gaining the ability to be flexible and adaptive in one's thought rather than placing reliance on a store of static knowledge.
Other educators have emphasised the need for tertiary level students to take responsibility for their own learning, a clear criticism of the "transmission model" of teaching and learning where full responsibility relies on the instructor to ensure that prescribed information is given and received. One prominent writer has pointed out the inconsistency between many traditional practices of higher education and the ideals of independence, thoughtfulness and critical analysis which are also espoused [8]. Assessment practices have in the past ignored the possibility of a student role in the process, thereby perpetuating an intellectual dependency and perhaps lack of self-confidence on the part of students. One way to remedy this is to give students the responsibility to participate in the assessment of their own and fellow students' work. Here, learning is again conceived of as a process of active engagement rather than passive reception.
Finally, it has been pointed out that peer reviewing skills are an essential component of expertise in professional, business and academic life. As two educators have noted:
Throughout their working lives, students will need to assess the quality of the work of their subordinates, their peers, their superiors and, realistically, themselves. Hence, building confidence through experience must take place at some stage. As part of "education for life" this can also assist in the essential task of allowing students to become self-learners, a measure of the quality of the educational programme they are undertaking [9].Particularly in the field of law (a largely self-governing profession) it has been suggested that the capacity for fair and accurate peer assessment is an important attribute of the successful practitioner [10]. Whether reviewing documents produced by another solicitor or considering the propriety of her actions in a disciplinary proceeding, the ability to make good judgements of another's work or actions is desirable. It is also expected that evaluating the work of peers will help to develop the ability to reflect critically on one's own professional activities in keeping with the ideal of the "reflective practitioner" [11].
However, the desirability of good assessment skills is not confined to professional fields. In order to succeed in academia one must learn the unique criteria which govern the valuation of work within a disciplinary tradition. One way of becoming familiar with those standards is to practise applying them. In a sense, most university education is concerned with introducing students to the standards by which relevant and valuable contributions to disciplinary knowledge are identified. Why shouldn't they be given the opportunity to try applying those standards themselves at some point in their tertiary education? While most students will not become academics and thus inevitably involved in peer review processes, all students may benefit from engagement with the criteria of academic excellence by assessing the work of others.
Before moving on to consider the possible perils of peer assessment another important but more practical advantage of that process should be mentioned. Whether or not the results of peer assessment are counted toward the grades awarded, student peer involvement in reviewing and commenting on student work can help to reduce the workload of the teacher. Particularly in those subjects where multiple items of student work are set, the contribution of students to giving feedback (and possibly marks) to each other can help to reduce the time demanded of the instructor to perform these tasks.
These then are the conceptual arguments for the promise of peer assessment. But there is also some evidence from teaching practice to support the view that self and peer assessment promises benefits for tertiary learning. Two educators have noted that:
The indications are that there is a direct relationship between assessment skills and the quality of work which a student produces. More excitingly, there is evidence that the skills can be exploited directly with a consequential improvement in student performance [12].Let us now see what the literature has identified as the potential perils of introducing peer assessment in tertiary setting.
So far as potential negative impact on student learning is concerned we might conveniently divide the difficulties into those that have to do primarily with peer assessment as a component in formative evaluation of students and those that have more to do with such assessment used for summative purposes.
Doubts about the place of peer assessment in formative evaluation of students relate to the question whether they are able to give each other valuable feedback. Here, the potential problems may concern both the content of such communications and their tone in so far as they may have an adverse effect on interpersonal relations and possibly academic self-confidence. As one writing teacher has noted, exposing one's work to peers can seem a risky proposition but peers can act as motivators in addition to being astute critics if a spirit of cooperation and not competition is fostered [16].
The constructiveness of feedback received from peers from an academic standpoint will depend on the application of appropriate criteria by students in making their evaluations. Several authors have pointed to the difficulty students may have in identifying relevant grounds for criticism of each others' work [17]. Students have taken up this point in their comments on peer assessment as in the following reported comments: "Students will need to be told what points to look for when assessing other people's work."; "Students will not usually have the same level of understanding of subject matter compared to the teacher" [18]. Other educators however have pointed to the social factors involved in gaining understanding of the accepted criteria and standards of judgment of work within a discipline. For them learning these is a process of acculturation in which peer assessment may play a positive role:
Peer and self assessment provide a mechanism for the acculturation of the student. ...Peer assessment facilitates and encourages the generation of consensual objectivity by requiring students to discuss, challenge and justify their subjective constructs. The teacher mediates and guides this process of legitimization by selectively approving some constructs above others [19].As regards the tone of student peer assessments it has been pointed out that maintaining the constructiveness of the feedback may depend crucially on conveying an appropriately positive message. Here the problems may be related to: student reluctance to criticise; student fear of criticism; confusion of feedback with criticism; students feeling unsafe; students taking a risk; fear of retaliation [20]. As two writers have concluded: "If the students do not feel a sense of trust, respect and rapport in the classroom, they will be unwilling to give honest feedback or accept feedback given to them" [21].
The key question arising out of the use of student peer assessment in summative evaluation is the issue of the validity of marks awarded. Typically, this has been framed in terms of the degree of correlation between marks students award themselves and their peers and the marks instructors would give for the same work. A closely related problem concerns the use of different criteria for assessment by students and teachers.
To the extent that tertiary education performs an accrediting function, particularly in professional fields, the issue of the validity of student-awarded marks if they are to count towards the academic requirements for a degree is a crucial one. For this reason, and also perhaps because the question appears suitable for quantitative analysis, there have been many published studies of student marking. In 1989 two researchers surveyed the literature on student self-assessment in an attempt to determine if there were any clear trends [22]. They found research which seemed to indicate both under and over marking by students compared with teachers and problems with the methodology of much of the published work. Since that review was performed there have been further reports of inquiries into the validity of student self and peer assessment, and the results should probably continue to be described as inconclusive [23].
In addition to the question of students' level of knowledge as it may affect the accuracy of their judgements of student work there is also the issue of the impact of personality and related interpersonal factors. Researchers have been alive to the possibility that such personal factors may affect the validity of student-awarded marks and some have reported attempts to avoid this problem, for example, by ensuring anonymous marking. On the other hand one teacher has concluded that a group which is well-known to each other and relatively cohesive offers the right climate for reliable and valid peer assessment [24].
Finally, in the area of summative evaluation by students there is the question of using appropriate teacher-sanctioned criteria for marking. The literature shows that many teachers have attempted to involve students in formulating the criteria for evaluation as part of an approach to sharing responsibility with them. It seems most instructors have been pleased with the degree of "fit" between student-generated criteria and their own. However, there is some indication that students may tend to weight "effort", the apparent amount of work undertaken as distinguished from the intrinsic value of the product of that labour, more heavily than would teachers [25].
Clearly, there are some formidable problems and unresolved issues connected with the use of student peer assessment whether for formative or summative evaluation in the tertiary setting. Nevertheless, for me the promise of peer assessment appeared to be worth risking the perils and I have introduced it into my law teaching. What follows is a description of my practice and some of the dilemmas I have encountered in the course of that experience.
Peer assessment can of course be structured in a wide variety of ways and the literature records many permutations. It has been noted that one of the difficulties with the research in this area is frequent failure to adequately specify the details of the assessment which is being examined [26] Therefore, although my use of peer assessment in teaching law was not intended as a research project, I will use the format for reporting on my experience which has been recommended by leading investigators [27], as follows:
What was being assessed?
Each student was required to prepare six individual documents throughout the year in a format specified by me. These ranged from business letters of a legal nature to contracts and deeds. Half of the class were designated as "originators" for each assignment. The originators prepared their document first and each of them then sent it to a paired student (the "responders") who replied by sending their own document back, incorporating desired amendments to the originators' draft. It was these documents which were self and peer assessed by the students. Documents and assessments were exchanged by students using electronic mail with their student identification numbers functioning as addresses, so that anonymity of the author was the rule.
What were the participant characteristics?
The class of 84 students were mostly in their last or penultimate year of law study and included both school leavers and mature students of both sexes. A range of ability as indicated by previous grades was present in the class although it is reasonable to suggest that due to the existence of quotas for entry to law school the average level of ability amongst law students is likely to be higher than amongst the university student population as a whole. The students' experience with self or peer assessment prior to taking this subject is unknown.
What subject matter, type and level of content was being assessed?
The documents assessed were those types frequently prepared by practicing solicitors. I expected them to show a high level of understanding of the relevant legal concepts and principles and the governing legislative framework. A factual situation was presented to the class as the basis on which the documents were to be prepared as a simulation of actual practice. Although the legal knowledge required to produce the documents was largely taught in prior subjects I was aware that these students did not have much if any practice in preparing such documents.
What was the time period?
The subject was a full-year unit taught over 26 weeks with weekly classes.
What opportunities did it provide for practice and feedback?
All self and peer assessments of the 6 required documents were included in the formal assessment of the unit; no practice assessments were provided for. Students in the class were surveyed by the university by means of a written questionnaire immediately before the end of the year which provided feedback to the instructor. In several classes examples of student assessments were made available to all and discussed by me as feedback to the students.
Did the marks count towards formal assessment?
Yes, the mark for each of the 6 documents prepared by each student counted as 7.5% of their mark in the subject (total of 45%). These marks were arrived at by averaging the self and peer assessment marks awarded for each document, with a proviso that the instructor retained the discretion to modify averages which fell above 80% or below 50%, by up to 10 marks, based on a review of the document.
What criteria were used?
Criteria for assessment were generated by class discussion based upon guidelines set out in a document prepared by the Curtin University of Technology School of Design entitled "Assessment Explained". The result of that class discussion was a document embodying the agreed criteria which is attached (Appendix A). Each student was required to complete a form containing the self-assessment of their own work which accompanied the document they sent to their paired student, and to prepare a similar form for the peer-assessment of the document received from their paired student. The forms for self and peer assessment were prepared by me based upon recommendations made by members of the same School of Design [28]. The peer assessment form is also attached (Appendix B) for reference.
What marking scale was used?
As shown in the document containing the assessment criteria (Appendix A) percentage marks were tied to grades according to the usual scale used by Murdoch University with which the students were familiar.
Observations?
For those students attaining average marks of over 80% my review of their documents usually resulted in the maximum reduction by me of the mark (up to 10%) although a trend was noticed to less reduction being made as the year progressed.
Very quickly it seemed that a majority of students adopted the practice of regularly giving self and peer assessment marks in the range 75-79%.
Very few marks of less than 50% were awarded most low marks were the result of penalties being imposed for lateness in submitting a document.
The majority of students tended to merely quote, parrot-like, portions of the assessment criteria in their self and peer assessment forms with little attempt to relate the criteria to specific aspects or portions of the document assessed.
Most peer evaluations echoed closely the mark given through self-assessment as if there were a sense of reciprocity at work. In some more isolated instances retaliation for low marks seemed to occur.
My impression was that the students did not like the self and peer assessment process. Comments made in the open ended section of the student evaluations are indicative: "The idea of marking the work of peers was a good one, but unsuccessful. 'Normal' students want to do as well as they can get away with. Once one person abuses the system so will all others." Self and peer assessment contributed largely to the result that a majority of the class received final marks for the unit in the range 70-75%.
Peer-Assessment Form
Your Student Number:
Student Number of author of work assessed:
List three aspects of the project you consider have been covered particularly well having regard to the project instructions and agreed assessment criteria:
2.
3.
2.
3.
| Please cite as: Zariski, A. (1996). Student peer assessment in tertiary education: Promise, perils and practice. In Abbott, J. and Willcoxson, L. (Eds), Teaching and Learning Within and Across Disciplines, p189-200. Proceedings of the 5th Annual Teaching Learning Forum, Murdoch University, February 1996. Perth: Murdoch University. http://lsn.curtin.edu.au/tlf/tlf1996/zariski.html |