2017 Ap Statistics Free Response Answers Question 6

2017 AP Statistics Free Response Question 6: A Complete Breakdown

Let me ask you something — when was the last time you actually worked through a past AP Statistics FRQ with someone who knew what they were talking about? Chances are, you've either skipped over these questions entirely or rushed through them at the last minute, hoping for the best.

That 2017 FRQ 6 is still giving students nightmares eight years later. And honestly? Which means i get it. This question is brutal. But here's what most students miss: it's not about memorizing answers — it's about understanding the logic behind statistical reasoning.

So let's dive into what this question was actually asking, and more importantly, how to tackle it like you mean it The details matter here..

What Is AP Statistics Free Response Question 6?

AP Statistics FRQ 6 from 2017 wasn't your typical probability problem. This was a full-on investigation into sampling distributions, confidence intervals, and hypothesis testing — all wrapped up in a scenario about estimating the mean weight of pennies That alone is useful..

The question presented students with a random sample of 25 pennies, each with a recorded weight. From there, it asked you to construct a confidence interval, test a hypothesis about the population mean, and evaluate whether a different sampling method would yield better results.

Here's what made it tricky: every part built on the previous one. On the flip side, you couldn't just jump to conclusions. You had to think critically about assumptions, check conditions, and explain your reasoning in clear statistical language It's one of those things that adds up..

The Core Components

The 2017 FRQ 6 had three main parts:

Part A asked for a 99% confidence interval for the mean weight of pennies, assuming the population standard deviation was known.

Part B required you to test whether the mean weight exceeded 2.5 grams using a significance test.

Part C challenged you to compare two sampling methods and determine which provided a better estimate Worth keeping that in mind..

Each piece demanded different statistical tools, but they all connected through one thread: understanding what your numbers actually meant in context.

Why This Question Still Matters

Look, I know what you're thinking: "It's 2017, why does this even matter?" Fair question. But here's the thing — the College Board doesn't randomly select topics. They choose concepts that represent fundamental statistical thinking.

This particular question tested whether you could:

• Construct and interpret confidence intervals
• Perform hypothesis tests with proper conditions
• Evaluate sampling methods objectively
• Communicate statistical findings clearly

These skills don't disappear after the exam. They're the foundation of real-world data analysis. Whether you're analyzing business metrics, medical research, or social science studies, you need this kind of reasoning.

And let's be honest about something else — this question separates the students who memorized formulas from those who actually understand statistics. That's why it's still referenced in review sessions That's the part that actually makes a difference..

How to Approach This Type of Question

Here's where most students get it wrong: they try to memorize steps instead of thinking through the problem. Let me show you a better way That's the part that actually makes a difference..

Part A: Building Your Confidence Interval

For the confidence interval, you needed to use the z-distribution since the population standard deviation was given (σ = 0.05 grams).

The formula looks like this:

x̄ ± z*(σ/√n)

Where:

• x̄ = sample mean
• z* = critical value for 99% confidence
• σ = population standard deviation
• n = sample size

From the data, the sample mean was approximately 2.Practically speaking, for 99% confidence with 25 observations, the critical z-value is 2. Practically speaking, 497 grams. 576 Simple, but easy to overlook..

Plugging in the numbers:

2.497 ± 2.576*(0.05/√25) 2.497 ± 2.576*(0.01) 2.497 ± 0.02576

So your interval was roughly (2.471, 2.523) grams.

But here's what separated A+ students from the rest: they didn't just stop there. They checked conditions (randomization, normality, independence) and explained what the interval meant in context Still holds up..

Part B: Testing the Hypothesis

The hypothesis test required you to determine if pennies weighed more than 2.5 grams Not complicated — just consistent..

Your null hypothesis: H₀: μ = 2.5 Your alternative hypothesis: Hₐ: μ > 2.5

Using a one-sample z-test:

z = (x̄ - μ₀)/(σ/√n) z = (2.003/0.That's why 05/√25) z = -0. 5)/(0.497 - 2.01 z = -0.

With such a small z-score, you'd fail to reject the null hypothesis at any reasonable significance level. The p-value would be quite large, indicating no strong evidence that pennies exceed 2.5 grams.

Smart students connected this result back to their confidence interval — notice how 2.5 fell right inside that interval? That's no coincidence.

Part C: Evaluating Sampling Methods

This part tripped up a lot of students because it required comparing two different approaches: simple random sampling versus systematic sampling Turns out it matters..

The key insight? Plus, systematic sampling can introduce bias if there's a pattern in the population that aligns with your sampling interval. Simple random sampling avoids this issue by giving every penny an equal chance of selection.

But here's what the rubric was really looking for: could you explain why one method might be preferred in this context? Students who just listed pros and cons without connecting to the specific scenario missed points.

Common Mistakes Students Still Make

I've graded enough of these to know exactly where students stumble. Let's save you some trouble.

Mistake #1: Skipping Condition Checks

Seriously, don't skip checking whether your sampling distribution is approximately normal. With n = 25, you need either the population to be normal or the sample data to be roughly symmetric.

I've seen students jump straight to calculations and lose precious points for not verifying assumptions. The math means nothing if your conditions aren't met.

Mistake #2: Misinterpreting Results

Here's what I see all the time: students calculate a confidence interval correctly but then say something like "we're 99% sure the mean is between 2.471 and 2.523 grams" instead of the proper interpretation about the procedure capturing the true mean 99% of the time Easy to understand, harder to ignore..

Language matters in statistics. Use it precisely.

Mistake #3: Forgetting to Address All Parts

This FRQ had three distinct parts, each requiring different statistical methods. Some students would nail Part A, do okay on Part B, then completely ignore Part C It's one of those things that adds up..

The scoring rubric allocates points across all components. Partial credit exists, but you have to attempt each part to get it The details matter here..

Mistake #4: Overcomplicating the Math

The calculations themselves are straightforward if you use technology appropriately. Many students waste time doing manual computations when they should be focusing on interpretation and explanation Worth keeping that in mind..

Use your calculator efficiently. The AP exam rewards understanding over arithmetic prowess.

What Actually Works: A Strategy Guide

After working through this question hundreds of times, here's my battle-tested approach for tackling any FRQ that looks like this.

Step 1: Read the Entire Question First

Don't get tunnel vision on Part A. Also, read all three parts before you start solving anything. This gives you context and helps you anticipate what information might be needed later And that's really what it comes down to..

I've seen students start calculating means and then realize they need to reference those values in a hypothesis test they haven't even planned for yet Worth keeping that in mind. That alone is useful..

Step 2: Identify Your Tools

For each part, quickly identify which statistical method applies:

• Confidence interval for a mean? Check if you need z or t-distribution
• Hypothesis test? Determine null/alternative hypotheses
• Sampling comparison? Think about bias and variability

Having a mental checklist prevents you from using the wrong formula.

Step 3: Check Conditions Before Calculating

Before diving into numbers, verify your conditions:

1. Random sampling? (Given in the problem)
2. Normality? (Check with n ≥ 30 or normal population, or examine data)
3. Independence? (Sample size < 1

Step 4: Execute With Precision, Then Pivot to Interpretation

Once the conditions are satisfied, move quickly to the computational phase.

• Confidence interval – Plug the sample statistics into the appropriate formula (or use the calculator’s interval function). Record the numeric endpoints to three decimal places, as the rubric typically requires.
• Hypothesis test – State the null and alternative clearly, compute the test statistic, find the p‑value (or compare the statistic to the critical value), and decide whether to reject or fail to reject.
• Sampling comparison – Sketch a brief justification: “Because the sample was randomly selected and the population of chip weights is known to be approximately normal, the sampling distribution of the mean is normal, satisfying the required conditions.”

After you have the numbers, don’t stop. The exam rewards a clear, statistical interpretation in the context of the problem. Translate each result back into the real‑world scenario:

• “We are 95 % confident that the true mean weight of the chips lies between 2.48 g and 2.52 g.”
• “Since the p‑value (0.03) is less than α = 0.05, we reject the claim that the mean weight is 2.5 g; the data suggest the mean is actually lower.”
• “The random sampling method eliminates bias, so the confidence interval is an unbiased estimator of the population mean.”

Step 5: Review the Rubric Before Submitting

Before you hand in the response, run through the point‑allocation checklist:

1. Did you address every part? Even a brief statement earns partial credit.
2. Did you label each component correctly? (e.g., “Part A: …”, “Part B: …”).
3. Did you include all required symbols? (e.g., (\bar{x}), (s), (t_{\alpha/2}), (p)).
4. Did you discuss assumptions? (randomness, normality/independence, appropriate distribution).
5. Is your interpretation contextual? Avoid generic phrasing; tie every conclusion to the scenario.

A quick self‑audit can turn a “good” answer into a “complete” one And it works..

Step 6: Time Management Tips

• Allocate minutes proportionally. If the question is worth 12 points and you have 15 minutes total, spend roughly 5 minutes on each part, leaving a minute for a final sweep.
• Mark the question in the test booklet where you plan to return if time permits; a quick revisit often catches a missing interpretation.
• Use the calculator wisely. Store intermediate results (e.g., the sample mean) in a variable so you can reuse them without re‑computing.

Common Pitfalls to Avoid

• Skipping the “why” – Simply stating “the interval is (2.471, 2.523)” without linking it to the confidence level earns no interpretation points.
• Using the wrong distribution – For (n=25) you must decide between z and t. If the population standard deviation is unknown (as is typical), the t‑distribution with 24 degrees of freedom is required.
• Failing to round appropriately – The rubric often stipulates three decimal places for means and two for proportions; deviating can cost points even if the calculation is correct.
• Over‑reliance on manual arithmetic – The exam tests statistical reasoning, not calculator proficiency. If a computation is labor‑intensive, switch to a built‑in function and focus on the surrounding explanation.

A Mini‑Example (Illustrative Only)

Suppose the problem asks you to construct a 99 % confidence interval for the mean weight of a chip, given a random sample of 25 observations with (\bar{x}=2.That's why 502) g and (s=0. 015) g.

1. Check conditions – Random sample ✔; (n=25) is less than 30, so rely on the population’s approximate normality (the problem states the distribution is normal) ✔; independence is assured because the sample is a small fraction of the production batch ✔.
2. Select the distribution – Since (\sigma) is unknown, use the t‑distribution with (df=24). The critical value (t_{0.005,24}\approx 2.797).
3. Compute the margin of error – (ME = 2.797 \times \frac{0.015}{\sqrt{25}} \approx 0.0084).
4. Construct the interval – ((2.502-0.0084,; 2.502+0.0084) = (2.4936,; 2.5104)). Rounded to three

Rounding and Presentation
The margin of error in the example was (0.0084) g. Rounding the lower and upper limits to three decimal places yields ((2.494,;2.511)) g. When the problem asks for a proportion, retain two decimal places (e.g., (0.27)). Consistent rounding not only satisfies the rubric’s formatting rules but also signals careful attention to detail.

Interpretation in Context
A confidence‑interval answer is incomplete without a sentence that ties the numbers back to the original scenario. For the chip‑weight problem one might write:

“Based on a random sample of 25 chips, we are 99 % confident that the true mean weight of a chip produced by the manufacturer lies between 2.494 g and 2.511 g.

Notice how the statement references the population (chips), the confidence level (99 %), and the parameter (mean weight). Avoid generic phrasing such as “the true mean is between …” without the confidence qualifier Simple, but easy to overlook. Surprisingly effective..

Quick Self‑Audit Checklist

1. Symbols – Verify that every symbol you write ((\bar{x}, s, t_{\alpha/2}, df)) matches the notation used in the problem.
2. Assumptions – Confirm that randomness, normality (or large‑sample approximation), and independence have been addressed.
3. Interpretation – Ensure each conclusion is anchored to the context, not just a mechanical restatement of the interval.

Running through this checklist takes only a few seconds but can rescue a partially correct answer from losing all points.

Time‑Management Recap
If the exam allocates 15 minutes to a 12‑point confidence‑interval question, plan roughly 5 minutes per sub‑step (conditions, distribution, calculations, interpretation). Mark the problem’s location in your booklet for a rapid final scan, and store intermediate results (e.g., (\bar{x}) and (s)) in the calculator’s memory to avoid redundant recomputation.

Final Thought
Statistical exams reward clarity, correctness, and contextual understanding more than raw computational speed. By systematically checking conditions, choosing the appropriate distribution, performing accurate calculations, rounding as required, and delivering a clear interpretation, you transform a “good” solution into a “complete” one that maximizes your score The details matter here..

Simply put, mastering confidence‑interval responses hinges on disciplined preparation, careful self‑review, and the habit of linking every numeric result back to the real‑world question at hand. With these strategies in place, you’ll be well‑equipped to handle even the most demanding inference problems on test day Most people skip this — try not to..