Comparing Linear vs. Exponential Models

SECTION 1

Historical Context & Motivation

Humans have always tried to predict the future — how many crops a field will yield, how fast a population will grow, or how much money an investment will earn. For most of history, people assumed that quantities grew at a constant rate, adding the same amount each year. This assumption worked well for simple situations like tracking wages or distances traveled. However, as mathematicians studied phenomena like compound interest, population booms, and the spread of disease, they realized that some quantities don't just add — they multiply. Understanding the difference between these two types of growth became one of the most powerful tools in mathematics.

~300 BCE

Euclid's Elements

Euclid described arithmetic sequences (adding a constant) and geometric sequences (multiplying by a constant), laying the groundwork for linear and exponential thinking.

1614

Napier Invents Logarithms

John Napier published tables of logarithms, creating a tool to work with exponential relationships and making multiplicative growth easier to study.

1798

Malthus on Population

Thomas Malthus argued that human populations grow exponentially while food supply grows linearly, sparking debates about sustainability that continue today.

2020

COVID-19 Modeling

During the pandemic, the world saw exponential growth in action. Public health officials used exponential models to predict case counts and justify interventions.

The central question this lesson addresses is deceptively simple: given a real-world situation with changing quantities, how do you decide whether a linear or exponential model fits best? And once you choose, how do you justify that choice with mathematical evidence? These skills are essential not just for passing a test, but for making sense of data you encounter every day — from social media follower growth to car depreciation.

SECTION 2

Core Principles & Definitions

Before comparing models, you need to understand what each one describes. A linear model represents a situation where a quantity changes by the same amount in every equal time period. An exponential model represents a situation where a quantity changes by the same factor (or percentage) in every equal time period. That single distinction — adding versus multiplying — is the key to everything in this lesson.

Constant Rate of Change

In a linear model, the slope (rate of change) stays the same. If you earn $50 per week, that's $50 added every single week — no more, no less.

Constant Multiplier

In an exponential model, the quantity is multiplied by the same growth factor each period. If a population doubles every year, that's ×2 each year — the amount added keeps increasing.

First Differences vs. Ratios

To identify a linear pattern in data, check if first differences (successive outputs subtracted) are constant. For exponential, check if consecutive ratios (successive outputs divided) are constant.

Graphs Tell the Story

A linear model produces a straight line. An exponential model produces a curve that starts slow and then accelerates (growth) or starts fast and then levels off (decay).

✦ KEY TAKEAWAY

Think of it like walking versus rolling a snowball. When you walk, you cover the same distance every minute — that's linear. When you roll a snowball downhill, it picks up more snow as it gets bigger, so it grows faster and faster — that's exponential. The question isn't which model is 'better' — it's which one matches the pattern of change in your situation.

SECTION 3

Visual Explanation — Linear vs. Exponential on a Graph

Both models start at the same value (50), but the linear model (blue) adds 15 each period while the exponential model (pink) multiplies by 1.4 each period. Notice how they start similarly but diverge dramatically over time.

The diagram above reveals the fundamental visual difference between these two models. The linear function (blue) forms a perfectly straight line — it grows by the same vertical amount between every pair of consecutive points. The exponential function (pink) starts out looking almost flat, but then curves upward more and more steeply. Early on, the two models produce similar values, which is exactly why it can be tricky to tell them apart from just a few data points. However, over enough time, the exponential model always overtakes the linear model, no matter how large the linear slope is. This visual distinction — straight line versus accelerating curve — is your first clue when deciding which model fits a data set.

SECTION 4

Mathematical Framework

Each model has a standard equation form. Knowing these forms lets you write an equation from a context, read off key information, and make predictions.

LINEAR MODEL

y = mx + b

m = slope (constant rate of change — how much y changes per unit increase in x); b = y-intercept (the starting value when x = 0).

EXPONENTIAL MODEL

y = a · bˣ

a = initial value (when x = 0); b = growth/decay factor. If b > 1, the model shows growth. If 0 < b < 1, the model shows decay.

How to Identify the Model from a Table

FIRST DIFFERENCES (LINEAR TEST)

Δy = y₂ − y₁

Subtract consecutive y-values. If all first differences are equal (for equally spaced x-values), the data is linear.

CONSECUTIVE RATIOS (EXPONENTIAL TEST)

r = y₂ ÷ y₁

Divide consecutive y-values. If all ratios are equal (for equally spaced x-values), the data is exponential. This common ratio becomes the base b in y = a · bˣ.

⚠️ Important Note

The first-differences and ratios tests only work when the x-values are equally spaced (e.g., x = 0, 1, 2, 3, … or x = 0, 5, 10, 15, …). If the spacing is uneven, you need to adjust your calculations or use regression tools.

SECTION 5

Identifying the Model from Data

Let's see both tests in action with two data tables side by side. Study each table and notice how the differences and ratios reveal the pattern.

Data Set A has constant first differences of +6, confirming a linear pattern. Data Set B has constant ratios of 3, confirming an exponential pattern. The decision rule at the bottom summarizes the approach.

Look carefully at Data Set B. The first differences are +10, +30, +90, +270 — they are clearly not constant, which rules out a linear model. But every ratio is exactly 3.0, which tells us each output is three times the previous one. This is the hallmark of exponential behavior. Meanwhile, Data Set A has perfectly constant differences of +6 and varying ratios, so it's linear. In practice, real-world data rarely gives you perfectly constant differences or ratios, but one pattern will be much more consistent than the other. That's your signal.

SECTION 6

Worked Example

A biologist is tracking the number of bacteria in a lab sample. She records the count every hour. Which model — linear or exponential — is more appropriate, and what is the equation?

Bacteria count every hour

Hour (x)	Bacteria (y)
0	200
1	300
2	450
3	675
4	1012.5

Choosing and Writing the Model

Step 1 — Compute First DifferencesSubtract consecutive y-values: 300 − 200 = 100, 450 − 300 = 150, 675 − 450 = 225, 1012.5 − 675 = 337.5.

First differences: 100, 150, 225, 337.5 → NOT constant

Step 2 — Compute Consecutive RatiosDivide consecutive y-values: 300 ÷ 200 = 1.5, 450 ÷ 300 = 1.5, 675 ÷ 450 = 1.5, 1012.5 ÷ 675 = 1.5.

Consecutive ratios: 1.5, 1.5, 1.5, 1.5 → CONSTANT

Step 3 — Choose the ModelBecause the consecutive ratios are constant, the data fits an exponential model. The common ratio is 1.5, meaning the bacteria population increases by 50% every hour.

Step 4 — Write the EquationThe initial value (when x = 0) is a = 200. The growth factor is b = 1.5. Substituting into y = a · bˣ:

y = 200 · 1.5ˣ

Step 5 — Verify with a PredictionCheck x = 4: y = 200 · 1.5⁴ = 200 · 5.0625 = 1012.5. This matches the data, confirming the model.

y(4) = 1012.5 ✓ — model confirmed

SECTION 7

Linear vs. Exponential — Side by Side

Comprehensive comparison of linear and exponential models

Feature	Linear Model	Exponential Model
Equation form	y = mx + b	y = a · bˣ
Type of change	Additive (constant amount)	Multiplicative (constant factor)
Graph shape	Straight line	Curve (J-shape or decay)
Data test	Constant first differences	Constant consecutive ratios
Real-world examples	Hourly wages, distance at constant speed, water filling a pool at steady rate	Compound interest, population growth, radioactive decay, viral spread
Long-term behavior	Grows (or shrinks) steadily without bound	Growth: skyrockets; Decay: approaches 0 but never reaches it
Key parameter	Slope (m) = constant rate of change	Growth factor (b); growth rate = b − 1

✦ KEY TAKEAWAY

When choosing a model, ask yourself one question: "Does this quantity change by a fixed amount or by a fixed percentage?" A $200/month raise is linear. A 3% annual raise is exponential — even though it sounds small, over decades the exponential raise produces far more total income. Context and the pattern of change are what drive the choice, not just the numbers.

SECTION 8

Connection to Advanced Models

Linear and exponential models are the two most fundamental types of growth you'll study, but they're also the building blocks for more complex models you'll encounter in later math courses. Understanding their differences now will prepare you for several advanced topics.

How this topic connects to future coursework

What You Know Now	Where It Leads
Linear model: y = mx + b	Systems of linear equations, linear programming, and linear regression (line of best fit)
Exponential model: y = a · bˣ	Logarithmic functions (the inverse of exponentials), the natural base e, and continuous growth models
Comparing models with data	Regression analysis and residuals — statistical methods to find the best-fitting model among many candidates
Growth factor b > 1 or 0 < b < 1	Logistic growth models that combine exponential growth with a carrying capacity (used in ecology and epidemiology)

One especially interesting extension is the logistic model, which starts out looking exponential but eventually levels off at a maximum value called the carrying capacity. This model describes many real populations more accurately than a pure exponential, because resources are limited. You'll also eventually learn about logarithms, which let you "undo" an exponential equation and solve for the exponent — a skill that's crucial in chemistry, physics, and finance.

SECTION 9

Practice Problems

PROBLEM 1 — CONCEPTUAL

A savings account earns $25 in interest every month, regardless of the balance. Would you model the account balance over time with a linear or an exponential function? Explain your reasoning.

PROBLEM 2 — BASIC CALCULATION

A table shows: x = 0, y = 8; x = 1, y = 24; x = 2, y = 72; x = 3, y = 216. Determine whether the data is linear or exponential, and write the equation.

PROBLEM 3 — INTERMEDIATE

Two companies offer you a job. Company A starts at $40,000/year and gives a $2,000 raise each year. Company B starts at $35,000/year and gives a 6% raise each year. Which company pays more in Year 10? Write both models and evaluate them at x = 10.

PROBLEM 4 — APPLIED

A car is purchased for $28,000. Its value over the next four years is recorded as: Year 0: $28,000; Year 1: $23,800; Year 2: $20,230; Year 3: $17,195.50; Year 4: $14,616.18. Is the depreciation linear or exponential? Write the model and predict the car's value in Year 7.

PROBLEM 5 — CRITICAL THINKING

A student examines a data set and finds that both first differences and consecutive ratios are approximately constant. She's confused. How can both seem constant, and what should she do to decide between a linear and exponential model? Provide a clear strategy.

SUMMARY

Summary

A linear model (y = mx + b) describes situations where a quantity changes by a constant amount each period, producing a straight-line graph with constant first differences. An exponential model (y = a · bˣ) describes situations where a quantity changes by a constant factor each period, producing a curved graph with constant consecutive ratios. When b > 1, the exponential model shows growth; when 0 < b < 1, it shows decay.

To choose the right model, compute both first differences and consecutive ratios from a data table (with equally spaced inputs) and see which set is constant. Always consider the real-world context — fixed dollar amounts suggest linear, while percentages or multiplicative processes suggest exponential. Over long time horizons, exponential growth always overtakes linear growth, no matter the starting values or rates. Mastering this comparison equips you for regression analysis, logarithms, and real-world data modeling in future courses.