Linear approximation in AP Calculus AB/BC

Linear approximation (or linearization) is the technique of using the equation of a tangent line at a point to estimate a function's value at a nearby x-value, based on the idea that a differentiable function looks locally linear when you zoom in close enough (AP Calc Topic 4.6, LO 4.6.A).

What is linear approximation?

Linear approximation is the payoff of one big idea in calculus, which is that if you zoom in far enough on a differentiable function, it looks like a straight line. That straight line is the tangent line. So if you know a function's value and its derivative at one point, you can write the tangent line there, L(x) = f(a) + f'(a)(x − a), and plug in a nearby x-value to estimate f(x) without knowing the actual function value.

The CED puts it this way in Topic 4.6. The tangent line is the graph of a locally linear approximation of the function near the point of tangency. The word "locally" matters. The estimate is good close to the point of tangency and gets worse as you move away. There's also a second layer the exam loves to test. The function's behavior near the point (specifically its concavity) determines whether your tangent line estimate is an underestimate or an overestimate of the true value. If the curve is concave up, the tangent line sits below the curve, so the approximation is an underestimate. Concave down means the tangent sits above, so it's an overestimate.

Why linear approximation matters in AP® Calculus

Linear approximation is the entire job of Topic 4.6 in Unit 4 (Contextual Applications of Differentiation), and it directly supports learning objective 4.6.A, "Approximate a value on a curve using the equation of a tangent line." It's one of the most reliable FRQ moves on the exam because it bundles several skills together. You have to find a derivative (often implicitly), build a tangent line, evaluate it, and then justify whether the estimate is too big or too small using the second derivative. That last step is where the points hide. Computing the estimate is the easy part; the exam wants the concavity-based justification. For the full walkthrough, head to the Topic 4.6 study guide.

How linear approximation connects across the course

Tangent Line Approximation (Unit 4)

This is the same procedure under a slightly different name. "Tangent line approximation," "linear approximation," and "linearization" all mean writing L(x) = f(a) + f'(a)(x − a) and plugging in a nearby x. If a problem says any of the three, do the same thing.

Tangent Line (Unit 2)

Unit 2 teaches you that the derivative IS the slope of the tangent line. Linear approximation is what that fact is for. You learned to build the line in Unit 2; in Unit 4 you finally use it to estimate function values.

Implicit Differentiation (Unit 3)

FRQs love pairing these. The 2024 FRQ Q5 gave the curve x² + 3y + 2y² = 48 with dy/dx = −2x/(3 + 4y), found implicitly. You can't solve for y, so the tangent line is the only realistic way to estimate nearby points on the curve.

Concavity and the Second Derivative (Unit 5)

Concavity is your over/underestimate detector. Concave up means the tangent line lies below the curve (underestimate); concave down means it lies above (overestimate). This is exactly the "function's behavior near the point of tangency" line in the CED, and it's the standard justification the exam asks for.

Is linear approximation on the AP® Calculus exam?

In multiple choice, expect stems like "use the tangent line at x = a to approximate f(a + 0.1)" or conceptual questions about why linearization works (the derivative gives the slope of the locally linear model). In FRQs, linear approximation usually shows up as one part of a multi-part problem, often after implicit differentiation, like the 2024 FRQ Q5 setup where dy/dx comes from the curve x² + 3y + 2y² = 48. The full task is usually three steps. Write the tangent line at the given point, evaluate it at the nearby x-value, then state whether the result is an over- or underestimate and justify it with concavity (the sign of the second derivative near the point). Accuracy questions also appear, and the answer is that the approximation improves the closer your x-value is to the point of tangency.

Linear approximation vs Linearization

These aren't actually different things, and that's the point. "Linearization" is the name for the linear function L(x) = f(a) + f'(a)(x − a), while "linear approximation" is the act of using it to estimate f(x). The College Board uses both terms (Topic 4.6 is literally titled "Local Linearity and Linearization"), so don't panic if a question swaps vocabulary mid-problem. Same formula, same tangent line, same answer.

Key things to remember about linear approximation

• Linear approximation estimates f(x) near x = a using the tangent line L(x) = f(a) + f'(a)(x − a), because differentiable functions are locally linear.

• The estimate is only trustworthy near the point of tangency, and accuracy improves as your x-value gets closer to that point.

• Concavity determines the direction of the error. Concave up means the tangent line gives an underestimate, and concave down means it gives an overestimate.

• On FRQs, linear approximation often follows implicit differentiation, like the 2024 FRQ where dy/dx = −2x/(3 + 4y) came from the curve x² + 3y + 2y² = 48.

• The justification points come from the second derivative, not the arithmetic, so always explain over/underestimate using concavity near the point of tangency.

Frequently asked questions about linear approximation

What is linear approximation in AP Calculus?

It's using the tangent line at a known point, L(x) = f(a) + f'(a)(x − a), to estimate a function's value at a nearby x-value. It works because differentiable functions look like straight lines when you zoom in, which is the "local linearity" idea in Topic 4.6.

Is linear approximation the same as linearization?

Yes. Linearization is the tangent line function itself, and linear approximation is the act of plugging a nearby x-value into it. The AP exam uses the terms interchangeably, so treat any of "linearization," "linear approximation," or "tangent line approximation" as the same task.

How do you know if a linear approximation is an overestimate or underestimate?

Check concavity near the point of tangency. If f''(x) > 0 (concave up), the tangent line sits below the curve and the approximation is an underestimate; if f''(x) < 0 (concave down), it's an overestimate. This justification is what FRQ rubrics typically award points for.

Does linear approximation give the exact answer?

No, it's an estimate with error that grows as you move away from the point of tangency. The exam exploits this. You're often asked whether the estimate is too high or too low, which you answer with the second derivative, not by computing the true value.

How is linear approximation tested on the AP Calc exam?

Usually as part of a multi-part FRQ. You find dy/dx (sometimes implicitly, like the 2024 FRQ with x² + 3y + 2y² = 48), write the tangent line at a given point, evaluate it at a nearby x-value, then justify over/underestimate using concavity. MCQs test the same skill in one step.