||International Conference on English Language Teaching Instruction and Assessment|
Ontario Institute for Studies in Education, University of Toronto
I argue that research and development in the field of language assessment needs to pursue three directions signaled by exemplary inquiry in this field in recent years. First, we need to broaden the scope of inquiry, particularly to understand ordinary assessment practices in classrooms and other learning contexts better, to consider various situations of language assessment around the world, and to enhance assessment practices for unique populations. To achieve these purposes, we also need to expand the range of methods of inquiry about language assessment, incorporating various kinds of qualitative inquiry. Second, a more profound set of theories need to inform language assessment. While Messick's framework has been widely adopted as a basis to validate formal language tests, new philosophies, epistemologies, and empirical data are needed to define what language assessments are and should be. In turn, the core concepts that guide language assessment require definitions, critical analyses, and continuing refinements in view of empirical phenemona. Third, there needs to be ongoing consolidation and re-evaluation of the knowledge that exists about language assessment. To this end, critical, historical analyses are vital as are techniques such as meta-analyses for synthesizing empirical research findings. I cite studies that exemplify each of these directions, showing how they help to improve and substantiate our knowledge about language assessment.