AP Statistics FRQ Guide 2026 — Free Response Tips, Format & Scoring

The AP Statistics free response section is worth 50% of your total score. It has 6 questions — 5 standard FRQs and 1 longer Investigative Task. Here's how to approach each one.

AP Stats FRQ Format

The 60 FRQ raw points scale to 50 composite points (50% of 100).

How AP Stats FRQs Are Scored

AP Statistics uses a unique E/P/I scoring system — not simple point-by-point rubrics like most AP exams.

Each part of each FRQ is scored as:

• E (Essentially Complete) = 4 points worth of credit
• P (Partially Complete) = 2-3 points credit
• I (Incorrect) = 0 points

Final score per FRQ combines the E/P/I ratings across all parts into a total out of 4, then converts to the 9-point scale.

What this means practically: Partial credit is available, but vague or incomplete answers consistently land at P, not E. The difference between E and P is usually one specific detail — the correct conclusion, the right condition, or a complete justification.

The 5 Standard FRQs — What They Test

Each standard FRQ covers one or two major statistics topics. Common question types:

Probability & distributions:

• Calculate probability using normal distribution, binomial, or geometric
• Interpret a probability in context
• Find expected value and standard deviation

Inference (most common):

• Set up and conduct a hypothesis test (t-test, z-test, chi-square, ANOVA)
• Construct and interpret a confidence interval
• Check conditions for inference (independence, normality, sample size)

Data analysis:

• Describe a distribution (shape, center, spread, outliers)
• Interpret regression output (slope, r², residuals)
• Compare two distributions using parallel boxplots or histograms

Experimental design:

• Identify the response variable, explanatory variable, and confounding variables
• Distinguish observational study from experiment
• Explain how randomization reduces bias

The Investigative Task (FRQ 6)

FRQ 6 is worth 15 points — 25% of the entire FRQ section. It's longer and multi-part, testing your ability to synthesize statistical reasoning across several connected ideas.

What makes it different:

• Requires applying multiple statistical concepts together
• Often introduces a novel scenario you haven't seen before
• Tests statistical thinking, not just procedure recall
• Partial credit is very accessible if you show your reasoning

Strategy: Don't skip this question. Even if you can't complete every part, partial credit on FRQ 6 is worth more than full credit on a standard FRQ.

The 4 Conditions Students Miss Most

1. Conditions for inference — always check them Every hypothesis test and confidence interval requires checking conditions. Graders deduct when you skip them.

• For t-tests: random sample, independence (n < 10% of population), approximately normal (n ≥ 30 or stated normal)
• For proportions: np ≥ 10 and n(1-p) ≥ 10
• For chi-square: all expected counts ≥ 5

2. Write conclusions in context — always "Reject H₀" earns 0. "We reject H₀. There is sufficient evidence that the mean weight of packages differs from 16 oz" earns the point.

3. Interpret, don't just calculate "r = 0.87" earns 0. "r = 0.87 indicates a strong positive linear association between study hours and test scores" earns the point.

4. Distinguish between association and causation If the study is observational (not an experiment), you cannot conclude causation. Graders specifically watch for this error.

FRQ Answer Template for Hypothesis Tests

Use this structure every time:

1. State hypotheses: H₀ and Hₐ in terms of the parameter (μ, p, etc.)
2. Name the test: "I will use a one-sample t-test for means because..."
3. Check conditions: Random, independent, normal — with justification
4. Calculate: Test statistic + p-value (show work)
5. Conclude in context: "Because p = [value] < α = 0.05, we reject H₀. There is sufficient evidence that..."

Missing any step costs partial credit even if your math is perfect.

FRQ Answer Template for Confidence Intervals

1. State: "I will construct a 95% confidence interval for [parameter]"
2. Check conditions: Random, independent, normal
3. Calculate: Point estimate ± margin of error (show formula)
4. Interpret: "We are 95% confident that the true [parameter] is between [lower] and [upper]"

The interpretation must include: the confidence level, "true [parameter]", and the interval bounds in context.

Score Impact of FRQs

With compositeMax = 100 and FRQs worth 50 pts total:

FRQ performance matters enormously. Each FRQ point ≈ 0.83 composite points, while each MC question ≈ 1.25 composite points — but you have 60 FRQ points vs. 40 MC, making FRQs collectively more impactful.

Related Resources

• AP Statistics Score Calculator
• AP Statistics Score Curve 2026
• AP Statistics Practice Test — 20 Questions
• Is AP Statistics Hard?
• AP Biology FRQ Guide
• How to Score a 5 on Any AP Exam