Skip to content

AP Calculus BC · technique

How to answer the free-response section

On this page · 5 sections

Section II of the AP Calculus BC exam is six free-response questions in 90 minutes, and it counts for 50% of your score. Each question is worth 9 points, so the section has 54 raw points. Part A gives you 30 minutes for questions 1 and 2, and a graphing calculator is required. Part B gives you 60 minutes for questions 3 through 6 with no calculator. During Part B you may go back to the Part A questions, but your calculator stays put away.

Each question is broken into lettered parts, usually (a) through (d), and each part carries its own points. Trained readers score your written work against a point-by-point rubric. A point is tied to a specific thing on the page: a correct setup, a correct answer, a justification, an interpretation with units. That means a bare number with no work earns little or nothing, even when it is right. It also means you can earn the setup point on a part where your final answer is wrong.

Readers also score each part consistently with your earlier answers. If you got part (a) wrong and used that value correctly in part (b), you can still earn the points in (b). The BC lineup is predictable. Expect a rate-in/rate-out or table question, a particle-motion or parametric or polar question, a differential equation with a slope field or Euler's method, and question 6 on Taylor series almost every year.

The steps

  1. 1

    Scan every part before you write

    Give yourself about 15 minutes per question, since 90 minutes across six questions is 15 each. Read all the parts first, because later parts are often independent of earlier ones. Part (d) of a rate question may only need an integral, while part (b) needs a hard justification. Do the parts you can finish quickly first, then return. If a part stalls you for three minutes, leave space and move on, because a skipped part costs only its points while a lost five minutes can cost a whole question.

  2. 2

    Write the setup in math notation first

    Before any number appears, write the expression that produces it. For total water entering a tank, write the definite integral with its limits and integrand, such as the integral from 0 to 8 of E(t) dt. Readers award a point for that setup on its own. Never write calculator syntax like fnInt(E,t,0,8), because readers do not accept it as mathematical work. If your setup is right and you mistype into the calculator, the setup point is still yours. In Part B you do not need to simplify a numerical answer, so leaving e - 1 or 3 + 2(5) as is earns full credit. Every extra step of arithmetic is just another chance to slip.

  3. 3

    Use the calculator only for its four allowed jobs

    In Part A, a bare calculator result counts only for four tasks: graphing a function, solving an equation, finding a numerical derivative at a point, and finding a definite integral. Anything else, like finding a maximum, needs calculus written on the page. Report decimals correct to three places after the decimal point. Store intermediate values in calculator memory instead of rounding them, because rounding 5.0185 to 5.02 and reusing it can push your final answer outside the accepted range.

  4. 4

    Justify with a named function and a stated condition

    A justification point needs a reason that names the right function and the right fact about it. Write 'h has a relative maximum at t = 3 because h' changes from positive to negative there,' not 'the graph goes up then down.' For the Mean Value Theorem, state both hypotheses before you use it. For example: 'Because f is differentiable on [2, 6], f is continuous on [2, 6] and differentiable on (2, 6). By the MVT, there is a c in (2, 6) with f'(c) = (f(6) - f(2))/4.' For the Intermediate Value Theorem, the hypothesis you need is continuity on the closed interval, so write 'f is continuous on [2, 6]' before you draw a conclusion. For series, name the test and check its conditions, including the endpoints of an interval of convergence.

  5. 5

    Interpret in context with time and units

    When a part asks what a value means, your sentence needs three things: the quantity, the time, and the units. For E'(5) = -4.807, write 'at t = 5 hours, the rate water enters the tank is decreasing by 4.807 gallons per hour per hour.' Units of a derivative are the output units divided by the input units. Units of an integral are the output units times the input units. Leaving out the time or using the wrong units is one of the most common ways to lose this point.

  6. 6

    Keep going after a mistake, and do not erase

    Because readers grade later parts consistently with your earlier answers, a wrong value in part (a) does not doom part (c). Use the value you found and finish the work. If you change your mind, cross out the old work with a single line rather than leaving two answers. Readers score everything that is not crossed out, so two conflicting answers can cost you the answer point.

  7. 7

    Run the BC series checklist on question 6

    Question 6 is almost always Taylor series, and its parts repeat. Expect to write the first few nonzero terms of a Taylor polynomial, to find a radius or interval of convergence with the ratio test, and to bound an error. Write the general term with the factorial, as f^(n)(a)/n! times (x - a)^n. Use the alternating series error bound when the series alternates and its terms decrease in absolute value toward 0. Write that condition in your answer, because it is your justification. Use the Lagrange bound when the problem gives you a bound on a higher derivative. Test both endpoints separately, because the ratio test says nothing there.

Worked through

Here is an invented calculator question and a full write-up of the kind readers reward.

The problem: For 0 ≤ t ≤ 8 hours, water enters a tank at a rate of E(t) = 30 + 12 sin(t/2) gallons per hour. Water leaves at a rate of W(t) = 5t + 12 gallons per hour. At t = 0 the tank holds 200 gallons. (a) Find E'(5) and explain what it means. (b) How many gallons of water enter the tank from t = 0 to t = 8? (c) Is the amount of water in the tank increasing or decreasing at t = 6? Give a reason. (d) At what time is the amount of water in the tank greatest? Justify your answer.

Part (a), 2 points. Write E'(5) = -4.807, using the calculator's numerical derivative. Then interpret it: at t = 5 hours, the rate at which water enters the tank is decreasing by 4.807 gallons per hour per hour. One point is for the value, and one is for a sentence with the time, the meaning and the units.

Part (b), 2 points. Write the integral from 0 to 8 of E(t) dt = 279.687 gallons. The setup with correct limits earns one point, and the value earns the other. Notice that the answer is written to three decimal places.

Part (c), 2 points. Let A(t) be the amount of water, so A'(t) = E(t) - W(t). Compute E(6) = 31.693 and W(6) = 42, which gives A'(6) = -10.307. Since A'(6) < 0, the amount of water is decreasing at t = 6. The reason must compare the two rates or state the sign of A'. Saying only 'it is decreasing' earns nothing.

Part (d), 3 points. Set A'(t) = E(t) - W(t) = 0 and solve with the calculator, which gives t = 5.018. That is the only critical point. Now use the candidates test, because the question asks for the greatest amount on a closed interval. Write A(t) = 200 + the integral from 0 to t of (E(x) - W(x)) dx. Then A(0) = 200, A(5.018) = 270.729 and A(8) = 223.687. The amount of water is greatest at t = 5.018 hours, since A(5.018) is the largest of the three values. The points go to setting A' = 0, to evaluating A at the critical point and both endpoints, and to the conclusion with its reason.

Two habits made this response full credit. First, every number sits next to the expression that produced it. Second, the value 5.018493 was stored in memory and reused, so the 270.729 is not thrown off by early rounding.

Common mistakes

  • Leaving out the constant of integration when solving a separable differential equation. If you add +C after you have already solved for y, or skip it entirely, you lose most of the points for that part, because the particular solution is then wrong.

  • Mixing up Riemann sum endpoints from a table. A left sum uses the values at the start of each subinterval, a right sum uses the end values, and five table entries give four subintervals, not five. With unequal widths, each height must be multiplied by its own width.

  • Treating the Fundamental Theorem the wrong way around. If g(x) is the integral from 0 to x of f(t) dt, then g'(x) = f(x): you plug in x and do not differentiate the integrand. For a definite integral you must subtract the lower-limit value, so the integral of e^x from 0 to 1 is e - 1, not e.

  • Keeping the old x-limits after a u-substitution. For the integral from 0 to 2 of 2x(x^2 + 1)^3 dx, let u = x^2 + 1, so du = 2x dx. When x = 0, u = 1, and when x = 2, u = 5, so the new integral is from 1 to 5 of u^3 du, which equals 156. Writing it as the integral from 0 to 2 of u^3 du gives 4, which is wrong.

  • Writing an equal-sign chain that is not true. If you mean f'(3) = 2(3) = 6 but write f'(x) = 2x = 6, the line claims that 2x always equals 6. Readers treat each equal sign as a statement, so a false chain like this can cost you the point for that line.

  • Giving a justification that does not name the function. 'It changes from increasing to decreasing' about a graph earns no point; 'f' changes sign from positive to negative at x = 2' does.

  • Rounding too early in Part A. Rounding an intermediate value to two decimals and reusing it can push the final answer outside the three-decimal accuracy readers accept, which costs the answer point.

Checklist

  • I can name the six questions' timing: 2 calculator questions in 30 minutes, then 4 no-calculator questions in 60 minutes.
  • I write the integral or derivative expression before writing any numerical answer, and in Part B I leave numerical answers unsimplified.
  • I know the four things a calculator answer may stand on alone, and I write calculus for everything else.
  • I give decimal answers to three places and store intermediate values instead of rounding them.
  • My interpretation sentences include the time, the meaning and the correct units.
  • My justifications name the function and the condition, like a sign change in f', and state theorem hypotheses.
  • Every equal sign I write is a true statement.
  • On a series question I write the general term with its factorial, test both endpoints, and state the conditions for the error bound I choose.
  • I finish later parts using my earlier answers, even if I suspect those answers are wrong.

More AP Calculus BC technique