The edumetric quality of new modes of assessment: Some issues and prospects

Abstract

Assessment has played a crucial role in education and training since formal education commenced. Certainly, assessment of learning has been seen as the cornerstone of the learning process since it reveals whether the learning process results in success or not. For many decades, teachers, trainers and assessment institutes were the only partners seen as crucial in the assessment event. Students were seen as subjects who were to be tested without having any influence on any other aspect of the assessment process. Recently, several authors have called our attention to what is often termed 'new modes of assessment' and 'assessment for learning'. They stress that assessment can be used as a means to reinforce learning, to drive learning and to support learning-preferably when assessment is not perceived by students as a threat, an event they have to fear, the sword of Damocles. These authors also emphasize that the way we assess students should be congruent with the way we teach and the way students learn within a specific learning environment. As such, the 'new assessment culture' makes a plea for integrating instruction and assessment. Some go even further: students can play a role in the construction of assessment tasks, the development of assessment criteria, and the scoring of performance can be shared amongst students and teachers. New modes of assessment that arise from such thinking are, for example, 908 or 1808 feedback, writing samples, exhibitions, portfolio assessments, peer-and co-assessment, project and product assessments, observations, text-and curriculum-embedded questions, interviews, and performance assessments. It is widely accepted that these new modes of assessment lead to a number of benefits in terms of the learning process: encouraging thinking, increasing learning and increasing students' confidence (Falchikov, 1986, 1995). However, the scientific measurement perspectives of both teachers and researchers, stemming from the earlier, highly consistent framework, where assessment was separate from instruction and needed to be uniformly administered, can form a serious hindrance to a wider introduction of these new assessment methods. As Shepard (1991) and Segers, Dochy and Cascallar (2003) indicated, instruction derives from the emergent constructivist paradigm, while testing has its roots in older paradigms. So, we argue, the traditional criteria for evaluating the quality of assessment need to be critically revised. The question that needs to be asked is whether we can maintain traditional concepts such as reliability and validity, or if these concepts need to be considered more broadly in harmony with the development of the new assessment contexts. In this chapter, attention is paid to the new evolution within assessment and especially to the consequences of this new approach to screening the edumetric quality of educational assessment.