The future of assessment as a human and social endeavour: Some inconvenient truths
Testing and examination have important consequences for high quality performance, such as graduation, promotion, certification, scholarship, and others. Assessments are also assumed to be accurate indicators of quality. Psychometric research evaluates the statistical properties of tests on the assumption that students are well prepared, make maximal effort, and that irrelevant factors – such as parental income or education, ethnicity, and others – do not interfere with performance.
Recently education systems have chosen to use alternative assessment methods (eg, performances, portfolios, peer assessments, self-assessments and others) to ensure students develop higher-order thinking skills. Simultaneously, systems want teachers and students to use assessments to improve student learning, teacher instruction, and institutional outcomes.
Identification of learner strengths and weaknesses is used to generate insights as to who needs to be taught what next and what learners themselves can do to keep learning. Fulfilling these valued goals requires an assessment-capable teaching force that can exploit the in-the-moment, on-the-fly processes of classroom interaction into valid judgements about student learning, the generation of constructive feedback, and the adjustment of teaching practices. These emphases are normally described as assessment for learning.
“Our conventional understanding of assessment is inadequate. Assessment performs a useful and necessary function in society. All assessments of any kind contain unknown amounts of random and systematic error. Despite robust statistical methods, our current assessment theory and practice is inadequate for the complex human and social factors. Thus, much of what we think we know based on assessment is much less certain than our intuitive theories of testing would suggest.
“While new developments are in progress to resolve these challenges, the best we can do now is be open to the possibility that all we have done despite our best efforts could be wrong and treat our results as tentative subject to further evaluation and testing.” Professor Brown argues.
“This lecture is a chance for me to overview the world of assessment in light of what we know about things that can invalidate the interpretations and actions we need to make. It summarises my careers, thus far, as a teacher, standardised test developer, and researcher into the psychological and cultural factors that make assessment difficult. It points to directions that we need to explore and highlights the importance of a generally open and tentative mind-set towards the use of all assessments.”