Evidence for What?

There was an article published recently in The Age, titled ‘This style of teaching gets results, say governments. But resistance to it is growing’, reporting an apparent backlash against Australia’s 'evidence-based' teaching reforms.



For context, there are two key things being referred to as ‘evidence-based' teaching reforms. The adoption of direct instruction in schools and uniform standards for teacher training at universities. Those backing the reforms said previous approaches were not delivering and not grounded in evidence – bring on the reforms, here’s the evidence. Those pushing back say the reforms are too restrictive, too narrow and they impede purposeful, intuitive teaching and long-term learning. Have another go.

The article quotes Jenny Donovan, chief executive of the Australian Education Research Organisation (AERO), who strongly supports the reforms and insists that teachers and policymakers must 'stay the course' despite difficulties in implementing the large-scale changes amid growing, vocal pushback. 'We're humans, we get distracted really easily,' she said. Also quoted is Nikki Brunker from the University of Sydney, a leading critic of the reforms, or, as she calls it, the new 'orthodoxy.' She argues that the current evidence-based agenda has become too prescriptive: ‘Teachers and school leaders are the ones on the ground seeing the problems arise as a result of trying to follow evidence-based practice.’ Her stance feels more robust than a case of being 'easily distracted.' But let's push on.



I want to clarify what 'evidence' means here because the term is doing a lot of work in this conversation. Evidence for what? Which questions are being asked, what counts as proof and who decides? Is it a scientific question, a philosophical one, or both?


Let's acknowledge what the new evidence does show. First, the focus on assessment and measurable outcomes is in response to lower literacy and numeracy rates than we'd like for our kids. Let’s be clear: the research underpinning systematic phonics instruction, or decoding, is strong. Whether we're parents, teachers or policymakers, the results were not cutting it – and now they have improved. Too many students were leaving primary school unable to read fluently and that had to change. When Donovan suggests education has not traditionally been driven by science – and that AERO is changing that – it feels, as a parent or a classroom teacher, like something we should get on board with. We all want what is best for young people.
 
But evidence is only as good as the question it’s answering.



In this case, the primary evidence used is NAPLAN data. Students are assessed, categorised and tracked throughout their school lives. If one apple falls below the others, or sits above, that is hard evidence right there. It tells us something concrete, and it’s useful. Within the confines of a question such as 'Does this approach improve short-term, measurable cognitive outcomes?', the answer is likely – yes. Phonics improves decoding and explicit teaching can lift performance on specific, targeted skills. These are true and real findings.



But thinking critically for a moment, if you ask a group of people to choose the best ice cream flavour between vanilla and chocolate, one flavour will come out ahead. But if the choice doesn't include the thousands of other flavours that exist, the evidence is accurate, but its value is limited entirely to that question. It tells us nothing about what wasn't included. I see you, salted caramel.



As Brunker puts it, 'When we look at what evidence-based practice is, it's so narrow.'

NAPLAN data doesn't tell us what a teacher sees when a child who has been disengaged for months suddenly engages because the learning landed in the right way. It doesn't help us understand the long game of a young person’s relationship with learning, their capacity to connect ideas and think beyond the here and now. It doesn’t account for the sustained motivation that flows from autonomy, belonging and competence; the three psychological needs at the heart of Self-Determination Theory, developed across decades of research in educational settings worldwide.



What it definitely doesn't capture is what Brunker and her colleagues are making clear: the disengagement that teachers experience when their professional purpose becomes flattened and prescriptive. Teachers are not immune to the effects of controlling mandates. Their motivation, too, is shaped by whether they feel trusted and autonomous in their work. 



Rod Davies, Director Primary and Early Years at Woodleigh School, recently wrote about what he calls the false binary that exists between explicit instruction and inquiry-based learning. He says that, in his decades of experience in educational leadership, the universal truth is that the strongest teachers have the skills to read the room and move between methods as the moment requires. These teachers hold two kinds of knowledge at once: a sound understanding of research and an even deeper understanding of the young people in front of them. They know when to follow the plan and when to blow it up.



This shocked me; around one in four Australian students does not complete Year 12. There are countless reasons, but NAPLAN markers do not tell us what they are. Student engagement declines as they progress through 13 years of schooling. This is widely documented and was described as alarmingly consistent in a 2025 study published in the Medical Journal of Australia. It found that students feel the system values conformity over creativity, forcing them along uninspired pathways rather than allowing them to pursue their own interests and curiosities.



It doesn't have to feel like that, though. Students in classrooms that meaningfully support mentored autonomy can buck that trend entirely, with engagement increasing rather than declining – if that is the goal.