Choosing Linear vs. Exponential — I can compare linear vs exponential models for a context and justify the choice. - Grade 9 Math

SECTION 1

Historical Context & Motivation

People have been modeling change for centuries. Ancient farmers tracked the steady growth of their grain stores season by season, while early bankers noticed that compound interest made savings grow faster and faster over time. These two patterns — steady addition versus repeated multiplication — became the foundation for two of the most important mathematical models in history. Understanding when to use each one has helped scientists predict population booms, engineers design safe structures, and economists forecast market trends.

~300 BCE • Euclid & Proportional Reasoning

Euclid's Elements formalized proportional relationships, laying groundwork for what we now call linear functions.

1614 • Napier Publishes Logarithms

John Napier introduced logarithms to simplify calculations involving exponential growth, giving scientists a tool to work with repeated multiplication.

1798 • Malthus & Population Growth

Thomas Malthus argued that human populations grow exponentially while food supplies grow linearly, sparking a debate about which model fits which context.

1900s–Today • Data-Driven Modeling

Modern data analysis uses regression tools to decide whether a linear or exponential model best fits real-world data, from viral social-media trends to climate projections.

The central question this lesson addresses is deceptively simple: Given a real-world situation, how do you decide whether a linear or an exponential model is the better fit, and how do you justify that choice? By the end, you'll be able to look at a table of values, a graph, or a verbal description and confidently pick the right model.

SECTION 2

Core Principles & Definitions

Before you can choose the right model, you need to understand what makes linear and exponential functions fundamentally different. Both describe how a quantity changes, but they do so in completely different ways. A linear model adds (or subtracts) the same amount in each time step, while an exponential model multiplies (or divides) by the same factor in each time step.

Constant Rate of Change

In a linear model, the first differences (the change between consecutive y-values) are always the same. Think of earning $10/hour — every hour adds exactly $10.

Constant Ratio of Change

In an exponential model, the ratio between consecutive y-values stays the same. Think of a rumor that doubles every day — each day multiplies the count by 2.

Growth vs. Decay

Both models can increase or decrease. Linear growth has a positive slope; linear decay has a negative slope. Exponential growth has a base > 1; exponential decay has a base between 0 and 1.

Long-Run Behavior

Over time, an exponential function will always overtake a linear function (if both are growing). The exponential curve starts slowly but eventually rockets upward.

✦

KEY TAKEAWAY

Think of it like walking up stairs versus riding an escalator that speeds up. With stairs (linear), each step lifts you the same height. On the accelerating escalator (exponential), each second carries you higher than the last. The question "Is it adding the same amount, or multiplying by the same factor?" is your single most powerful tool for choosing the right model.

SECTION 3

Visual Explanation — Side-by-Side Graphs

The quickest way to tell a linear model from an exponential model is to look at their graphs. A linear function always produces a straight line, while an exponential function produces a curve that bends upward (or downward) increasingly steeply. The diagram below compares y = 2x + 1 (linear) with y = 2^x (exponential).

Notice how the blue linear line rises at a constant slope, while the pink exponential curve starts slowly and then shoots upward, overtaking the line between x = 2 and x = 3.

When you look at data graphically, a straight-line pattern strongly suggests a linear model, while a pattern that curves away from the x-axis — slowly at first, then steeply — points toward an exponential model. Keep in mind that exponential decay curves toward the x-axis from above, getting closer and closer without touching it.

SECTION 4

Mathematical Framework

Each model has a standard equation form. Knowing these forms — and what each variable represents — lets you write the function once you've identified the model type.

LINEAR MODEL

y = mx + b

m = slope (constant rate of change, i.e., the amount added per unit of x). b = y-intercept (starting value when x = 0).

EXPONENTIAL MODEL

y = a · bˣ

a = initial value (y-value when x = 0). b = common ratio (the constant multiplier per unit of x). If b > 1, the function models growth; if 0 < b < 1, it models decay.

The Difference-vs.-Ratio Test

When you have a table of evenly spaced x-values, you can run two quick tests. First, compute the first differences: subtract each y-value from the next one. If these differences are roughly constant, the data is linear. Second, compute the consecutive ratios: divide each y-value by the previous one. If these ratios are roughly constant, the data is exponential.

FIRST DIFFERENCES (LINEAR TEST)

Δy = y₂ − y₁ = y₃ − y₂ = … = constant

If all first differences are approximately equal, the model is linear.

CONSECUTIVE RATIOS (EXPONENTIAL TEST)

r = y₂ / y₁ = y₃ / y₂ = … = constant

If all consecutive ratios are approximately equal, the model is exponential.

SECTION 5

Decision Flowchart & Comparison Table

The flowchart below walks you through the decision process step by step. Start at the top and follow the branches based on what you observe in the data, the graph, or the verbal description of the situation.

Follow this flowchart when you encounter a new modeling problem. If you have numerical data, check first differences and ratios. If you only have a verbal description, look for key phrases like "per" (linear) or "doubles/triples" and "percent per" (exponential).

Feature	Linear Model	Exponential Model
Equation form	y = mx + b	y = a · bˣ
How it changes	Adds/subtracts a constant amount	Multiplies/divides by a constant factor
Graph shape	Straight line	Curve (J-shape or decay curve)
Table test	First differences are constant	Consecutive ratios are constant
Key phrases	"$5 per hour," "3 cm each day"	"doubles every week," "loses 10% per year"
Real-world examples	Constant-speed driving, hourly wages, filling a pool at a steady rate	Bacterial growth, radioactive decay, compound interest

Side-by-side comparison of the two model types

SECTION 6

Worked Example — Choosing & Justifying a Model

A biologist records the number of bacteria in a petri dish every hour. The data are shown below. Determine whether a linear or exponential model fits the data, write the equation, and justify your choice.

Hour (x)	0	1	2	3	4
Bacteria (y)	50	150	450	1 350	4 050

Bacteria count every hour

Choosing the Right Model for Bacteria Growth

1
Step 1 — Check First Differences

Compute the differences between consecutive y-values: 150 − 50 = 100, 450 − 150 = 300, 1 350 − 450 = 900, 4 050 − 1 350 = 2 700. These differences (100, 300, 900, 2 700) are not constant, so a linear model is not a good fit.

First differences are not constant → NOT linear.
2
Step 2 — Check Consecutive Ratios

Divide each y-value by the previous one: 150 / 50 = 3, 450 / 150 = 3, 1 350 / 450 = 3, 4 050 / 1 350 = 3. Every ratio equals 3, so the data has a constant multiplicative factor.

Consecutive ratios are constant (r = 3) → EXPONENTIAL model.
3
Step 3 — Identify a and b

The initial value at x = 0 is a = 50. The common ratio is b = 3.

a = 50, b = 3
4
Step 4 — Write the Equation

Substituting into the exponential form y = a · bˣ gives us the model equation.

y = 50 · 3ˣ
5
Step 5 — Justify the Choice

An exponential model is appropriate because the consecutive ratios are constant at 3, meaning the bacteria population triples every hour. The first differences are increasing, which rules out a linear model. The context also supports this choice: bacterial reproduction involves each cell splitting into multiple cells, a multiplicative process.

Justified: constant ratio, multiplicative growth context.

SECTION 7

Strengths & Limitations of Each Model

Neither model is universally "better" — each shines in specific situations and breaks down in others. Understanding these trade-offs helps you make smarter modeling decisions and recognize when a model's predictions may no longer be trustworthy.

	Strengths	Limitations
Linear	Simple, predictable, easy to calculate. Works well for short-term predictions when the rate of change is roughly constant.	Cannot capture accelerating or decelerating change. May predict impossible negative values (e.g., negative population).
Exponential	Captures rapid growth or decay realistically. Natural fit for processes driven by multiplication (interest, populations, half-lives).	Predicts infinitely large values, which is unrealistic long-term for real populations (resources run out). Harder to compute by hand for large exponents.

Strengths and limitations of linear vs. exponential models

✦

KEY TAKEAWAY

Models are like maps: a road map is great for driving across town, but useless for hiking a mountain trail. A linear model is your road map — perfect for situations with a steady pace. An exponential model is your topographic map — built for terrain that rises (or falls) dramatically. Always ask yourself, "Does this model still make sense for the range of values I care about?"

SECTION 8

Connection to Advanced Modeling

In Algebra 2, Pre-Calculus, and beyond, you'll encounter more sophisticated models that build directly on what you've learned here. Choosing between linear and exponential is the first step in a larger modeling toolkit.

What You Know Now	What Comes Next
Linear model: y = mx + b	Piecewise linear models, linear regression (best-fit lines for scattered data)
Exponential model: y = a · bˣ	Logarithmic models, logistic growth (exponential that levels off), exponential regression
Choosing between two models by inspecting data	Residual analysis, R² values, and statistical tests to quantify which model fits best
Constant difference (linear) vs. constant ratio (exponential)	Constant second differences → quadratic model; constant percentage change → exponential (reinforced in Algebra 2)

How today's skills connect to future coursework

One especially important extension is the logistic model, which starts like exponential growth but then levels off at a carrying capacity — much like a real population that runs out of food. You'll also learn to use logarithms to "undo" exponential equations, which is essential for solving problems like "How long until my investment doubles?" The skills you're building now — identifying the type of change and justifying your model — are exactly the reasoning skills those advanced topics require.

SECTION 9

Practice Problems

Question 1: CONCEPTUAL

A friend says, "My savings account adds $25 every month, so it must be exponential growth because the balance keeps getting bigger." Explain why your friend is incorrect and identify the correct model type.

Show answer

Your friend is confusing "getting bigger" with exponential growth. The key distinction is not whether the quantity increases, but how it increases. Because the account adds a fixed amount ($25) every month, the first differences are constant at $25. A constant additive change means the growth is linear, not exponential. An exponential model would require the balance to grow by a constant percentage (or multiplicative factor) each month, such as earning 2% interest on the current balance.

Question 2: BASIC CALCULATION

A plant is 4 cm tall and grows 1.5 cm per day. Write a linear equation for the plant's height h after d days and find its height on day 10.

Show answer

Since the plant grows by a constant 1.5 cm each day, this is a linear model with slope m = 1.5 and initial value b = 4. The equation is h = 1.5d + 4. To find the height on day 10, substitute d = 10: h = 1.5(10) + 4 = 15 + 4 = 19 cm. The constant rate of "1.5 cm per day" confirms the choice of a linear model.

Question 3: INTERMEDIATE

The table below shows the value of a car over time. Determine whether a linear or exponential model best fits the data, and write the equation. Year (x): 0, 1, 2, 3, 4 Value ($y): 20 000, 16 000, 12 800, 10 240, 8 192

Show answer

First, check first differences: 16 000 − 20 000 = −4 000; 12 800 − 16 000 = −3 200; 10 240 − 12 800 = −2 560; 8 192 − 10 240 = −2 048. These are not constant, so the model is not linear. Next, check consecutive ratios: 16 000 / 20 000 = 0.8; 12 800 / 16 000 = 0.8; 10 240 / 12 800 = 0.8; 8 192 / 10 240 = 0.8. The ratios are all 0.8, confirming an exponential model. With a = 20 000 and b = 0.8, the equation is y = 20 000 · 0.8ˣ. Because b = 0.8 is between 0 and 1, this represents exponential decay — the car loses 20% of its value each year.

Question 4: APPLIED

A city's water department reports that a reservoir loses water at a rate of 5 000 gallons per day due to consumption. Meanwhile, a bacteria colony in a lab doubles every 6 hours starting from 200 cells. For each situation, (a) state whether a linear or exponential model is appropriate and why, and (b) write the equation.

Show answer

Reservoir: The reservoir loses a constant 5 000 gallons per day, which is a fixed amount subtracted each time period. This matches a linear model. If W₀ is the starting water level, then W = W₀ − 5 000d, where d is the number of days. The constant subtraction (negative slope) confirms the linear choice. Bacteria: The colony doubles every 6 hours, which means it is multiplied by 2 each period. This is a constant multiplicative factor, so an exponential model is appropriate. Using 6-hour periods as the unit for x: y = 200 · 2ˣ, where x is the number of 6-hour periods. The key justification is the constant ratio of 2, which comes from the biological process of cell division.

Question 5: CRITICAL THINKING

A small business earned $1 000 in its first month. In month 2 it earned $1 500, in month 3 it earned $2 250, and in month 4 it earned $3 375. A student claims this data is linear because the earnings always increase. Another student claims it is exponential because the ratios are constant. (a) Who is correct? Justify with calculations. (b) If the pattern continues, how much will the business earn in month 7? (c) Discuss one reason why an exponential model might become unrealistic for this scenario over a very long time frame.

Show answer

(a) The second student is correct. First differences: 500, 750, 1 125 — these are not constant, so the data is not linear. Consecutive ratios: 1 500/1 000 = 1.5, 2 250/1 500 = 1.5, 3 375/2 250 = 1.5 — all equal to 1.5, confirming exponential growth. The equation is y = 1 000 · 1.5ˣ⁻¹ (or equivalently y = 1 000 · 1.5⁽ˣ⁻¹⁾ where x is the month number starting at 1). (b) For month 7: y = 1 000 · 1.5⁶ = 1 000 · 11.390625 = $11 390.63 (approximately $11 391). (c) An exponential model predicts unlimited growth, but in reality a small business faces market saturation, competition, and limited resources. Eventually earnings would level off or fluctuate rather than continuing to multiply by 1.5 every month. A logistic model or piecewise model might better capture long-term behavior.

SUMMARY

Lesson Summary

Summary

A linear model (y = mx + b) describes situations with a constant rate of change — the same amount is added or subtracted in each time step, producing a straight-line graph and constant first differences. An exponential model (y = a · bˣ) describes situations with a constant ratio of change — the quantity is multiplied by the same factor in each time step, producing a curved graph and constant consecutive ratios.

To choose and justify the right model, use the difference-vs.-ratio test on tables, look for straight lines vs. curves on graphs, and listen for key phrases like "per" (linear) or "doubles/percent per" (exponential) in word problems. Always support your choice with at least one piece of numerical or contextual evidence.