Research should support teaching, not drive it

“The research shows that …”

For teachers, these words can be difficult to argue with. They are often used to introduce new teaching strategies, professional learning initiatives and system-wide approaches to classroom practice, carrying an implicit assumption that if research has identified a more effective way to teach, teachers should simply adopt it. But research evidence doesn’t always readily translate into classroom practice.

There are very good reasons for education to take research findings seriously. Research has substantially increased our understanding of how students learn, how cognitive load can be managed, and how teachers can support motivation and engagement. The difficulty comes when a nuanced finding, often based on an average effect, begins its journey towards becoming one-size-fits-all advice for every teacher and student.

In a recent commentary published in Learning and Instruction, I examined this research-practice gap from the perspective of having spent more than two decades as a secondary school teacher before moving further into educational research.

The argument is not that teachers should disregard research, but that researchers, policymakers and school leaders need to “go the distance” when translating evidence into practice. Knowing that something works on average is an important starting point; understanding for whom it works, under what circumstances, and how it interacts with a particular teacher’s expertise is considerably more difficult.

What do we mean by educational success?

Academic achievement matters. Schools must ensure students acquire knowledge and skills, and examination results, grades and standardised assessments provide useful information about this. The problem arises when academic achievement becomes synonymous with educational success.

Education is also concerned with students’ motivation, engagement, confidence, effort, perseverance, relationships, self-belief and broader development. A four-year Australian study of several thousand students, for example, found that perceived teaching support played an important role in students’ motivation and engagement trajectories through adolescence.

Think back to a teacher who had the greatest positive influence on you. Why did that person come to mind? Perhaps they helped you achieve outstanding examination results, but for many people the answer will be harder to quantify: a teacher who showed belief in them, gave them confidence, recognised potential they did not yet see, or motivated them to pursue something they had never considered possible.

The significance of that teacher cannot be captured simply by improved examination results. Research has long pointed to important connections between interpersonal relationships and students’ motivation, engagement, effort and achievement.

When I first started teaching, I didn’t think motivating students was part of my role. I thought, if students came to class wanting to learn, I could teach them.

After two decades in secondary classrooms, however, perhaps the most important change in my understanding of teaching was the realisation that motivating students is one of the most important things teachers do.

Effective explanations and well-designed instruction are essential, but students ultimately must invest their own cognitive and behavioural effort in learning. Even an expertly constructed lesson has limited impact on a student who is disengaged.

Motivation therefore sits upstream of many outcomes we eventually measure. An excellent teacher may develop a student’s confidence, curiosity, perseverance and willingness to invest effort. These may contribute to improved achievement, but they are also worthwhile educational outcomes in themselves.

From average effects to real classrooms

Educational research necessarily explores patterns. Large studies and meta-analyses can establish whether an approach has a positive effect overall. Such evidence is extremely useful, but an average effect does not mean every student benefited equally, every teacher implemented the approach equally well, or that the same intervention will be optimal in every classroom.

There is no “average” student walking into an “average” teacher’s “average” classroom on Monday morning. Students differ in prior knowledge, confidence, motivation and readiness to learn; teachers differ in experience, expertise, interpersonal skills and pedagogical approach.

Teaching involves navigating these differences in real time: judging whether students understand, whether to move on, whether a different explanation is needed, or whether a planned activity should be abandoned because something more important has emerged. This adaptability is part of the craft of teaching and develops over many years; it cannot be reduced to a prescriptive set of procedures.

Teaching strategies don’t operate in a vacuum

Even when a particular teaching approach is proving effective, factors beyond a teacher’s control can profoundly influence what happens next.

An example from my own teaching involved a low-achieving Year 8 mathematics class in which many students had diagnosed learning challenges. Engagement was initially low, but gradually improved as I endeavoured to differentiate questions, carefully scaffold concepts and create opportunities for students to experience success.

Then came a common assessment task that all Year 8 classes were required to sit so they could be compared against each other. The test was far beyond the level at which my students were working, but I had no influence to change it.

My class averaged around 20 per cent on the test. Students who had tentatively begun to believe they could succeed were suddenly presented with compelling evidence that they could not. Some felt betrayed because a teacher who had been encouraging them had apparently given them an assessment that they had almost no chance of passing. As a direct result their motivation and engagement collapsed.

No evidence-based instructional technique could have solved that problem the following morning. The issue was not one of appropriate pedagogy, but instead involved assessment policy, student confidence, relationships, motivation and a lack of teacher autonomy. Teachers’ classroom strategies do not operate independently of these conditions.

Evidence-informed, not evidence-dictated

The distinction between evidence informing practice and prescribing it can be seen in the debate around explicit teaching and inquiry-based learning.

There is strong evidence for the benefits of explicit and structured instruction, particularly when students encounter new or complex material.

The Australian Education Research Organisation recommends explicit explanation, demonstration and modelling, while the NSW Department of Education describes explicit teaching as a powerful evidence-based practice but also says teachers should select the appropriate strategy “at the right time for the right purpose”.

That qualification matters. Evidence for a teaching approach does not mean it should dominate every lesson, for every learner, at every stage. Cognitive load research itself demonstrates the expertise reversal effect: guidance that benefits learners with little prior knowledge can become less effective or even counterproductive as expertise increases and students need more autonomy support to become independent learners.

The useful question is therefore not the false dichotomy of whether explicit instruction is “good” and inquiry learning is “bad”, or vice versa. It is: for which students, at what point in their learning, for what content, and how much guidance is needed? These are professional judgements informed by evidence.

Accounting for teacher expertise

There is a further complication when research is translated into practice: teachers themselves are at different stages of development and expertise.

A beginning teacher may benefit enormously from clear guidance about lesson structure, modelling, questioning and classroom routines. An accomplished teacher may already employ these principles fluently while adapting them in ways that are difficult to capture in a standardised framework.

My commentary raises the possibility of something analogous to an expertise reversal effect for teachers themselves.

Recommendations derived from average effects may substantially assist a novice or struggling teacher but offer much less to (or even hinder) an expert practitioner who has already developed an effective approach. For the latter, research may be most useful as a way of understanding, refining and occasionally challenging existing practice. Evidence-informed professional learning should therefore be differentiated, just as good classroom teaching is differentiated.

Teachers’ research fatigue

Experienced teachers also possess an institutional memory of previous debunked research-informed initiatives. “Learning styles” are a clear example of this.

Teachers were encouraged to identify students as visual, auditory or kinaesthetic learners and adjust their instruction accordingly, before subsequent research found no evidence that matching instruction to purported learning styles improved learning. “Multiple intelligences” has faced similar scrutiny and criticism.

This does not justify teachers automatically rejecting new research; science advances because evidence is tested and refined. But it helps explain why “the research shows…” may not always persuade teachers as policymakers expect. Asking how new advice fits with previous wisdom, whether it applies to particular students or classroom contexts, and whether it genuinely improves a teacher’s existing practice are reasonable professional questions to ask.

Translating research into real-world classrooms

Professional judgement uninformed by evidence can result in ineffective practice. Equally, evidence applied without professional judgement can be counterproductive. Teachers need access to high-quality research, but research-based advice intended to influence classrooms also needs to engage seriously with differences between teachers, students and contexts.

To do this, researchers are encouraged to identify the conditions under which effects occur and involve practitioners when translating findings into recommendations. School leaders might endeavour to distinguish between providing evidence and requiring uniform implementation. Policymakers and media might resist turning probabilistic findings into universal prescriptions.

Most importantly, educational success needs to be understood more broadly. Academic achievement remains central, but so should the motivation, confidence, curiosity, effort and relationships that make learning possible and academic growth achievable.

Returning to that influential teacher from your own schooling. Their impact may still be evident decades later in what interests you, what you believe yourself capable of, or the choices you made. Educational research should help more teachers have that kind of impact. Doing so requires more than identifying what works on average and converting it into a list of things all teachers should do. It requires understanding how evidence interacts with teacher expertise, student development and the complex social-emotional environment of a classroom.

That is the difference between research driving teaching and research supporting it.

Dr Robin Nagy is a former teacher and adjunct lecturer at the School of Education, University of NSW.

Do you have an idea for a story?
Email rcox@intermedia.com.au