Developing Valid and Reliable Rubrics for Writing Assessment: Research and Practice
Book chapter, 2011
Writing assessment has long been a particular challenge for academic staff. A broad range of theoretical approaches have resulted quite varied practices, from traditional attempts to have single academics read through piles of student manuscripts to systematic, programme-wide efforts involving grading teams and multiple readers. The diversity of practices results from diverse perceptions and beliefs, training and levels of experience, influences and motivations, resources and institutional policies. Fundamentally, these endeavours are intended to provide writing assessments that are valid and reliable. A valid method should provide accurate assessment of student learning in connection with desired outcomes, and a reliable assessment should offer fair and consistent evaluation of learning in connection with task requirements and expectations. There is and will continue to be ongoing tension between these two aims. For example, while a multiple-choice test can provide perfectly reliable grading, what valid assessment of a student’s learning with respect to academic writing could it offer? This guide explores how academic staff can pursue the aims of valid and reliable writing assessment in the context common disciplinary, programme and institutional challenges.
validity and reliability in assessment