Avoid over-reliance on tests and exams
Amid calls for a more holistic approach towards children’s development and learning, the recently announced changes in the scoring for the Primary School Leaving Examination (PSLE) are a welcome step in the right direction. These changes have once again brought to the fore issues around high-stakes testing and exams.
The sciences of human cognition and learning tell us that standardised tests provide a reasonably objective measure of human ability in applying a learnt set of knowledge and skills. Part of what contributes to the sense of objectivity is a seductive reduction of human ability to a quantifiable metric such as a percentage score or a letter grade.
Yet, what often goes unnoticed is that this measurement is done under very controlled settings: Students take the test individually, in one sitting, over a short period of time, without the use of resources, and in settings largely removed from the settings in which that knowledge is to be applied. It turns out that the very factors that give standardised tests objectivity and other desirable psychometric properties, such as reliability, are also the factors that limit the inferences we can derive from them.
What are these limitations and what should be done moving forward?
First, tests measure our ability to apply what we know, not invent or generate new knowledge. Arguably, it is the capacity to generate new knowledge in the face of novel situations, which is of greater need today. A substantial amount of research, including my own, suggests that children have a sophisticated capacity to generate new ideas in novel situations, an ability that often goes unexploited and undeveloped. Further, research shows that when this capacity is exploited, children learn better.
Second, resourcefulness matters. Standardised tests measure performance largely in the absence of resources such as tools, access to expertise, working with others, getting feedback, and so on. Yet, in the real world, it is performance in the presence of resources and our ability to productively use these resources in our environment that gives us a competitive edge over others.
Third, if tests demand that we solve problems within minutes and in one sitting, then learning to solve problems that require persistence over time will suffer. Indeed, research suggests that individual performance on the same task varies over time, and that persistence and grit are significant predictors of learning and success in the longer run.
One could counter these limitations with the argument that the function of schooling is mainly to build basic knowledge and skills, and used later to develop more advanced knowledge and skills. However, the acquisition and use of knowledge is not the same thing.
Research suggests that knowledge is not something we can automatically use once we acquire it. How we acquire knowledge and the context in which we acquire it are stronger determinants of how well we can use it outside of those contexts.
Research suggests that transfer of learning across contexts is hard. The more we acquire knowledge in test-driven environments, the more the use of this knowledge will be constrained to test-taking contexts. This is inimical to the development of creativity and innovation demanded by our changing economy. Performance in tests does not necessarily imply deep learning and mastery — something that employers and universities around the world have already started taking into account in their selection processes.
Finally, there is the issue of cognitive development, the idea that the mind has to be sufficiently developed before it can learn and do certain things. This developmental variation is at its greatest in the growing years from birth to adolescence. Therefore, it may not be that a child does not have a specific ability, but that he or she has not reached developmental maturation.
What this means is that we may be wrongly attributing poor performance in tests to a lack of ability — the younger the child, the greater the chance of making such an attribution error. In other words, we may be prematurely making inferences about a child’s potential if we rely too much on the outcomes of tests.
So where does that leave us?
Policy has been on the move. In addition to the broadening of the PSLE scoring system, several other initiatives by the Ministry of Education, such as support for teacher-led innovations in teaching, emphasis on teachers’ professional development, removal of ranking of schools, and funding of research in education, represent efforts that emphasise the holistic development of the child. The idea is that when teachers are given the necessary training and incentives to design evidence-based teaching methods, alongside a reduction in emphasis on ranking and finely split scoring, then students’ learning experiences, in and out of classrooms, should become more holistic and not overly focused on tests and exams.
As parents, we also need to get on the move. Often, I hear parents say they have no choice because of the way the system is. While this is understandable, we could instead choose to move not because of, but in spite of, the system. The mindset needs to change, from one that over-relies on the importance of tests and exams as the be all and end all of a child’s learning to a focus on the holistic development of the child. This may be easier said than done, but if we as parents are not prepared to change our mindsets for our children, then who will?
Over time, one hopes that classroom, school, and societal cultures will move, for what is required is a cultural shift. But no one thing can in and of itself engender this shift. Policy, people, and practice need to move together, and not wait on the other to move first.
ABOUT THE AUTHOR:
Manu Kapur is a Professor of Psychological Studies at The Education University of Hong Kong. He previously headed the Learning Sciences Lab as well as the Curriculum, Teaching and Learning Department at the National Institute of Education of Singapore. The opinions here are his own.