Author:

Keywords:

peer assessment, formative evaluation

Abstract:

The dissertation includes three theoretical contributions and three empirical studies on peer assessment, a general introduction and final reflections including a discussion of the results, a discussion of the educational implications and a discussion of some methodological issues. The first contribution delineates the role that peer assessment can play in raising the consequential validity of an assessment system. First, it clarifies the type of effects that assessment in general can have on learning, and formulates the design principles for increasing the consequential validity of an assessment system. Then, it is shown that peer assessment helps to meet the identified design principles that enhance consequential validity of an ‘assessment system’. More specifically, this dissertation shows that peer assessment can make it more feasible to include challenging and authentic tasks in one’s assessment system; it can help making the assessment demands more clear to the students; it can provide a supplement or a substitute for formative staff assessment; and finally, it can support the response to teacher feedback. The second contribution goes beyond the impact of peer assessment on the consequential validity, and addresses the problem that the output of peer assessment is evaluated against a variety of quality criteria in the literature, resulting in a cluttered picture. The different conceptualisations of quality that appear in the literature are analysed. It is shown that discussions about the most appropriate quality criteria for the output of peer assessment should be brought back to the underlying differences in goals. The most obvious goal is its use as an assessment tool. The learning goal of peer assessment has also been well-established. Investigating the literature more closely yields three additional goals: installation of social control in the learning environment; preparation of students for self-monitoring and self-regulation in lifelong learning; and active participation of students in the classroom. Each of these goals results in different quality criteria. It is argued that only the criteria that are congruent with the goal that one is trying to achieve should be considered when evaluating the quality of peer assessment. The third contribution starts from the observation that, together with the expansion of peer assessment research in the last decade also the diversity of peer assessment practices has increased exponentially. This diversity poses difficulties for practitioners as well as researchers. An inventory of peer assessment diversity is developed that may be of interest to practitioners, as a checklist of important decisions to take or an overview of possible alternatives to a specific practice, and to researchers as a guideline of which information to provide on the particularities of their peer assessment design. The fourth contribution compares the impact of peer feedback and teacher feedback on student learning, addressing the question whether peer feedback can serve as a substitute for expert feedback. A pretest posttest control group design examines the long term learning effects of individual peer feedback and collective teacher feedback on writing assignments in secondary education (N=85). Moreover, it examines the addedvalue of two measures to support the response of the assessee to peer feedback: an a priori question form and an a posteriori reply form. The study showed no significant difference in students’ progress on essay marks between the condition with plain substitutional peer feedback and the control condition with teacher feedback. However, both groups (plain peer feedback ànd teacher feedback) appeared to make significantly less progress then the groups in the ‘extended’ feedback conditions with the question or the reply form. The fifth contribution examines a group of 68 first year students in secondary education who experienced formative peer assessment for three successive writing assignments. They were divided in two experimental conditions (similar to the ‘extended’ feedback conditions in the previous contribution) and a control condition with plain peer feedback. Students’ progress in writing performance is examined against the constructiveness of the peer feedback they gave and received, and against the condition in which they participated. The effect of the constructiveness of feedback is studied from two directions: from the point of view of the receiver of the peer feedback (‘assessment for learning’) and from the point of view of the assessor who gave peer feedback (‘assessing for learning’). The results of a repeated measures analysis show a significant positive effect of the composition of the received peer feedback on student performance. The constructiveness of feedback that students provided themselves was not found to improve their learning. Nevertheless, the overall level of constructiveness of the feedback was low. Possible barriers preventing students from providing good feedback, and solutions to these, are discussed in the paper. Finally, the study could not replicate the effect of condition that was found in the fourth contribution. The sixth contribution compares strengths and weaknesses of peer feedback and staff feedback, from the student’s perspective. The study is situated in a university course with 192 first year students in educational sciences. Generic, collective staff feedback on the draft versions of a series of cumulative assignments is complemented with a formative peer assessment system. Starting from a hypothetical forced choice, a further in-depth study addresses the perceived characteristics of both sources of feedback and their perceived contribution to a learning environment that attends the learner’s needs. These perspectives are complemented with reasons reported by students to prefer one of both sources of feedback. Closed-ended questionnaire items are triangulated with qualitative data from open-ended questions. Results show that approximately half of the students were willing to trade in the credibility of staff feedback for the specificity of peer feedback if they have to choose. However, both sources of feedback showed to have their own strengths and weaknesses from the student’s perspective. They were complementary and they even provided the conditions under which the complementary source became better.