Comparing Exponential vs. Logarithmic — I can compare exponential and logarithmic growth and explain when each model is appropriate. - Math 3

SECTION 1

Historical Context & Motivation

Throughout human history, people have needed ways to describe how quantities grow and change. Some quantities—like money earning interest or bacteria reproducing—double over and over again, growing faster the larger they get. Others—like the intensity of sound perceived by our ears or the magnitude of an earthquake—compress enormous ranges of values into a manageable scale. These two patterns of change, exponential growth and logarithmic growth, are deeply connected—yet they behave in strikingly opposite ways. Understanding their relationship began centuries ago and transformed science, finance, and engineering.

1614 • Napier Publishes Logarithm Tables

Scottish mathematician John Napier introduced logarithms as a computational tool, turning tedious multiplication into simple addition and revolutionizing astronomy and navigation.

1683 • Jacob Bernoulli and Compound Interest

While studying compound interest, Bernoulli discovered the constant e ≈ 2.718, the natural base of exponential growth that appears throughout mathematics and science.

1935 • Richter Develops the Earthquake Scale

Charles Richter created a logarithmic scale to measure earthquake magnitudes, demonstrating how logarithms compress vast ranges of data into understandable numbers.

1965 • Moore's Law Predicts Exponential Growth

Gordon Moore predicted that the number of transistors on a microchip would double roughly every two years—a prediction that held for decades and is a classic example of exponential growth in technology.

The central question this lesson tackles is straightforward but powerful: when a real-world quantity is changing, how do you decide whether an exponential model or a logarithmic model best describes the situation? To answer that, you first need to understand what each function does, how they are related, and what their graphs look like.

SECTION 2

Core Principles & Definitions

Before comparing the two models, let's nail down exactly what each one means. An exponential function has the variable in the exponent, while a logarithmic function essentially asks, "What exponent do I need?" These two ideas are inverse operations—just like addition undoes subtraction, logarithms undo exponentiation.

Exponential Functions

A function of the form y = a · b^x where b > 0 and b ≠ 1. The output changes by a constant ratio for each unit increase in x.

Logarithmic Functions

A function of the form y = log_b(x), the inverse of the exponential. It answers: "To what power must b be raised to get x?" The output changes by a constant amount when x is multiplied by a constant factor.

Inverse Relationship

If y = b^x, then x = log_b(y). Their graphs are reflections of each other over the line y = x.

Growth Behavior

Exponential functions grow without bound, accelerating as x increases. Logarithmic functions also grow without bound, but they do so more and more slowly—they decelerate.

✦

KEY TAKEAWAY

Think of exponential and logarithmic functions like a volume knob and a loudness meter. When you turn the knob (input) by equal steps, the sound energy (exponential output) doubles each time—it explodes. But the loudness meter (logarithmic output) barely budges because your ear compresses those huge energy jumps into small perceived differences. One model captures the explosive, multiplying behavior; the other captures the compressing, leveling-off behavior.

SECTION 3

Visual Explanation — Graphs Side by Side

The best way to see the difference between exponential and logarithmic functions is to plot them on the same coordinate plane. The diagram below shows y = 2^x (the exponential curve in cyan) and y = log₂(x) (the logarithmic curve in violet), along with the line y = x (dashed). Notice how the two curves are mirror images across that dashed line—this is the visual proof that they are inverse functions.

The cyan curve (y = 2^x) shoots upward as x increases, while the violet curve (y = log₂(x)) climbs slowly. The dashed line y = x shows their symmetry as inverse functions.

Several features jump out from this graph. The exponential curve crosses the y-axis at (0, 1) and has the x-axis as a horizontal asymptote — the curve approaches y = 0 but never reaches it as x goes to the left. Meanwhile, the logarithmic curve crosses the x-axis at (1, 0) and has the y-axis as a vertical asymptote — it plunges toward negative infinity as x approaches 0 from the right. If you swap the x- and y-coordinates of any point on one curve, you land on the other curve, confirming they are inverses.

SECTION 4

Mathematical Framework

Let's formalize the two function families and their key properties. Understanding the algebra behind each model helps you manipulate them confidently in problems and convert between forms.

EXPONENTIAL FUNCTION

y = a · bˣ

a = initial value (y-intercept when x = 0); b = base (growth factor if b > 1, decay factor if 0 < b < 1); x = independent variable (often time).

LOGARITHMIC FUNCTION

y = log_b(x)

b = base (same base as the related exponential); x = input (must be positive); y = the exponent to which b must be raised to produce x.

INVERSE RELATIONSHIP

b^(log_b(x)) = x and log_b(b^x) = x

Applying one function and then the other returns you to the original value. This is the defining property of inverse functions.

RATE OF CHANGE COMPARISON

Exponential: Δy/Δx grows proportionally to y | Logarithmic: Δy/Δx shrinks as x increases

For exponentials the rate of change accelerates — each step produces a bigger jump. For logarithms the rate of change decelerates — each step produces a smaller gain.

💡

Converting Between Forms

If you know the exponential form b^y = x, you can always rewrite it in logarithmic form as y = log_b(x). For example, 2³ = 8 becomes log₂(8) = 3. Practice this conversion — it's the single most useful skill for moving between the two models.

SECTION 5

Growth Patterns — When to Use Each Model

Choosing between an exponential and a logarithmic model depends on how the quantity you're measuring behaves. The table below summarizes the key differences, and the diagram that follows shows real-world data patterns alongside their matching models.

Feature	Exponential Model	Logarithmic Model
General Form	y = a · bˣ	y = c · log_b(x) + d
Domain	All real numbers	x > 0 only
Range	y > 0 (for a > 0)	All real numbers
Asymptote	Horizontal (y = 0)	Vertical (x = 0)
Rate of Change	Accelerating — grows faster and faster	Decelerating — grows slower and slower
Key Pattern in Data	Equal Δx → output multiplied by constant ratio	Input multiplied by constant ratio → equal Δy
Real-World Examples	Population growth, compound interest, radioactive decay	Richter scale, decibel scale, pH scale, learning curves

Exponential vs. Logarithmic at a Glance

A decision flowchart to help you choose: if your data shows a constant multiplicative ratio in outputs for equal input steps, use an exponential model. If it shows diminishing returns as input grows, use a logarithmic model.

Here's a quick test you can apply to any data set. Make a table with equally spaced x-values. Calculate the ratio of consecutive y-values. If those ratios are roughly constant, the data is exponential. If instead you notice that multiplying the x-values by a constant factor adds a constant amount to y, the data is logarithmic. In practice, real data may not be perfectly one or the other, but these checks point you toward the best-fit model.

SECTION 6

Worked Example — Identifying and Using the Right Model

A biologist tracks the number of bacteria in a petri dish every hour and also measures the perceived loudness (in decibels) of a speaker as its power is increased. Let's determine which model fits each scenario and make a prediction.

Scenario A: Bacteria Population

1
Step 1 — Examine the Data

The biologist records: at t = 0 h, population = 500; t = 1 h, population = 1 000; t = 2 h, population = 2 000; t = 3 h, population = 4 000. Each hour the population doubles.
2
Step 2 — Check Ratios

Compute successive ratios: 1 000 ÷ 500 = 2, 2 000 ÷ 1 000 = 2, 4 000 ÷ 2 000 = 2. The ratio is constant at 2, confirming an exponential model.

Constant ratio = 2 → Exponential
3
Step 3 — Write the Equation

Using y = a · bˣ with a = 500 (initial value) and b = 2 (doubling), the model is y = 500 · 2^t.

y = 500 · 2ᵗ
4
Step 4 — Predict

At t = 5 hours: y = 500 · 2⁵ = 500 · 32 = 16 000 bacteria.

Predicted population at 5 hours: 16 000

Scenario B: Sound Loudness vs. Power

1
Step 1 — Examine the Data

The speaker data shows: power = 1 W → 60 dB; power = 10 W → 70 dB; power = 100 W → 80 dB; power = 1 000 W → 90 dB. Each time power is multiplied by 10, the decibel reading increases by 10.
2
Step 2 — Check for Logarithmic Pattern

Equal multiplicative changes in the input (×10) produce equal additive changes in the output (+10 dB). This is the hallmark of a logarithmic model.

Constant additive Δy when x is multiplied → Logarithmic
3
Step 3 — Write the Equation

Using the common-log form: y = 10 · log₁₀(P) + 60, where P is power in watts. Check: log₁₀(1) = 0 → y = 60 ✓; log₁₀(10) = 1 → y = 70 ✓.

y = 10 · log₁₀(P) + 60
4
Step 4 — Predict

At P = 10 000 W: y = 10 · log₁₀(10 000) + 60 = 10 · 4 + 60 = 100 dB.

Predicted loudness at 10 000 W: 100 dB

SECTION 7

Strengths & Limitations of Each Model

No model is perfect for every situation. Exponential and logarithmic models each have strengths and limitations, and recognizing these helps you avoid misusing them.

Aspect	Exponential Model	Logarithmic Model
Strength	Accurately captures processes that compound — populations, investments, viral spread.	Compresses wide data ranges into manageable scales — earthquakes, sound, acidity.
Limitation	Predictions can become unrealistically large; real systems often hit resource limits (logistic growth).	Cannot model situations with rapid acceleration; not defined for x ≤ 0.
Common Mistake	Extrapolating too far into the future without considering real-world constraints.	Forgetting that the input must be positive; trying to evaluate log(0) or log(negative).
Best For	Short- to medium-term predictions of compounding phenomena.	Measuring or ranking quantities that span many orders of magnitude.

Strengths and Limitations

✦

KEY TAKEAWAY

Think of it this way: exponential models are great storytellers for the "how fast is it growing?" question, while logarithmic models are great answerers of the "how many doublings (or tenfold increases) has it undergone?" question. In a sense, exponentials look forward (predict the future), and logarithms look backward (measure how far you've come).

SECTION 8

Connections to Advanced Topics

Once you are comfortable switching between exponential and logarithmic perspectives, a range of more advanced mathematical ideas opens up. The table below previews how these models evolve in later courses.

This Course	Future Topic
y = a · bˣ with b > 1 (exponential growth)	In calculus, the derivative of eˣ equals itself — the rate of growth of an exponential is proportional to its current value.
Exponential models for population	Logistic growth models add a carrying capacity, curving the exponential into an S-shape: y = K / (1 + Ae⁻ᵏᵗ).
y = log(x) and its slow growth	In computer science, algorithms with O(log n) time complexity are prized for efficiency — binary search is a classic example.
Converting between exponential and log forms	Logarithmic differentiation in calculus simplifies products, quotients, and variable exponents.

From Math 3 to Future Courses

Even without calculus, you can already appreciate the power of these models. Every time you use a logarithmic scale on a graph (like a semi-log plot), you are converting exponential data into a straight line—making trends easier to spot. This technique appears in chemistry (pH), astronomy (star brightness), and data science (feature scaling).

SECTION 9

Practice Problems

Question 1: CONCEPTUAL

In your own words, explain how you can tell from a graph whether a function is exponential or logarithmic. Name at least two visual features that distinguish them.

Show answer

An exponential graph starts slowly (near a horizontal asymptote, typically the x-axis) and then curves steeply upward, passing through (0, a) where a is the initial value. A logarithmic graph starts by rising steeply near its vertical asymptote (the y-axis) and then levels off, passing through (1, 0). The exponential curve is concave up (curving upward), while the logarithmic curve is concave down (curving and flattening). Additionally, they are mirror images across the line y = x, confirming their inverse relationship.

Question 2: BASIC CALCULATION

A savings account starts with $800 and grows by 5% per year. Write an exponential equation for the account balance B after t years, then calculate the balance after 6 years.

Show answer

The exponential model is B = 800 · (1.05)ᵗ. Substituting t = 6: B = 800 · (1.05)⁶. Computing (1.05)⁶ ≈ 1.3401, so B ≈ 800 × 1.3401 ≈ $1 072.08. The account earns compound interest, and each year the 5% is applied to the previous year's balance, which is why the model is exponential rather than linear.

Question 3: INTERMEDIATE

A researcher measures the following data: when x = 1, y = 0; when x = 3, y = 1.585; when x = 9, y = 3.17; when x = 27, y = 4.755. Determine whether this data best fits an exponential or logarithmic model, and find the equation.

Show answer

Check ratios of consecutive y-values: 1.585/0 is undefined, so instead check the input pattern. The x-values multiply by 3 each step (1 → 3 → 9 → 27), while the y-values increase by approximately 1.585 each step. Equal multiplicative changes in x producing equal additive changes in y is the signature of a logarithmic function. Since log₃(1) = 0, log₃(3) ≈ 1, log₃(9) = 2, log₃(27) = 3, and multiplying by a constant c ≈ 1.585 we get y ≈ 1.585 · log₃(x). Using the change-of-base formula, 1.585 ≈ ln(3)/ln(2) which simplifies to y = log₂(x). Checking: log₂(1) = 0 ✓, log₂(3) ≈ 1.585 ✓, log₂(9) ≈ 3.17 ✓, log₂(27) ≈ 4.755 ✓. The model is y = log₂(x).

Question 4: APPLIED

The pH of a solution is defined as pH = −log₁₀[H⁺], where [H⁺] is the hydrogen ion concentration in moles per liter. If a cleaning solution has [H⁺] = 10⁻¹² mol/L and lemon juice has [H⁺] = 10⁻² mol/L, find the pH of each. Then explain why the pH scale is logarithmic rather than linear.

Show answer

For the cleaning solution: pH = −log₁₀(10⁻¹²) = −(−12) = 12. For lemon juice: pH = −log₁₀(10⁻²) = −(−2) = 2. The hydrogen ion concentration ranges from about 10⁻¹⁴ to 10⁰ mol/L — a span of 14 orders of magnitude. A linear scale would make it nearly impossible to compare everyday substances. The logarithmic pH scale compresses this enormous range into a manageable 0-to-14 scale, where each whole-number step corresponds to a tenfold change in [H⁺]. This makes differences intuitive: lemon juice (pH 2) is 10¹⁰ = 10 billion times more acidic than the cleaning solution (pH 12).

Question 5: CRITICAL THINKING

A viral video's view count triples every day for the first week, starting from 200 views. After the first week, growth slows dramatically, and the total views follow a logarithmic pattern. (a) Write an exponential model for the first week and predict views on Day 7. (b) Explain why switching to a logarithmic model after Day 7 makes sense in the real world. (c) Could you use a single function to model the entire lifespan of the video's viewership? Suggest one and explain.

Show answer

(a) During the first week: V = 200 · 3ᵗ, where t is days since posting. On Day 7: V = 200 · 3⁷ = 200 · 2 187 = 437 400 views. (b) Exponential growth cannot continue indefinitely because the number of potential viewers is finite, interest wanes, and algorithms stop promoting the video. A logarithmic model captures this 'diminishing returns' phase where each additional day adds fewer and fewer new views — the curve flattens. (c) A logistic function, y = K / (1 + Ae⁻ᵏᵗ), could model the entire lifespan. It behaves like an exponential early on (steep growth), then transitions through an inflection point, and finally levels off near a maximum value K (the carrying capacity, i.e., total potential viewers). This S-shaped curve naturally combines the explosive early phase with the plateauing later phase in one equation.

SUMMARY

Lesson Summary

Summary

Exponential functions (y = a · bˣ) model quantities whose rate of change is proportional to their current value — they accelerate as they grow. Logarithmic functions (y = log_b(x)) are the inverses of exponentials — they decelerate as x grows, compressing wide ranges of input into a manageable output scale. Their graphs are reflections of each other across the line y = x.

To choose between the two models, examine your data. If equal input steps produce a constant output ratio, use an exponential. If multiplying the input by a constant produces a constant additive change in output, use a logarithmic model. Real-world applications range from compound interest and population growth (exponential) to the Richter scale, decibels, and pH (logarithmic). Mastering the relationship between these two function families gives you the tools to model a huge variety of phenomena in science, finance, and technology.