A percentile is the value in a distribution below which a given percentage of observations fall. In AP Stats, percentiles measure relative position, so a test score at the 90th percentile beat roughly 90% of all scores, no matter what the actual score was.
A percentile describes where a data value sits relative to everything else in the distribution. If your exam score is at the 84th percentile, about 84% of scores are below yours. The percentile says nothing about your raw score itself, only about your position in the pack.
In the CED, percentiles show up in two places. Topic 1.7 treats them as a measure of position alongside quartiles (Q1 is the 25th percentile, the median is the 50th, Q3 is the 75th). Topic 1.10 takes it further with normal distributions, where you find percentiles by converting a value to a z-score and using a standard normal table or technology like normalcdf. Either way, the core idea is the same. You're answering the question "what fraction of the data falls below this value?"
Percentiles live in Unit 1 (Exploring One-Variable Data) and directly support three learning objectives. AP Stats 1.7.A asks you to calculate measures of position, and percentiles are the main one. AP Stats 1.10.B asks you to determine proportions and percentiles from a normal distribution using z-scores and technology. AP Stats 1.10.C is the big payoff, and its essential knowledge says it plainly: percentiles and z-scores may be used to compare relative positions of points within a data set or between data sets. That comparison skill is what the exam loves. Percentiles let you compare an SAT score to an ACT score, or a rent price in one city to another, even when the raw numbers aren't on the same scale.
Keep studying AP® Statistics Unit 1
Quartiles, Q1, and Q3 (Unit 1)
Quartiles are just percentiles with famous names. Q1 is the 25th percentile, Q3 is the 75th, and the IQR (Q3 minus Q1) tells you the spread of the middle 50% of the data. If you understand percentiles, every boxplot question becomes a percentile question in disguise.
Median (Unit 1)
The median is the 50th percentile, full stop. Half the data sits below it. This is also why the median is resistant to outliers. It only cares about position in the ordered list, not how extreme the values at the ends are.
Empirical Rule (Unit 1)
The empirical rule is secretly a percentile cheat sheet for normal distributions. One standard deviation above the mean lands at about the 84th percentile, because 50% sits below the mean plus half of the middle 68%. Knowing these landmark percentiles lets you sanity-check answers without a calculator.
normalcdf and z-scores (Unit 1)
For a normal distribution, you find a percentile in two steps. Convert the value to a z-score with (x − μ)/σ, then use a standard normal table or normalcdf to get the area to the left. That area to the left IS the percentile. This z-score machinery comes back constantly in later units, so getting it down now pays off all year.
Multiple choice questions usually hand you a mean and standard deviation, then ask you to find the percentile of a value (or work backward from a percentile to a value, which means using invNorm). A classic setup: exam scores are normal with μ = 500 and σ = 100, the top 5% earn scholarships, and a student scores 650. You'd find that 650 sits at about the 93rd percentile, just short of the 95th percentile cutoff. On FRQs, percentiles appear in context-heavy problems. The 2019 exam included a question about a smartphone battery's life span (normal distribution percentile calculations) and an investigative task about rental prices, where relative position within a distribution drove the analysis. The key skills you must show are converting to a z-score, finding the correct area, and interpreting the percentile in context, in a sentence, with units. "The value 650 is at approximately the 93rd percentile, meaning about 93% of scores are below 650" is the kind of interpretation that earns the point.
Both measure relative position (that's exactly what LO 1.10.C says), but they speak different languages. A z-score counts standard deviations from the mean, so z = 1.5 means "1.5 SDs above average." A percentile counts the percentage of data below you, so the 93rd percentile means "above 93% of values." For normal distributions they're convertible (z = 1.5 corresponds to about the 93rd percentile), but z-scores require knowing the mean and SD, while percentiles work for any distribution, normal or not. On comparison questions, either one can justify which value is "relatively higher," but you have to pick one and use it correctly.
A percentile is the value below which a given percentage of observations fall, so the 90th percentile beats about 90% of the data.
The median is the 50th percentile, Q1 is the 25th, and Q3 is the 75th, which means quartiles and boxplots are built entirely out of percentiles.
For a normal distribution, find a percentile by converting the value to a z-score and finding the area to the left using a table or normalcdf.
Percentiles and z-scores both measure relative position and let you compare values across different data sets, which is exactly what LO 1.10.C tests.
A high percentile does not mean a high raw score; it only means the value is high relative to the rest of its own distribution.
When interpreting a percentile on an FRQ, write a full sentence in context, like 'about 93% of scores fall below 650.'
A percentile is the value in a distribution below which a given percentage of observations fall. It measures relative position, so a value at the 80th percentile is greater than about 80% of the data, regardless of the actual numbers involved.
No. The 90th percentile means you did better than about 90% of everyone else, not that you got 90% of anything right. You could score 60% on a brutal test and still land in the 90th percentile if most people scored lower.
A z-score measures how many standard deviations a value is from the mean, while a percentile measures the percentage of data below the value. For normal distributions they're interchangeable (z = 0 is the 50th percentile, z = 1 is roughly the 84th), but percentiles work for any distribution shape.
Convert the value to a z-score using z = (x − μ)/σ, then find the area to the left of that z-score using a standard normal table or normalcdf on your calculator. That left-tail area, written as a percentage, is the percentile.
Yes. The median is defined as the middle value of ordered data, which means 50% of observations fall below it. That's also why Q1 and Q3 are the 25th and 75th percentiles, since they're the medians of the lower and upper halves.
Connect this key term to the AP exam workflow: review the course, practice questions, and check related study tools.
Review units, study guides, and course resources.
Check this vocabulary in multiple-choice context.
Apply key concepts in written AP responses.
Estimate the exam score you are working toward.
Review the highest-yield facts before practice.
Put the full course together before test day.