Fluency Is Not One Thing: Let’s Get Clear About Arithmetic, Computational, and Procedural Fluency

Last week I was at the National Council of Teachers of Mathematics (NCTM) Spring Conference in Indianapolis, and one session in particular stuck with me.

Yvette Lehman from Make Math Moments spoke about fragmentation in math systems. She made a simple but powerful point: clarity can become a barrier.

In many schools and districts, everyone says the same word.

“We’re focusing on fluency.”

But unless people are aligned on what type of fluency they mean (and how they’re defining it) the work stalls.

One team may mean basic fact speed.
Another may mean flexible computational reasoning.
Another may mean procedural accuracy in multi-step algebra.

Everyone agrees on the word.
No one agrees on the definition.

And without alignment, growth is nearly impossible.

You cannot improve what you have not clearly defined.

That conversation at NCTM reinforced something I’ve been thinking about for years:

We use the word fluency constantly in mathematics education.
But we rarely define it precisely.

And when we don’t define it, we end up talking about three very different ideas as if they are interchangeable.

So let’s get clear.

There is a difference between:

  • Arithmetic (basic fact) fluency

  • Computational fluency

  • Procedural fluency

They are related. They are layered. But they are not the same.

And that distinction matters.

The Relationship of Different Fluency Terms in Mathematics from Bay-Williams & SanGiovanni, 2021

Basic (Arithmetic) Fact Fluency: A Critical Foundation (But Only a Foundation)

Arithmetic fact fluency refers specifically to the quick, accurate retrieval of basic single-digit arithmetic combinations and the ability to use that knowledge efficiently.

In our peer-reviewed article in Teaching Exceptional Children, Combining Explicit Strategy Instruction and Mastery Practice to Build Arithmetic Fact Fluency, my co-authors and I focused on this narrow but important slice of mathematics learning.

Yes, students should be able to retrieve basic facts accurately within a few seconds. But more importantly, they must be able to use those facts efficiently in service of larger mathematical thinking.

Fact fluency is not just speed. It is strategic retrieval that supports more complex reasoning.

Arithmetic fact fluency matters because it frees up working memory. As LeFevre et al. (2005) explain, when students expend all of their cognitive energy solving basic combinations, they have less capacity available for higher-level reasoning and conceptual understanding.

But here’s the key:

Arithmetic fact fluency is only one small subset of something much larger.

Computational Fluency: Bigger Than Basic Facts

Computational fluency extends far beyond single-digit combinations.

It includes:

  • Multi-digit operations

  • All four operations

  • Whole numbers, fractions, decimals, and integers

  • Flexible strategy selection

  • Efficiency and accuracy across contexts

A student can know their basic facts and still lack computational fluency.

For example, a student might quickly recall 7 × 8 but struggle to reason through 0.75 × 32. That is not a fact fluency issue. That is a broader computational fluency issue involving number sense, proportional reasoning, and flexible strategy use.

When schools equate fluency with “fast facts,” we narrow instruction in ways that do not prepare students for real mathematical demands.

Procedural Fluency: The Broader Strand

The National Research Council’s Adding It Up identifies procedural fluency as one of the five strands of mathematical proficiency. Procedural fluency includes carrying out procedures:

  • Accurately

  • Efficiently

  • Flexibly

The Five Strands of Mathematical Proficiency from the National Research Council, 2001

This goes beyond recall and beyond computation. It includes multi-step algorithms, symbolic manipulation, and strategic selection of methods.

Procedural fluency is part of a larger ecosystem of mathematical proficiency that also includes conceptual understanding, strategic competence, adaptive reasoning, and productive disposition.

When we zoom out, we see the layering:

  • Arithmetic fact fluency supports

  • Computational fluency, which supports

  • Procedural fluency, which sits inside

  • Mathematical proficiency as a whole

That hierarchy matters for instruction.

Why This Distinction Matters in Classrooms and MTSS

When we blur these categories, several problems show up:

  • We overemphasize timed drills without strategy development

  • We label students as “not fluent” without clarifying what kind of fluency is missing

  • We design interventions that target speed instead of understanding

  • We misalign Tier 2 and Tier 3 supports

In an MTSS framework, clarity is non-negotiable. If we are going to design effective Tier 1 instruction and targeted intervention, we must know exactly what skill set we are addressing.

If the issue is arithmetic fact fluency, explicit strategy instruction paired with mastery practice may be appropriate.

If the issue is computational fluency across number sets, the solution is different.

If the issue is procedural fluency within algebraic reasoning, that is different again.

Precision matters.

I unpack more about how instructional clarity and assessment alignment fit within MTSS in this post on Every Student, Every Tier: Making MTSS Work in Elementary Math. When we are clear about what we are measuring and why, intervention becomes far more effective.

So What Should We Do?

  1. Stop using “fluency” as a catch-all

  2. Be explicit about which layer of fluency we are targeting

  3. Align instruction and assessment to the right level of need

If we are talking about arithmetic fact fluency, then we should use research-backed approaches like explicit strategy instruction followed by purposeful mastery practice. That work is grounded in evidence and well documented.

If we are talking about computational or procedural fluency, then we must widen the lens.

Fluency is not one thing.

And our students deserve more than vague language.

They deserve precision.

Previous
Previous

Motivation Matters in Mathematics

Next
Next

Which One Doesn't Belong (WODB)