It was a Tuesday in May. Thousands of high schoolers sat in dimly lit gyms, clutching TI-84 calculators and praying the "Investigative Task" wouldn't be about something obscure like linguistics or hydraulic fracturing. Looking back, the 2018 AP Stats FRQ answers reveal a lot more than just who knew their formulas. They show exactly where the gap lies between "doing math" and "thinking statistically." Honestly, most people who walked out of that room felt fine, but the scoring guidelines later proved that a lot of students were basically speaking a different language than the graders.
The Problem with the Environmental Group (Question 1)
The first question was supposed to be the "easy" one. It focused on a regional environmental group and the heights of trees. You've got a frequency relative histogram. It looks simple. But then the College Board throws a curveball by asking students to estimate the median.
A lot of kids just looked at the graph and took a wild guess at the middle. Big mistake. The real trick in the 2018 AP Stats FRQ answers for this section was showing the work by identifying the $n = 25$ trees and finding the average of the 13th value. If you didn't mention that the median must fall in the 10-15 foot interval because that's where the cumulative frequency hits the halfway mark, you lost credit. It’s about the "why," not just the "what."
Then came the description of the distribution. Remember S.O.C.S.? Shape, Outliers, Center, Spread. If you forgot even one, the graders were ruthless. You had to say the distribution was skewed to the right. You had to mention the range or the interquartile range. If you just said "it's spread out," you were toast.
That Tricky Significance Test (Question 4)
Question 4 was the one that made people sweat. It was about those "Adler" and "Standard" methods of drying seeds. Students had to determine if there was a significant difference in the proportions of seeds that germinated.
📖 Related: False eyelashes before and after: Why your DIY sets never look like the professional photos
Here is where the 2018 AP Stats FRQ answers get technical. You had to run a two-sample z-test for proportions. Sounds easy? Not quite. Most students failed the "Large Counts" condition. You can't just say "n is big enough." You actually had to calculate the pooled proportion, which was $140/200 = 0.7$. Then you had to show that $100(0.7)$ and $100(0.3)$ were both greater than or equal to 10.
The p-value ended up being around 0.13. Since that's bigger than 0.05, you fail to reject the null. But here's the kicker: the interpretation. You couldn't say "the methods are the same." That’s a statistical sin. You had to say "we do not have convincing evidence that the proportions are different." It’s a subtle distinction, but it’s the difference between a 3 and a 4 on the exam.
Why Context Is Everything
I’ve talked to plenty of readers who grade these for the College Board in Kansas City. They all say the same thing. Students know the numbers, but they forget they’re talking about seeds or trees or people. If your answer for Question 4 didn't mention "germination rates" or "drying methods," you were basically just doing algebra. Statistics is numbers with a story. If you strip the story away, you're not doing stats anymore.
The Infamous Question 6: The Investigative Task
Every year, Question 6 is the boogeyman. In 2018, it was about a company that produces silicon wafers. They were looking at the relationship between the number of chips and the diameter of the wafer.
👉 See also: Exactly What Month is Ramadan 2025 and Why the Dates Shift
The first part was easy—linear regression. But then, they introduced a non-linear model. Students had to compare the two. This is where the 2018 AP Stats FRQ answers required some actual intuition. You couldn't just look at $R^2$. You had to look at the residual plot. If the residual plot has a pattern, the model is a bad fit. Period.
The last part of Question 6 asked for a "proportion of the area." This was basically asking students to think like a calculus student without using calculus. You had to understand that as the diameter increases, the area increases by the square of the radius. If you didn't catch that $Area = \pi r^2$, you were guessing in the dark.
Common Mistakes That Tanked Scores
Looking at the data from that year, certain errors popped up over and over.
- Confusing Correlation with Causation: In Question 2 (about the shoppers), many students implied that being a certain age caused people to use the pharmacy. You can't say that in an observational study. You can only say there's an association.
- Poorly Defined Parameters: When writing the hypotheses for Question 4, you couldn't just write $H_0: p_1 = p_2$. You had to define what $p_1$ and $p_2$ actually were. If you didn't say "where $p_1$ is the true proportion of seeds that germinate using the Adler method," you lost points.
- Rounding Too Early: Some students rounded their z-scores to one decimal place, which messed up their p-values. In the world of AP Stats, precision matters until the very end.
How to Use These Answers to Study
If you're looking at these old prompts to prep for a future exam, don't just read the solutions. That's passive. It's useless.
✨ Don't miss: Dutch Bros Menu Food: What Most People Get Wrong About the Snacks
- Set a timer for 13 minutes per question (25 for Question 6).
- Write your answer out in full sentences.
- Grade yourself using the official rubric, but be mean. If you missed a label on a graph, give yourself a zero for that part.
- Rewrite the answer correctly.
The 2018 AP Stats FRQ answers are a goldmine because they represent a "classic" year. There weren't many weird outliers in the question styles, which makes them perfect for practicing the core logic that the College Board loves.
The Psychology of the Exam
There's a weird mental hurdle with this specific year. Question 3 was about a probability distribution for a raffle. It seemed too simple. A lot of high-achieving students overthought it, trying to use complex binomial formulas when they just needed to multiply a few decimals. Honestly, sometimes the simplest answer is the right one. Don't go looking for ghosts where there aren't any.
The 2018 exam reminds us that "statistical literacy" is the goal. It’s not about being a human calculator. It’s about being able to look at a pile of data and explain what it means to someone who doesn't know a standard deviation from a standard error.
Actionable Steps for Mastering FRQs
To truly master the logic found in the 2018 exam, focus on these three habits:
- Always Define Your Variables: Before you write a single equation, write down what $x$, $y$, $p$, or $\mu$ represents in the context of the problem.
- Check Your Conditions Constantly: Whether it's $n < 10%$ of the population or the "Large Counts" condition, write it down every time. It’s a "free" point that people leave on the table constantly.
- Use the Phrase "In Context": Every conclusion should end with a reference to the actual subject matter—whether it's silicon wafers, tree heights, or seed germination.
If you can explain the 2018 AP Stats FRQ answers to a friend who isn't in the class, you've actually learned the material. If you're just memorizing the steps to a 1-proportion z-test, you're going to struggle when the College Board throws a weirdly phrased question your way next May.