Why Precision Matters in Mathematics Instruction
By Esther White, CEO & Founder, Maths Australia
Walk into two classrooms on the same morning. Both teachers are using explicit maths instruction. Both have manipulatives out on the tables. Both are drawing on evidence-informed approaches such as Cognitive Load Theory, the Concrete-Representational-Abstract (CRA) approach, and mastery learning. On paper, you'd struggle to tell them apart.
By the end of the year, one group of students demonstrates secure mathematical understanding, while the other is still relying on guesswork.
I've watched this happen more times than I can count, over more than 25 years of working alongside teachers, tutors, intervention specialists and parents across Australia. It's almost never a case of one teacher doing it "right" and another doing it "wrong." Most of the teachers I meet are careful, thoughtful and genuinely invested in their students getting there. What separates the two classrooms isn't which approach they've chosen. It's the precision behind how that approach is actually carried out, lesson by lesson, term by term.
That, in a nutshell, is what I mean by precision in mathematics instruction: not a new principle to add to an already long list, but the care taken in applying the evidence-informed principles teachers already know are important.
Precision in mathematics instruction isn't about perfection or rigid teaching. It's about being deliberate. It means choosing mathematical language carefully, connecting representations explicitly, sequencing learning thoughtfully, and making important ideas visible rather than assuming students will discover them for themselves.
It's not a small point. In my experience, it's the whole point. Sometimes the difference really is found in the subtleties.
Maths Is a Language
I say this often enough that people who know me could probably finish the sentence: maths is a language. I don't mean that as a nice turn of phrase. Mathematics has its own vocabulary, its own symbols, and its own conventions and structures for representing and communicating an idea. A fraction bar means something specific. An equals sign means something specific. Even a small word like “of”, as in “a third of twelve”, carries real mathematical weight.
Precise mathematical language reduces ambiguity. When a term is used consistently and accurately, a student isn't left guessing at what a teacher, a textbook or a test question actually means, and that precision frees up attention for the mathematical thinking itself.
Mathematical understanding also depends on consistent exposure to that vocabulary, much the way a toddler learns to talk because the same words mean the same things, over and over, until the pattern is secure enough to build on. Mathematics teaching benefits from a similar kind of consistency, not because variety is a problem, but because a student needs a stable base of meaning before variety becomes useful rather than confusing.
A maths classroom works the same way, whether we plan for it or not. If one teacher describes division as “sharing” and another describes it as “how many groups”, and nobody ever draws an explicit line between the two, a student is left to work out that translation alone. Some manage it. Plenty don't, and quietly decide maths isn't for them, when the real issue was never the mathematics. It was that the language and representations around it never quite lined up.
I explore this idea further in my TEDx talk, Why Maths Is A Simple Language. Taught with precision, maths is one of the more logical, learnable subjects a student will meet. It becomes far harder to access when the words, symbols and representations around it keep shifting without explanation.
Why the Small Things Trip Students Up
It's worth being specific about what this kind of inconsistency actually looks like in a classroom, because it rarely looks dramatic. Nobody is doing anything obviously wrong. It's smaller than that.
A teacher uses base ten blocks for place value in Term 1, then moves to a different material in Term 2 without ever linking the two. A worksheet uses slightly different wording for a strategy than the one written on the board that morning. One resource labels the “whole” in a fraction model differently to the next. Each choice, on its own, seems harmless.
A student with a secure, flexible grasp of the concept barely notices these small shifts. They can translate between one representation and the next without much effort, because the understanding underneath is strong, and this is actually one of the reasons multiple representations are valuable in mathematics teaching. A student who understands a concept deeply can move between a number line, an array and an equation with relative ease, and that flexibility is itself a sign of genuine mathematical understanding.
For a student who's still building that understanding, and at any given time, that's a considerable number of students in a classroom, every shift in language or model asks something of working memory. Cognitive Load Theory is useful here as an explanation rather than a slogan. Working memory has limited capacity, and when a student has to constantly reinterpret changing language or representations before they can even begin thinking about the mathematics itself, unnecessary cognitive load increases, leaving fewer resources available for the understanding the task was designed to build.
Misconceptions often take root in exactly this gap. Rather than a sign that a student hasn't been paying attention, a misconception is usually a sign that they've built a reasonable but incomplete explanation from what they've seen and heard, because the relationship between one representation and the next was never made explicit. The student isn't wrong to think the way they do. They simply haven't yet been shown how the pieces connect.
Variety isn't the problem here. Multiple representations, used well, are one of the more powerful tools in mathematics teaching. The trouble starts when that variety arrives without an explicit bridge between the old and the new. Precision has little to do with limiting a teacher to one model or one form of words, and everything to do with doing that connecting work out loud, rather than assuming a student will find it unaided.
What the Research Actually Says
None of this is new or fringe. It sits inside a well-established evidence base that most experienced teachers will already recognise, even without the academic labels attached.
Rosenshine's Principles of Instruction, synthesised from decades of classroom research, describe a set of practices, clear modelling, carefully sequenced instruction, guided practice and checking for understanding before moving on, that together create coherent learning. A worked example only supports learning if the next one follows a logic the student already recognises, and guided practice only builds fluency if it uses the same language and structure the student has just been taught.
Explicit instruction, as the research describes it, isn't simply “the teacher talks and the students listen.” Done well, it's clear, deliberately sequenced and largely free of ambiguity, which matters most for students who are still novices with a concept and don't yet have the background knowledge to patch over gaps themselves. Until that expertise develops, unclear or inconsistent teaching can place a real burden on students that has little to do with the mathematics itself.
Research into mastery learning suggests a similar principle: consolidate understanding of a concept before moving to the next, because later concepts are frequently built directly on top of earlier ones. Retrieval practice and cumulative review support this by keeping earlier learning active, rather than filed away after a unit test and rarely revisited. Much of this also aligns with how schools approach Response to Intervention, where effective support depends on knowing precisely what a student does and doesn't yet understand, rather than assuming it from a year level or a single test score. Numeracy intervention built on that accurate picture tends to be more targeted, and more likely to close a gap than paper over it.
Although these bodies of research emerged from different traditions, they point to a common principle: students learn best when instruction is carefully sequenced, mathematical language is used consistently, and important connections are made explicit rather than left for students to infer.
Maths Doesn't Forget What Came Before
It's easy to say “maths builds on itself” and move on. It's worth sitting with what that actually means for a lesson, and for the student sitting in front of you.
A student can't understand equivalent fractions with real security if their underlying sense of what a fraction represents is still shaky. They can't work confidently with equations if the equals sign was only ever taught as “the answer goes here,” rather than as a symbol of balance. Later mathematics doesn't sit beside earlier mathematics. It's built directly on top of it, and it depends on that foundation holding weight.
A gap from years earlier often resurfaces at the worst possible time, looking like a brand-new problem when it isn't. A Year 8 student struggling with algebra may well be struggling because of an incomplete understanding of multiplicative relationships formed three or four years earlier. Precision in the early years tends to pay off later. Imprecision tends to resurface, often right before an assessment, when there's the least time available to address it.
Knowing exactly where a student's understanding currently sits matters more than knowing what year level they're officially in. Two students in the same class can be standing on quite different foundations, and teaching them as though they're in the same place can leave one without the support they need. This, in many respects, is the starting point of any well-run maths intervention: an honest, specific picture of what a student already understands, rather than an assumption based on age.
Hands-On Only Works When It's Going Somewhere
Manipulatives have earned their place in mathematics teaching. The Concrete-Representational-Abstract (CRA) approach is supported by a substantial body of research, and tends to be most effective when students move purposefully from physical materials, to an accurate drawn representation, to an abstract symbol, rather than encountering each stage as a separate, disconnected activity.
That word “purposefully” is doing a lot of work in that sentence.
Hands-on learning does real work when it's used consistently enough that the materials become part of the mathematical language a student is learning, and explicitly linked to the representation and symbol that follow it. This doesn't mean a teacher should only ever use one set of materials. It means that whatever materials are used, the connection between what a student built, what they drew, and what they're now asked to read as a symbol needs to be made clear, rather than assumed.
Hands-on learning turns decorative rather than instructional the moment it becomes an activity to complete, rather than one stage in a sequence heading toward abstraction. A student can enjoy using the blocks, finish the task, and still never connect what they built to the symbols on the page. The materials themselves were never really the point. What they're used to reveal, consistently and on purpose, is the point.
You Can't Rush What Isn't Built Yet
There's a real pull, especially with a scope and sequence document sitting on the desk, to keep moving even when a student hasn't yet consolidated the concept in front of them.
But mathematics doesn't pay much attention to pacing guides. A student who moves on to multiplication without a secure grasp of what multiplication actually represents isn't being helped along by that progression. They're being asked to build on a floor that isn't load-bearing yet.
Mastery approaches emphasise something worth holding onto: getting there matters more than getting there quickly. A concept needs to be genuinely understood, not just correctly ticked off on a worksheet, before the next layer goes on top. A student can produce correct answers from a memorised procedure without holding the understanding needed to adapt it when a problem changes shape. Research in this area suggests both conceptual understanding and procedural fluency matter, and that one doesn't reliably produce the other.
Simple Isn't the Same as Simplistic
There's a common assumption that more complex resources, more varied activities and more moving parts in a lesson add up to better teaching. In my experience, it's often the other way round.
None of this means using only one representation or one resource. What matters is that whatever representations and resources are used are chosen deliberately and explicitly connected, rather than introduced one after another with no bridge between them. A lesson built around two or three carefully connected representations, precise language and a clear model can easily outperform one stitched together from five different resources with no explicit links, because it leaves a student's working memory free to focus on the mathematics itself, rather than reconciling what's just changed.
This isn't a case of teachers overcomplicating things. Most are working inside real constraints: mixed-ability classrooms, packed curriculum documents, resources never designed with consistency across year levels in mind, and rarely enough time to plan collaboratively as a whole staff. When students experience inconsistency, it's rarely down to any individual teacher's choices. More often it's structural, a school that hasn't yet settled on shared mathematical language and shared models across year levels.
What teachers and schools need isn't necessarily more elaborate resources. They need a framework that is precise and consistent enough that what a student learns about fractions in Year 2 still makes sense - and still connects - when they encounter fractions again in Year 6.
How I Bring This Together
The I-CRAVE Maths® Methodology grew out of watching evidence-informed principles become disconnected between one classroom, one resource, or one year level and the next. Rather than presenting a new theory of mathematics instruction, it provides a practical implementation framework that brings together explicit instruction, the Concrete-Representational-Abstract (CRA) approach, mastery learning and other evidence-informed practices into one coherent, teachable methodology for everyday mathematics instruction.
Take a student working on fractions. Before anything else, I want to know exactly where their understanding starts, not what a syllabus says a Year 4 student should already know. That's Identify, and skipping it is one of the more common ways a gap gets carried forward instead of closed.
From there, the student builds the concept with real materials, physically making a whole and splitting it into equal parts. That's Concrete. Next, they draw what they've just built, proportionately and accurately, so the picture on the page still represents the same idea as the blocks on the table. That's Representation. Only once that connection is solid do we introduce the numerator and denominator, linking the symbols back to something the student has already built and drawn with their own hands. That's Abstract.
Then I ask the student to explain it back to me, in their own words but using accurate mathematical language. That's Verbal, and it's often where I find out whether the earlier steps genuinely landed, or just looked like they did on the day. A student who can explain why a fraction works, in language that matches how a mathematician would describe it, understands something meaningfully different from a student who can only repeat the steps.
Underneath all of it sits Explicit teaching: the same language, models and sequence, whether it's this lesson or next term, this teacher or the one next door. That consistency is what stops a student from having to do private translation work between one part of their mathematics education and the next, and it's what allows explicit instruction, CRA, mastery learning, cumulative review and mathematical language to survive contact with an actual timetable and an actual staffroom, rather than staying theory on a page.
The Subtleties Are Not Small
It's tempting to treat language, sequence, modelling and representation as the fine print of mathematics teaching, details that matter less than the big decisions about pedagogy. I'd argue it's the other way round.
Most experienced educators already agree on the big picture. Explicit instruction, purposeful concrete learning, mastery before progression, cumulative review: none of that is particularly controversial. What separates the classrooms, schools and systems that get strong, lasting outcomes from those that don't is rarely a disagreement about these principles. It's the precision behind how they're carried out, day after day, across every year level a student moves through.
Precision in mathematics instruction isn't a separate strategy to layer on top of everything else a teacher is already doing. It's the quality that determines whether explicit instruction, CRA, mastery learning and consistent mathematical language actually deliver the mathematical understanding they're capable of building. The subtleties aren't small details sitting on top of good teaching. They're what make genuine understanding possible in the first place.
If you're looking to strengthen mathematics teaching through clearer, more precise and evidence-informed practice, Maths Australia offers free professional learning and information sessions, comprehensive educator training, and accreditation pathways built around the I-CRAVE Maths® Methodology. To learn more, visit mathsaustralia.com.au/training.