At the 2022 budget debate in parliament (7 March), Minister for Education Chan Chun Sing tried to explain why tests and exams in our schools cause so much stress:
Because we use it to compare ourselves/ our children to others, rather than “an opportunity for us to know our children better - where they stand, where they're strong, and where they're weak."
How do we present the results of tests, exams, even PSLE, in our local school system? Children are given a score, e.g. between 0-100, which is written on top of the test paper, or printed on a result sheet, and returned to them. The most natural next step would be to look around the room and compare results, don’t you think?
If we really want to achieve what Minister Chan is suggesting, here are some better ways to test and provide feedback on children’s learning throughout their primary school years. A combination of these methods can also replace PSLE. (This list is not exhaustive, we are just sharing better ways of assessing learning that we have encountered in international schools):
i. More regular but shorter, bite-sized assessments, in multiple formats e.g. multiple-choice questions, short written answers, individual or group projects, presentations, class discussions, hands-on challenges, etc.
These are more age-appropriate and accurate gauges of learning for primary school students. They allow children to demonstrate their different strengths as well, without penalizing those who have shorter attention spans, difficulty with writing long answers, etc.
They are also a more accurate reflection of how children will be expected to use their knowledge and skills in the working world, in collaboration with others, rather than memorizing and regurgitating key words and phrases and sitting through long exams.
ii. Use of computerised testing, so that the sub-areas of strength and weakness can be flagged by the automated marking, and listed right at the top along with the grade. This actually moves the focus very quickly to the self-improvement feedback that the child should be seeing. And teachers don’t need to plough through papers manually if they want to highlight children’s areas for improvement. These tests can be standardised nationally.
iii. Computer Adaptive Testing, which can hone in more quickly and accurately on the child’s level of understanding, without destroying confidence. Questions are guided by the level of understanding demonstrated in previous questions, so e.g. if a child is getting a lot of questions correct, subsequent questions will grow harder to hone in on their true level of understanding. Similarly, if they are getting a lot of questions wrong, subsequent questions will be easier.
At the national level, it allows each child to be benchmarked against themselves as well as the national standard for their age. It is less confidence-destroying for the child, who for example does not see the 2 traditionally impossible-to-solve questions at the end of every PSLE Maths paper, unless they are one of the rare few likely to be able to solve them. It has been used for major standardised exams like the GRE for at least the past two decades, and more recently the SAT.
iv. Also easier with computerised testing - provide a graph that shows the child’s progress in the subject over the past few terms, i.e. their learning trajectory. Children will naturally be drawn to start comparing the slopes (showing how much/ how fast they have improved), which can be displayed more prominently than the actual grades. This also nurtures a growth mindset and teaches an important life lesson that the direction/ pace of growth is more important than where one starts.
iii. Also easier with computerised testing - Provide a comparison graph that tracks the national average performance, so there is some element of comparison for child/ teacher/ parent on how far off the child is from expected levels of achievement at their age. But the focus of the comparison is this ‘faceless average’ rather than the other students around them.
iv. Yes, also easier with computerised testing - Send these results home by email to parents, rather than giving them out in class. Again, this keeps the focus on the individual child’s progress. Teachers can discuss individually with children and parents on sub-areas that have improved (to encourage) or are still weak (to focus further effort), both of which would have been helpfully and automatically collated for them for every child.
Below is an example of a results sheet from an international school in Singapore. It keeps the focus on the individual child’s progress and areas of strength/ weakness. The blue line shows the child’s score at different timepoints through multiple grade levels. The orange line shows the average score in the school or school cluster. The yellow line shows the expected average score for a child that age:
All the above would be made much easier if we adopted computerised testing (here’s a list of benefits). Let’s harness the digital and data age for the benefit of our primary school children, and in the process, make things easier for our overstretched and overworked teachers too.
Besides individual progress and performance, it will give teachers (and their supervisors/ coaches) plenty of data to see at a glance if there are specific sub-areas that many of their children are struggling with.
When conducted at the start and end of the academic year (pre-and-post testing), such testing also allows teachers (and schools) to be rated based on the improvement they have made in children’s academic performance, rather than raw scores.
It enables our schools to accurately identify and reward the teachers who make the real difference in the trajectory of our children’s development, the ones who can move children from much below the national average to just under, or even over.
It would also allow MOE to have much more fine-grained data on schools and their relative strengths and challenges, and MOE could then provide them targeted help to really get us to ‘every school is a good school’.
The use of regular, bite-sized, pre-and-post testing of children also solves the other mystery that MOE is apparently still grappling with, according to Minister Chan (budget debate, 7 March 2022):
“Pinpointing the best age to test students in a national examination that balances children's different pace of development is something the Education Ministry will continue to study. Testing too early might be detrimental for some late bloomers, while testing too late might mean being unable to apply the interventions necessary to help children progress at their pace.”
In this day and age, there is no need for a one-off high-stakes national exam in the first 9-10 years of education, and so MOE should stop worrying about what the “best age” for this single high-stakes test is, and implement regular computer-based bite-sized testing throughout the primary school years.
The testing should be continuous, it should provide quick and accurate feedback to parents and teachers on children’s areas of strength and weakness so these can be encouraged/ tackled. Children should always be helped to progress at their individual pace, whether they are early or late bloomers.
"Finland doesn't waste time or money on low-quality mass standardised testing. Instead, children are assessed every day, through direct observation, check-ins and quizzes by the highest-quality "personalised learning device" ever created - flesh-and-blood teachers.
(Sydney Morning Herald, 25 Mar 2016: This is why Finland has the best schools)
Trying to reduce or remove exams between P1 and P5, while PSLE is still in place, is a futile exercise. If the KPI/ ultimate measure of success for primary school remains PSLE, then it is human nature to want to know if we are on track to meet our KPI. And so teachers and parents will continue to test. They might just change the name, or do it in the tuition centres. An underground testing economy, if you will. Removing mid-year exams will make end-year exams more high stakes.
“As long as this obsession with grades remains, the removal of mid-year exams will not achieve the shift in focus that MOE is hoping for - since it is likely that weighted assessments will be administered in place of the mid-year exams or a higher weighting will be accorded to the end-of-year exams.”
Kieira Teo, 16. (Straits Times, 14 Mar 2022: Voices of Youth: Mindsets need to change to achieve focus on learning)
In fact, reducing regular testing in schools, without removing PSLE itself, will only hurt the less well-off families. They will be deprived of the regular and realistic feedback on how their child might do in PSLE, and they will have insufficient information to direct their meagre resources at the out-of-school intervention that might benefit their child’s PSLE score the most. And so the income disparities in our secondary schools will worsen.