The first derivative test is the workhorse method on the AP Calculus exam for classifying a critical number as a local maximum, local minimum, or neither. Students recognise the phrase, but on FRQs the test is graded line by line, and a sloppy sign chart or a missing justification can cost an entire row of rubric credit. This article walks through the underlying calculus, the multiple-choice stem patterns that signal a first derivative test problem, and the exact scoring language the College Board uses to award (or withhold) points on free-response questions. The aim is to turn a one-line rule into a defensible argument that survives a 5-level reader's pencil.
What the first derivative test actually states
The first derivative test is a conditional argument, not a formula. Given a function f that is differentiable on an open interval containing a critical number c (where f'(c) = 0 or f'(c) is undefined), the sign of f' on the immediate left and right of c determines the local behaviour of f at c. If f' changes from positive to negative at c, then f has a local maximum at c; if f' changes from negative to positive, then f has a local minimum at c; if f' does not change sign, then c is not a local extremum, even though f'(c) = 0.
This conditional structure is what makes the test valuable on AP Calculus, because it forces a candidate to write a multi-step justification rather than a single computation. On FRQs the rubric typically splits the answer into at least three scored pieces: identifying the critical number, exhibiting a sign chart or equivalent sign analysis of f', and stating the conclusion in the correct vocabulary. A student who reports the correct x-value but writes 'it's a max because the derivative is zero' usually earns one of those three rows and loses the other two, which is the difference between a 3 and a 5 on that question.
It is also worth being precise about vocabulary. AP Calculus readers are trained to credit 'local maximum' and 'absolute maximum' only when the candidate distinguishes them, and the same applies to 'relative extremum' versus 'global extremum'. The first derivative test is silent on absolute behaviour: a function can have a local maximum at c and still not be the highest point on the closed interval of the question. Candidates who conflate the two are punished with a missing row on the FRQ rubric and with distractor options in the multiple choice. Keep the local/global distinction in your writing hand from the first practice set onward.
Finally, the first derivative test requires continuity only at the point under inspection, not across the entire domain. The 2018-released FRQ from the College Board tested a function with a corner at c, and the rubric accepted a sign analysis on each side of c provided the candidate noted the cusp. For most candidates reading this, that means: never write f'(c) = 0 when c is a corner; instead mark c as a critical number because f' is undefined there, and then run the sign test. The vocabulary is small but the credit is real, and it is the kind of one-line correction that moves a 4 to a 5.
Setting up the sign chart on an FRQ
The sign chart is the visible proof of work that the AP reader is hunting. A strong chart has four columns: the critical number c, an interval to the left of c, the value of c itself, and an interval to the right of c. Each row then records the sign of f' on that interval, a representative test point, and the behaviour of f (increasing, decreasing, or neither) as a direct consequence. Two of those columns are scored independently, so a missing sign in the left-hand column is roughly a one-row penalty even if the conclusion is correct.
The fastest way to construct the chart is to factor f' completely. If f'(x) = (x+2)(x-1)(x-3), the roots -2, 1, 3 are the critical numbers and the three factors carry a sign you can read off at a glance. Choose test points such as -3, 0, 2, 4 and tabulate the product. Candidates who skip the factoring step and plug arbitrary values into f' lose minutes and increase the chance of a sign error. For most candidates I tutor, the factoring takes 20 seconds and the test points take another 30; the whole chart fits in a 3-line box on the FRQ page and is enough to defend every conclusion downstream.
On the AP Calculus AB exam, the first derivative test typically appears as part of a multipart FRQ where the function and its derivative are given (or easily computed) and the candidate is asked to identify local extrema, intervals of increase, and the absolute maximum on a stated closed interval. The BC exam extends this with implicit or parametric derivatives, but the sign-chart argument is identical. In both cases the rubric reads: 'credit awarded for the value of the critical number,' 'credit awarded for the sign of f' on the interval (x1, x2),' 'credit awarded for the correct local maximum/minimum conclusion.' When a sign is wrong, the conclusion credit is withheld but the critical-number credit is preserved, which means the candidate still salvages one of the three rows.
The common pitfall here is double-counting. A candidate writes 'f' is positive on (-2, 1) and (1, 3), therefore f has a local max at x = 1' — and loses the conclusion point because f' did not actually change sign at x = 1. The chart forces you to look at the row for x = 1 itself: the sign on the left and the sign on the right must differ. A sign chart is not an inventory; it is a comparison. Train yourself to circle the column for c and ask, 'left vs right, same or different?' before writing the conclusion word.
Reading first derivative test MCQ stems
The multiple-choice section of the AP Calculus exam signals a first derivative test problem in predictable ways. Stems that give a graph of f' and ask about the behaviour of f are the most common. The vocabulary they lean on includes 'increasing', 'decreasing', 'local maximum', 'local minimum', and 'point of inflection' — and the trap answers almost always swap one of those for a near-synonym such as 'absolute maximum' or 'concave up'. Reading the stem twice for the exact adjective (local versus absolute) is the single fastest correction a student can make.
A typical AB-level stem reads: 'The graph of y = f'(x) is shown. At which value of x does f have a local minimum?' The four answer choices are roots or sign-change locations of f', and the correct answer is the point where f' changes from negative to positive. The distractor at the same x-value where f' is zero but does not change sign is the trap. The way to avoid it is to draw a tiny sign row under each candidate x: -, 0, + versus -, 0, -. The first gives a local min, the second gives neither. This 15-second sketch turns a guess into a defended answer.
BC-level stems add calculus-specific twists. A common one is: 'The function f is differentiable. Given f' on an interval and f'' on the same interval, which statement must be true?' The correct answer almost always hinges on a sign comparison of f' alone — the second derivative is a distractor, present to lure students into the second derivative test. If you see f' given as data, the first derivative test is the natural path; switch to the second derivative test only when f' is hard to factor or when the test point method is messy. The College Board has a habit of placing both derivatives in the same stem precisely to test which method the candidate picks.
Numerical answer stems are the third family. 'For what value of x does f have a local maximum, given f'(x) = x^3 - 6x^2 + 11x - 6?' The first derivative test is faster than the second derivative test here: factor f'(x) to (x-1)(x-2)(x-3), identify the critical numbers 1, 2, 3, and test the sign of f' on (1, 2) and (2, 3) by plugging 1.5. f'(1.5) = (0.5)(-0.5)(-1.5) = positive; f'(2.5) = (1.5)(0.5)(-0.5) = negative. Sign changes from positive to negative at x = 2, so x = 2 is a local max. Total time: under 90 seconds, and the answer is locked in by the sign row rather than a memory of the second derivative rule.
How the rubric scores the first derivative test on FRQs
AP Calculus free-response scoring operates on a points-per-row basis, and the first derivative test is unusually generous because it generates three to four scorable rows from a single problem. A typical rubric on a 'find the local extrema of f on the interval [a, b]' question breaks down as: (1) finding f'(x) correctly, (2) solving f'(x) = 0 to obtain the critical numbers, (3) producing a sign chart or sign analysis of f', and (4) stating the local max and local min at the correct x-values with the correct vocabulary. Each row is roughly 1 point, with a 1-point 'global answer' credit if the question asks for the absolute maximum on a closed interval.
The rubric language is precise. 'Credit is awarded for the value of the critical number' means the numeric answer must appear on the page; an erased or implied value is a lost point. 'Credit is awarded for a correct sign of f' on an interval' means the sign must be stated, not merely implied by an arrow on the chart; the reader should be able to circle 'positive' or 'negative' without inference. 'Credit is awarded for the local maximum/minimum conclusion' means the word 'local' must be present, and the x-value must match the sign change. A candidate who writes 'max at x = 2' without 'local' may still get credit depending on the rubric's wording for that year, but the safe move is always to write the full phrase.
When a sign row is wrong, the rubric usually preserves the points above and below. So if your chart says f' is positive on (-1, 2) when it should be negative, the points for 'found critical number x = 2' and 'identified the left interval' are typically still awarded, while the conclusion point is withheld. The exception is the conclusion point on a multipart FRQ: once the chart is wrong, the local-extremum answer is also wrong, and the absolute-maximum answer (if asked) follows from it. Reading a released FRQ rubric for a recent year shows that on average, candidates who got the chart wrong still scored 60 percent of the available points on the question, which is a useful calibration: the first derivative test rewards partial work.
One more rubric detail worth memorising: the College Board distinguishes between 'a critical number of f' and 'a local extremum of f'. The two sets overlap but are not identical. x = 0 is a critical number of f(x) = x^3 but not a local extremum; the rubric will not credit it as a local maximum or minimum, even though it is correctly listed as a critical number. On an FRQ that asks for both, the safe practice is to list the critical numbers first, then state explicitly which ones are local extrema and which are not. The reader is then able to award the critical-number row, the sign-chart row, and the conclusion row independently, and your score is decoupled from any single arithmetic slip.
First derivative test versus second derivative test: a comparison
The first derivative test and the second derivative test are not interchangeable, and the AP Calculus exam expects candidates to know which is appropriate in a given setting. The first derivative test requires the sign of f' on both sides of c, obtained either algebraically (factoring f') or graphically (reading a given f' graph). The second derivative test requires the sign of f''(c) only, but it is inconclusive when f''(c) = 0 and it requires f' to be continuous at c. Most AB and BC rubrics accept either test on a free-response problem, but the time cost differs sharply.
The first derivative test is faster when f'(x) factors into linear pieces, because the sign change is read directly from the factors. The second derivative test is faster when f'(x) is messy but f''(x) is simple, or when the candidate is asked to classify a specific x-value rather than describe the function on an interval. On a 15-minute FRQ, picking the right test is itself a strategic decision: a 90-second first derivative test on a hard derivative is a worse trade than a 60-second second derivative test on the same problem.
The table below summarises the trade-offs that show up most often on AP Calculus scoring guides.
| Feature | First derivative test | Second derivative test |
|---|---|---|
| Information required | Sign of f' on both sides of c | Value of f''(c) |
| Best when | f'(x) factors easily into linear factors | f'(x) is hard to factor, f''(x) is simple |
| Time cost on a typical FRQ | 60 to 120 seconds | 45 to 90 seconds |
| Works at corners (f' undefined) | Yes, with one-sided sign analysis | No, requires f' continuous at c |
| Inconclusive cases | Rare (only when f' has the same sign on both sides) | Common (when f''(c) = 0) |
| Rubric rows credited | 3 to 4 (critical number, sign, conclusion, sometimes global) | 2 to 3 (sign of f''(c), conclusion, sometimes critical number) |
In my experience tutoring AP Calculus students, the most common avoidable mistake is reaching for the second derivative test by reflex. The first derivative test is the default on the exam; the second derivative test is the fallback when factoring is impractical. Switching the order in your head — first derivative test first, second derivative test only if the first is too expensive — typically recovers 5 to 10 minutes across the free-response section, and that is the difference between finishing the last sub-question and leaving it blank.
Worked example: a multipart FRQ prompt
Consider the function f(x) = x^4 - 4x^3 + 4x^2 on the interval [-1, 3]. A typical AP Calculus AB FRQ might ask: (a) find the critical numbers of f, (b) find the intervals on which f is increasing and decreasing, (c) identify the local maxima and minima, and (d) find the absolute maximum and absolute minimum of f on [-1, 3]. Each part is a row in the rubric and each can be answered using the first derivative test.
Part (a) is algebraic. f'(x) = 4x^3 - 12x^2 + 8x = 4x(x^2 - 3x + 2) = 4x(x-1)(x-2). The critical numbers are x = 0, x = 1, x = 2. The factored form is what makes the rest of the problem tractable. The rubric awards the row for the set {0, 1, 2}, and a candidate who writes 'x = 0, 1, 2' on the page has the row.
Part (b) is the sign chart. The four intervals on the real line are (-infinity, 0), (0, 1), (1, 2), (2, infinity). On the relevant sub-intervals within [-1, 3], the signs are: negative on (-1, 0), positive on (0, 1), negative on (1, 2), positive on (2, 3). The rubric awards the row for stating 'f is decreasing on (-1, 0), increasing on (0, 1), decreasing on (1, 2), increasing on (2, 3)' with the intervals in brackets. Missing the sign on even one interval is a one-row penalty.
Part (c) reads the chart. f' changes from negative to positive at x = 0, so f has a local minimum at x = 0 with f(0) = 0. f' changes from positive to negative at x = 1, so f has a local maximum at x = 1 with f(1) = 1. f' changes from negative to positive at x = 2, so f has a local minimum at x = 2 with f(2) = 0. The rubric awards the row for each correct (x-value, function value, type) triple, and the word 'local' must appear.
Part (d) is the global question. On [-1, 3], f(-1) = 9, f(0) = 0, f(1) = 1, f(2) = 0, f(3) = 9. The absolute maximum is 9, attained at x = -1 and x = 3; the absolute minimum is 0, attained at x = 0 and x = 2. The rubric awards the row for stating the y-values, and a second row for the x-values. The first derivative test does not, by itself, answer part (d) — you also have to evaluate f at the endpoints. Candidates who skip the endpoints and report only the local extrema lose both rows.
The total points on a question like this are typically 9, and a candidate who executes the first derivative test cleanly on parts (a) through (c) can pick up 6 of those points in under 8 minutes. That is the structural payoff of the test: a single algebraic step (factoring f') unlocks a cascade of rubric rows.
First derivative test on implicit, parametric, and BC-specific functions
On the AP Calculus BC exam, the first derivative test extends naturally to functions defined implicitly or parametrically. The mechanics are the same: find the critical numbers, build a sign chart of the derivative with respect to the independent variable, and read the sign change. The only twist is the chain rule. For an implicitly defined function y satisfying x^2 + y^2 = 25, the derivative dy/dx = -x/y is defined wherever y is nonzero. A critical number in x occurs when dy/dx = 0, which is when x = 0 (and y is not zero, so y = ±5). The first derivative test on x: just to the left of 0, dy/dx is positive when y is positive (say y near 5) and negative when y is negative (y near -5). Sign change at x = 0 along the upper branch: local max at (0, 5). Along the lower branch: local min at (0, -5). The rubric on BC FRQs is more lenient about intermediate calculus and stricter about final vocabulary.
Parametric functions on the interval [a, b] are scored identically. Given x(t) and y(t) differentiable, the derivative dy/dx is (dy/dt) / (dx/dt) wherever dx/dt is nonzero. A critical number t0 satisfies dy/dt = 0 with dx/dt nonzero. The first derivative test on t0 uses the sign of dy/dt on the two sides of t0, just as in the explicit case. The only rubric trap on parametric FRQs is forgetting that dy/dt = 0 is the critical condition, not dy/dx = 0, and the candidates who internalise this on day one of BC review tend to score higher on the parametric question than the candidates who wait until spring.
For functions expressed as polar curves r = f(theta), the first derivative test applies to the radial behaviour but not directly to the Cartesian y-coordinate. Most BC rubrics do not require a polar-curve classification, but when they do appear in a multiple-choice stem, the trick is to convert to parametric form (x = r cos theta, y = r sin theta) and proceed. The exam never asks for a full local-extrema argument on a polar curve in a free-response setting; it asks at most for the r-value at a critical angle, and the first derivative test is overkill for that stem. Recognise the question type and choose the test that fits the rubric.
Common pitfalls and how to avoid them
The first pitfall is sign errors on the chart. They are the most common scoring-killing error and they survive into AP-exam week because students practice without a sign chart. A reliable countermeasure is to factor f'(x) first, then read signs directly from the factors. Each linear factor changes sign once at its root; the product's sign flips each time the cumulative count of negative factors crosses an odd number. Drawing the factor signs on a number line, then multiplying them, is faster than plugging test points and is less error-prone because the structure is visible.
The second pitfall is omitting the function value at the extremum. The rubric on a 'local maximum' row requires both the x-value and the f(x)-value. Candidates who write 'local max at x = 1' and skip f(1) lose half a row. The fix is mechanical: after writing the conclusion, immediately write 'f(1) = ...' underneath. This is the cheapest insurance on the exam, and it costs 5 seconds.
The third pitfall is conflating local and absolute. A candidate reports the absolute maximum as the answer to a 'local maximum' question and loses the conclusion point. The reverse happens too: reporting the local maximum when the question asks for the absolute. The fix is to underline the adjective in the question stem before you start writing. Underline 'local' or 'absolute' on every FRQ prompt, even when the question looks obvious. A 3-second underlining step prevents a one-row deduction.
The fourth pitfall is misclassifying an endpoint. Endpoints are not local extrema in the strict calculus sense unless the function is one-sided and the one-sided inequality holds. The AP exam sometimes asks for 'the largest value of f on the closed interval', and a candidate who reports the endpoint as a local maximum loses vocabulary points. The right move is to answer 'the absolute maximum is f(b) at the right endpoint x = b' and skip the word 'local'. The rubric does not require the word 'local' for an endpoint answer, and the candidate is protected from the deduction.
The fifth pitfall is testing the sign of f' at the critical number itself. f'(c) is zero (or undefined) by definition of a critical number, so it carries no sign information. A candidate who writes 'f'(1) = 0, therefore f has a local max at x = 1' is misreading the first derivative test. The test is a comparison of left and right signs, not a statement about the value at c. Train yourself to write the chart in two columns flanking c, with c in the middle, so the visual structure enforces the comparison.
The sixth pitfall is over-relying on the second derivative test because it is shorter. On a function where f'' vanishes at multiple critical numbers, the second derivative test is inconclusive and the candidate wastes time. The first derivative test is never inconclusive on a factored f'; it is only inconclusive when the sign of f' does not change, which is itself a useful answer. Default to the first derivative test in your preparation, and the second derivative test becomes a fallback rather than a habit.
Preparing the first derivative test for exam day
Preparation for the first derivative test on AP Calculus is a question of repetition with a rubric in hand. Three to five College Board-released FRQs from the past several administrations all contain a first derivative test prompt in either part (b) or part (c) of a multipart question, and the rubrics are public. A reliable six-week study plan: week one, factor 20 derivatives and build sign charts without a calculator; week two, solve the local-extrema parts of two released FRQs per day and compare your sign chart to the rubric; week three, add the absolute-extremum parts; week four, mix in BC-specific implicit and parametric prompts; week five, time yourself on a full FRQ section; week six, redo the problems you missed and isolate the rubric row that was lost.
On exam day, the first derivative test is one of the highest-yield techniques a candidate can deploy, because it converts a one-line computation into three or four rubric rows. Time budget: 90 seconds to factor f', 60 seconds to build the sign chart, 30 seconds to write the local-extrema conclusions, 30 seconds to evaluate f at the critical points and endpoints. Total: 3.5 minutes for a question that is worth roughly 4 to 5 points. The ratio of points to time is the highest on the free-response section, and it is the reason the first derivative test is a featured technique in every AP Calculus review book.
The soft target is to enter the exam able to write a sign chart from memory on a generic cubic or quartic f' in under 60 seconds, with a vocabulary that distinguishes local from absolute and a habit of writing the function value at every reported extremum. Candidates who reach that level reliably score in the 4 to 5 band on the AP Calculus AB or BC exam, and the first derivative test is the single largest contributor to that outcome.
Conclusion and next steps
The first derivative test is the default local-extrema argument on AP Calculus, and the exam rewards it with a multi-row rubric whenever it appears. Candidates who internalise the conditional structure, build sign charts from factored derivatives, and use the precise vocabulary of 'local maximum' and 'local minimum' on free-response questions convert a one-line rule into a defensible 5-level argument. The next study step is to take a released AP Calculus FRQ on local extrema, time yourself on the sign-chart construction, and grade your work against the published rubric. AP Courses' AP Calculus AB and BC programmes drill the first derivative test as a featured technique, with sign-chart critiques timed against the rubric language and a 5-point FRQ build-out that targets the local-extrema parts of the multipart prompt.