What Is a Mathematical Model? — Secondary Mathematics Guide

Students often think a model is simply “the formula in a word problem”. That is too narrow. A model begins before the formula. Someone must decide which quantities matter, how they relate, what assumptions are acceptable, and what outputs would count as useful.

This makes modelling one of the best places to connect arithmetic, algebra, graphs, geometry, statistics and reasoning. A model is where mathematics leaves the page, enters a situation, and then returns with an answer that must still be interpreted.

Models are useful because reality is complicated

Suppose a taxi-style pricing problem says a journey costs $4 initially plus $2 per kilometre. The model C=4+2d ignores traffic, waiting time, surge pricing, route differences and many other real details. That is not necessarily a weakness. If the question only asks how the stated price changes with distance under the simplified tariff, those extra details may be irrelevant.

A good model is not the most detailed possible description. It is a description detailed enough for the question being asked.

Every model has four parts

1. Quantities: what is being measured or counted?
2. Relationships: how do those quantities depend on one another?
3. Assumptions: what are we treating as fixed, negligible or idealised?
4. Purpose: what decision, estimate, explanation or prediction should the model support?

Without the fourth part, students may calculate correctly without knowing whether the answer is useful.

A model can be verbal, numerical, graphical or algebraic

“Cost increases by $3 per item after a fixed $10 charge” is a verbal model. A table of item counts and costs is a numerical model. A straight-line graph is a graphical model. C=10+3n is an algebraic model.

These representations can describe the same relationship. Strong modelling moves between them deliberately rather than treating each as a separate topic.

Example 1: constant speed

Suppose a vehicle travels at a constant 60 km/h for t hours. A suitable model is d=60t. The model assumes constant speed and treats distance travelled as proportional to time.

At t=2.5, the model predicts d=150 km. If the actual vehicle slows, stops or changes route, the model no longer describes the real journey exactly. It may still be useful over a period where constant speed is a reasonable approximation.

Example 2: cooling with a crude linear approximation

Imagine a liquid starts at 80°C and a simplified classroom exercise says it cools by 4°C each minute for five minutes. The model T=80−4t works over the stated interval 0≤t≤5.

It would be poor modelling to extend the rule indefinitely: at t=30 it predicts −40°C, even though the real cooling process and surrounding conditions would matter long before then. A formula can be mathematically valid while the model domain is physically inappropriate.

Example 3: area as a model of material needed

If a rectangular floor has length L and width W, area A=LW can model the amount of surface to cover. If each tile covers 0.25 m², an idealised tile count begins with A/0.25.

Real tiling may require extra material for cuts and wastage. The pure area calculation is a first model; a better practical model may include a wastage allowance.

A model is not the same as reality

The statement “the model predicts 120” should not be rewritten mentally as “reality must be 120”. Models are conditional. Their conclusions depend on assumptions, data and the mathematical relationships chosen.

This is why model answers should often be followed by interpretation: “Under the constant-rate assumption, the predicted time is 8 minutes.”

Models can be exact within a defined system

Not every model is approximate. If a game defines a score as S=5w+2d where w and d are exact counts under fixed rules, the formula can exactly model that scoring system. Approximation enters when the system itself is a simplification of a more complicated reality.

The role of assumptions

Assumptions are statements that make a model manageable. They may include constant speed, uniform density, no leakage, identical items, negligible thickness, independent events, or a stable percentage growth rate.

Assumptions should be explicit because they determine when the model can and cannot be trusted.

The role of variables

A variable is not merely a letter. It stands for a quantity whose value can change or is unknown. Choosing variables well compresses the situation.

For a rectangle, L and W are more informative than x and y if the meanings matter. In a generic algebra proof, x and y may be enough. Good notation serves the model.

The role of parameters

In C=a+bn, n may vary while a and b are fixed within one version of the model. Here a is a fixed charge and b a unit rate. They are parameters controlling the relationship.

Changing a or b produces a new member of the same model family. This prepares students for later function and statistics work.

When a model is useful

• To estimate a quantity that is difficult to measure directly.
• To compare scenarios.
• To predict outcomes under stated assumptions.
• To identify which variables matter most.
• To explain a pattern.
• To support a decision.

A complete modelling cycle

1. Understand the situation.
2. Choose quantities and variables.
3. State assumptions.
4. Build a mathematical representation.
5. Calculate or analyse.
6. Interpret the result in context.
7. Check whether the model is reasonable.
8. Revise if needed.

The last step matters. Modelling is iterative. A model can be improved when evidence shows that an assumption is too crude.

Common misconceptions

“A model is just a formula.” A formula is one possible representation inside a larger modelling process.

“More detail always makes a model better.” Extra detail can make a model harder to use without improving the answer to the intended question.

“If the mathematics is correct, the model is correct.” A mathematically correct calculation can still rest on poor assumptions.

“A model that fails once is useless.” Failure may reveal which assumption needs improvement.

How we would teach this from first principles

Start with familiar situations and ask students to remove details deliberately. What matters in a simple journey model? Distance, time and speed may matter; colour of the car does not. Then ask what assumption turns the situation into a manageable equation.

Next, give two possible models for the same situation and ask which is more suitable and why. This shifts modelling from formula retrieval to judgement.

Diagnostic checkpoints

• Can the student identify the quantities in a situation?
• Can they state an assumption explicitly?
• Can they distinguish a model from reality?
• Can they explain the meaning of a parameter or rate?
• Can they interpret a numerical result in context?
• Can they suggest one reason a model might fail?

Practice prompts

1. A tank fills at 8 L/min from empty. Write a model. Answer: V=8t under constant-flow assumptions.

2. A service charges $12 fixed plus $5 per use. Write a model. Answer: C=12+5n.

3. Give one reason V=8t may fail for a real tank. Possible answer: flow rate may not remain constant or the tank may reach capacity.

4. A rectangular field has area 600 m². Is L×W=600 a model? Yes: it captures the area constraint while leaving many possible dimensions.

The deeper mathematical habit

Modelling teaches students that mathematics is not only about solving equations supplied by someone else. It is also about deciding which equation should exist in the first place.

Continue to From a Real Situation to Variables, Assumptions and Equations, How to Test a Mathematical Model and When a Model Fails. Return to the Secondary Mathematics Master Index.