Know the Exam Format
40
Multiple Choice Questions
90 minutes • 50% of score
~2.25 min per question
6
Free Response Questions
90 minutes • 50% of score
5 short + 1 investigative task
3
Hours Total
No penalty for wrong answers
NEVER leave anything blank!
Scoring Insight: You don't need a perfect score. Historically, getting roughly 75%+ of available points earns a 5. Getting about 60-63% earns a 4. A 3 (passing) typically requires around 50%. Every single point matters.
Multiple Choice Strategies
Strategy 1: Answer Every Question
There is NO penalty for guessing. If you have no idea, eliminate what you can and pick from the rest. A blank answer is always wrong; a guess has a chance.
Strategy 2: Read All Five Options
The most common MC mistake is picking the first answer that looks right. Many questions have a "pretty good" answer and a "best" answer. Read all five before choosing.
Strategy 3: Watch for "Trap" Words
Key words that change everything: always, never, must, cannot, proves, causes. Statistics almost never deals in absolutes. Be suspicious of extreme language.
Strategy 4: Use Process of Elimination
If you can eliminate even one answer, your odds improve from 20% to 25%. Eliminate two and you're at 33%. Three eliminated = 50/50. Every elimination helps.
Strategy 5: Don't Overthink
Your first instinct is usually right. If you've studied the material, trust your preparation. Only change an answer if you have a clear reason.
Free Response Strategies
The #1 FRQ Rule: CONTEXT, CONTEXT, CONTEXT
Generic answers lose points. Never write "there is a positive association between x and y." Instead, write "there is a positive association between hours studied and exam score." Use the variable names and units from the problem.
The 4-Step Procedure (Inference Questions)
Most FRQ inference questions require all 4 steps. Missing a step = losing a whole E. The steps:
| Step | What to Include | Common Mistake |
|---|
| STATE | Define parameter, state H0 and Ha | Not defining what p or μ represents in context |
| PLAN | Name the test, check ALL conditions | Forgetting the 10% condition or using wrong values for Large Counts |
| DO | Show formula, calculate test stat & p-value | Not showing the formula or setup |
| CONCLUDE | Compare p-value to α, state conclusion in context | Saying "accept H0" or forgetting context |
Show Your Work (Even When You Use a Calculator)
You CAN use calculator notation (like "1-PropZTest") but you MUST also show:
- The formula or name of the procedure
- The values plugged in
- The test statistic and p-value
If your calculator breaks, you still get credit for correct setup!
FRQ Partial Credit Trick
Even if you can't do the full calculation, write down everything you know: state the hypotheses, name the test, check conditions. You can earn 2-3 out of 4 E/P/I components without even computing the test statistic.
DO on FRQs
- Answer in complete sentences
- Use variable names from the problem
- Include units when relevant
- Show formula setup
- Write something for every part
- Say "fail to reject H0" not "accept"
- Use "convincing evidence" language
DON'T on FRQs
- Leave any part completely blank
- Write just a number with no context
- Say "proves" or "accepts"
- Skip checking conditions
- Contradict yourself (hedging is fine)
- Erase work — cross it out neatly
- Write more than needed (brevity wins)
Top Common Mistakes (Point Killers)
Mistake 1: Confusing Association and Causation
WRONG: "The study shows that eating breakfast causes better test scores."
RIGHT: "The study shows an association between eating breakfast and test scores. Because this is an observational study, we cannot conclude causation."
Only randomized experiments can establish causation. This is one of the most tested concepts.
Mistake 2: Wrong Confidence Interval Interpretation
WRONG: "There is a 95% probability that the true mean is in this interval."
RIGHT: "We are 95% confident that the true mean of [context] is between [lower] and [upper]."
The true parameter is fixed — it's either in the interval or not. "95% confident" describes the METHOD, not this specific interval.
Mistake 3: "Accepting" the Null Hypothesis
WRONG: "We accept H0 and conclude there is no difference."
RIGHT: "We fail to reject H0. There is not convincing evidence of [Ha in context]."
Failing to find evidence against H0 is NOT the same as proving H0 is true. Absence of evidence ≠ evidence of absence.
Mistake 4: Describing Without Comparing
WRONG: "Group A has a mean of 75. Group B has a mean of 82."
RIGHT: "Group B has a higher mean (82) than Group A (75), suggesting Group B performed better."
When asked to COMPARE, you must use comparative language: "higher than," "more spread out than," "similar to."
Mistake 5: Using p-hat vs p0 in the Wrong Place
WRONG: Using p-hat in the standard error of a hypothesis test.
RIGHT: For TESTS, use p0 (the null value) in the SE. For INTERVALS, use p-hat.
In a test, you assume H0 is true, so you use p0. In an interval, you're estimating p, so you use p-hat.
Mistake 6: Subtracting Standard Deviations
WRONG: SD(X - Y) = SD(X) - SD(Y)
RIGHT: Var(X - Y) = Var(X) + Var(Y), then SD = √(Var). Always ADD variances.
Variability always increases when combining variables, whether adding or subtracting.
Mistake 7: Confusing Random Selection and Random Assignment
WRONG: "Random assignment makes the results generalizable to the population."
RIGHT: Random SELECTION → generalizable. Random ASSIGNMENT → causal conclusions.
These serve completely different purposes. Many students mix them up, and the AP exam specifically tests this distinction.
Mistake 8: Forgetting Conditions
WRONG: Jumping straight to the calculation in an inference problem.
RIGHT: Always check: Random? 10% condition? Large Counts / Normal?
Skipping conditions is one of the easiest ways to lose a full E on an FRQ. It takes 30 seconds to write and can be worth a full point.
Calculator Tips (TI-83/84)
Key Functions to Know
| What You Need | Where to Find It |
|---|
| Normal probabilities | 2nd → DISTR → normalcdf(lower, upper, μ, σ) |
| Inverse Normal (find z or x) | 2nd → DISTR → invNorm(area, μ, σ) |
| Binomial probability | 2nd → DISTR → binompdf(n, p, x) or binomcdf |
| 1-Prop z-test | STAT → TESTS → 1-PropZTest |
| 1-Prop z-interval | STAT → TESTS → 1-PropZInt |
| t-test (1-sample) | STAT → TESTS → T-Test |
| 2-Sample t-test | STAT → TESTS → 2-SampTTest |
| Chi-square test | STAT → TESTS → χ²-Test |
| Linear regression t-test | STAT → TESTS → LinRegTTest |
| Descriptive stats | STAT → CALC → 1-Var Stats |
| Regression equation | STAT → CALC → LinReg(ax+b) |
Exam Day Reminders
The Night Before
Get sleep. Seriously. A rested brain performs better than an extra hour of cramming. Lay out your pencils, calculator (with fresh batteries), and ID.
During the MC Section
First pass: answer everything you're confident about. Second pass: tackle the harder ones. Final pass: make sure nothing is blank. Budget ~2 min per question.
During the FRQ Section
Read ALL 6 questions first. Start with the ones you know best. Budget ~13 min for questions 1-5, ~25 min for question 6 (investigative task). Write something for every part.
If You're Stuck
Write what you know. State the hypotheses. Name the test. Check one condition. Partial credit is your friend. A blank page = zero points. Anything written = a chance at points.