In AP Statistics, the mean is the average of a quantitative data set, calculated as x̄ = (1/n)Σxᵢ (sum of values divided by the number of values). It measures center but is nonresistant, meaning outliers and skew pull it away from the median, which is why the AP exam constantly asks you to choose between the two.
The mean is the balance point of a distribution. Add up every data value, divide by how many values there are, and you get x̄ (for a sample) or μ (for a population). The CED gives the formal version in Topic 1.7: x̄ = (1/n)Σxᵢ. That notation matters on the exam, because x̄ is a statistic (computed from a sample) while μ is a parameter (a fixed value describing a population), and inference in later units is all about using x̄ to estimate μ.
The single most-tested property of the mean is that it's nonresistant. One huge outlier drags the mean toward it, while the median barely moves. Picture the mean as the point where a dotplot would balance on a seesaw. A skewed tail or an extreme value tips the balance, so in a right-skewed distribution the mean sits above the median, and in a left-skewed one it sits below. The mean also generalizes beyond raw data lists. In Unit 4 it becomes the expected value of a random variable (μ_X = Σxᵢ·P(xᵢ)), and in Unit 5 it describes the center of sampling distributions.
The mean first appears in Unit 1 under LO 1.7.A (calculate measures of center) and LO 1.7.C (explain which measure of center to use), where describing shape, center, and variability per LO 1.6.A is the bread-and-butter skill. But it doesn't stay there. The standardized values (xᵢ - x̄)/sₓ inside the correlation formula in Topic 2.5 are built on the mean. In Unit 4 it shows up as expected value (LO 4.8.A), as np for binomial variables (LO 4.11.A), and in the rules for combining and transforming random variables (LOs 4.9.A and 4.9.B). In Unit 5, sampling distributions have their own means, like μ_p̂ = p (LO 5.5.A). If you don't understand the mean cold, half the course's formulas stop making sense.
Keep studying AP Statistics Unit 4
Median (Unit 1)
The median is the mean's resistant counterpart. Comparing them is a built-in skew detector: mean greater than median suggests right skew, mean less than median suggests left skew, and roughly equal suggests symmetry. LO 1.7.C asks you to justify choosing one over the other based on outliers and shape.
Standard Deviation (Units 1 & 4)
Standard deviation measures the typical distance of data values from the mean, so it literally cannot exist without the mean. They travel as a pair: both are nonresistant, and both get reported together whenever a distribution is roughly symmetric.
Expected Value of Random Variables (Unit 4)
The mean of a discrete random variable, μ_X = Σxᵢ·P(xᵢ), is just a weighted average where probabilities replace the 1/n. Same idea, new clothes. For a binomial variable the shortcut is μ_X = np, and linear combination rules tell you the mean of aX + bY is aμ_X + bμ_Y.
Sampling Distributions (Unit 5)
Take many samples and each one has its own statistic, and those statistics form a distribution with its own mean. For sample proportions, μ_p̂ = p, which is the unbiasedness idea that makes confidence intervals and hypothesis tests in Units 6-9 work.
Multiple-choice questions love the mean-versus-median comparison. A classic stem gives you a five-number summary with one extreme maximum (say Max = 89 when Q3 = 31) and asks which statistic is most affected. The answer is the mean (or another nonresistant measure like the range or standard deviation), because the median and IQR shrug off outliers. Other MCQs ask when the mean and median are approximately equal (symmetric distributions). On FRQs, the mean appears constantly in context. The 2017 exam used the mean diameter of melons from two distributors in a normal-distribution problem, and the 2018 investigative task started from a reported mean systolic blood pressure of 122 for the U.S. population. Expect to calculate expected values, interpret a mean in context with units (required for full credit), and use μ_p̂ = p or μ_X = np when working with sampling and binomial distributions. The fastest way to lose points is interpreting a mean without context or confusing x̄ with μ in your hypotheses.
Both measure center, but they answer different questions. The mean is the arithmetic balance point and uses every value, so a single outlier yanks it around. The median is the positional middle and only cares about order, so it's resistant. AP Stats rule of thumb: report the median (with IQR) for skewed data or data with outliers, and the mean (with standard deviation) for roughly symmetric data. If a question asks which measure is 'most affected' by an extreme value, the mean is your answer; if it asks which 'best describes' a skewed distribution's center, pick the median.
The mean is the sum of all data values divided by the number of values, written x̄ = (1/n)Σxᵢ for a sample and μ for a population.
The mean is nonresistant, so outliers and skewed tails pull it toward the extreme values while the median stays put.
In a right-skewed distribution the mean is typically greater than the median; in a left-skewed distribution it's typically less; in a symmetric distribution they're approximately equal.
For a discrete random variable, the mean is the expected value μ_X = Σxᵢ·P(xᵢ), and for a binomial variable it simplifies to μ_X = np.
Linear transformations shift and scale the mean predictably: if Y = a + bX, then μ_Y = a + bμ_X, and the mean of aX + bY is aμ_X + bμ_Y.
Sampling distributions have means too, like μ_p̂ = p for sample proportions, which is what makes sample statistics useful for estimating population parameters.
The mean is the average of a quantitative data set: add every value and divide by the count, written x̄ = (1/n)Σxᵢ. It's one of the two main measures of center on the exam, alongside the median.
No. The mean is nonresistant, so when data are skewed or contain outliers, the median describes the center better. AP graders expect you to justify your choice using the shape of the distribution, per LO 1.7.C.
The mean is the balance point that uses every value, while the median is the middle value when data are ordered. An extreme value moves the mean a lot but barely affects the median, which is why the mean exceeds the median in right-skewed data.
x̄ is the sample mean, a statistic calculated from data; μ is the population mean, a fixed parameter you usually don't know. Mixing them up in hypotheses (writing H₀ about x̄ instead of μ) costs points on inference FRQs.
Yes. The expected value of a random variable is its mean, μ_X = Σxᵢ·P(xᵢ), which is a weighted average using probabilities instead of equal 1/n weights. For a binomial variable it's simply np.