AP Statistics random variables cheat sheet

By Jude Wallis · Published

For a discrete random variable, the mean is μX=∑xi⋅P(xi)\mu_X = \sum x_i \cdot P(x_i) and the standard deviation is σX=∑(xi−μX)2⋅P(xi)\sigma_X = \sqrt{\sum (x_i - \mu_X)^2 \cdot P(x_i)}. A binomial variable has mean npnp and standard deviation np(1−p)\sqrt{np(1-p)}. Normal probabilities are areas found with z-scores.

AP Statistics: Unit 2 (topics 2.8 Introduction to Random Variables and Probability Distributions, 2.9 Parameters of Random Variables, 2.10 The Binomial Distribution, 2.11 The Normal Distribution). Random variables are Unit 2, topics 2.8 through 2.11, of the Fall 2026 AP Statistics course. The geometric distribution and combining random variables were removed in the redesign and are named here only as outside the framework.

Discrete random variables

A discrete random variable XX takes a countable set of values xix_i, each with probability P(xi)P(x_i).

QuantityFormulaMeaning
Valid distributionEvery P(xi)P(x_i) is between 0 and 1, and ∑P(xi)=1\sum P(x_i) = 1Find a missing probability by subtracting the others from 1
Mean (expected value)μX=∑xi⋅P(xi)\mu_X = \sum x_i \cdot P(x_i)Long-run average value of XX over many repetitions
VarianceσX2=∑(xi−μX)2⋅P(xi)\sigma_X^2 = \sum (x_i - \mu_X)^2 \cdot P(x_i)Square of the standard deviation
Standard deviationσX=∑(xi−μX)2⋅P(xi)\sigma_X = \sqrt{\sum (x_i - \mu_X)^2 \cdot P(x_i)}Typical distance of XX from its mean
Cumulative probabilityP(X≤x)P(X \leq x)Sum of the probabilities of all values up to and including xx

μX\mu_X is read "mu of X" and σX\sigma_X is read "sigma of X". The mean does not have to be a value that XX can take. Interpret both in context with units: "over many households, the average number of pets is about 2.05". See how to find expected value.

Turning words into probability statements

For a random variable that takes whole-number values:

PhraseStatementIn terms of cumulative probability
Exactly 3P(X=3)P(X = 3)P(X≤3)−P(X≤2)P(X \leq 3) - P(X \leq 2)
At most 3P(X≤3)P(X \leq 3)P(X≤3)P(X \leq 3)
Fewer than 3P(X<3)P(X < 3)P(X≤2)P(X \leq 2)
At least 3P(X≥3)P(X \geq 3)1−P(X≤2)1 - P(X \leq 2)
More than 3P(X>3)P(X > 3)1−P(X≤3)1 - P(X \leq 3)
Between 2 and 5, inclusiveP(2≤X≤5)P(2 \leq X \leq 5)P(X≤5)−P(X≤1)P(X \leq 5) - P(X \leq 1)

For a discrete variable, << and ≤\leq give different answers. For a continuous variable such as a normal one, the probability of any single exact value is 0, so P(X<a)=P(X≤a)P(X < a) = P(X \leq a).

The binomial model

XX is binomial when all four conditions hold.

ConditionMeaning
Two outcomesEach trial is a success or a failure
Independent trialsThe result of one trial does not change the probability on another
Fixed number of trialsnn is set in advance
Same probabilityThe probability of success pp is the same on every trial
QuantityFormula
Probability of exactly xx successesP(X=x)=(nx)px(1−p)n−xP(X = x) = \binom{n}{x} p^x (1-p)^{n-x}
MeanμX=np\mu_X = np
Standard deviationσX=np(1−p)\sigma_X = \sqrt{np(1-p)}

(nx)\binom{n}{x} is read "n choose x" and counts the ways to place xx successes among nn trials. On a TI-84, binompdf(n, p, x) gives P(X=x)P(X = x) and binomcdf(n, p, x) gives P(X≤x)P(X \leq x). See binomial probability, how to tell if a situation is binomial, and the binomial probability calculator.

The normal model

A normal distribution is continuous, unimodal, bell-shaped, and symmetric. It is set by its mean μ\mu and standard deviation σ\sigma. Probabilities are areas under the curve, and the total area is 1.

TaskMethod
Standardize a valuez=x−μσz = \frac{x - \mu}{\sigma}
Area to the left of xxLook up zz in the z-table, or use technology
Area to the right of xx1−(area to the left)1 - (\text{area to the left})
Area between two valuesLarger left area minus smaller left area
Value for a given percentileFind the zz with that area to its left, then x=μ+zσx = \mu + z\sigma
Empirical ruleApproximate share of values
Within 1 standard deviation of the mean68%
Within 2 standard deviations of the mean95%
Within 3 standard deviations of the mean99.7%

Example: for a normal variable with μ=70\mu = 70 and σ=8\sigma = 8, the value 82 has z=82−708=1.5z = \frac{82 - 70}{8} = 1.5. The area to the left is 0.9332, so P(X>82)=1−0.9332=0.0668P(X > 82) = 1 - 0.9332 = 0.0668. The 90th percentile uses z=1.28z = 1.28, which gives x=70+1.28(8)=80.24x = 70 + 1.28(8) = 80.24. See how to find a z-score and the normal distribution calculator.

Outside the Fall 2026 framework

The College Board's list of topics removed for the Fall 2026 course includes two that older review materials place next to these:

  • The geometric distribution, which counts the trials until the first success.
  • Combining random variables, which gave rules for the mean and standard deviation of a sum or difference of two random variables.

The related idea that the variances of independent statistics add still appears in the course, inside the standard deviation formulas for a difference of two sample proportions and a difference of two sample means. See is the geometric distribution on the AP Statistics exam and what was removed from AP Statistics.

Mean and standard deviation of a discrete random variable

The number of pets XX in a randomly chosen household in one town takes the values 0, 1, 2, 3, and 4 with probabilities 0.10, 0.25, 0.30, 0.20, and 0.15. Find the mean and standard deviation of XX.

  1. Check: 0.10+0.25+0.30+0.20+0.15=1.000.10 + 0.25 + 0.30 + 0.20 + 0.15 = 1.00.

  2. Mean: μX=0(0.10)+1(0.25)+2(0.30)+3(0.20)+4(0.15)=0+0.25+0.60+0.60+0.60=2.05\mu_X = 0(0.10) + 1(0.25) + 2(0.30) + 3(0.20) + 4(0.15) = 0 + 0.25 + 0.60 + 0.60 + 0.60 = 2.05.

  3. Deviations from 2.05: −2.05,−1.05,−0.05,0.95,1.95-2.05, -1.05, -0.05, 0.95, 1.95. Squared: 4.2025,1.1025,0.0025,0.9025,3.80254.2025, 1.1025, 0.0025, 0.9025, 3.8025.

  4. Weight each square by its probability: 4.2025(0.10)+1.1025(0.25)+0.0025(0.30)+0.9025(0.20)+3.8025(0.15)=0.42025+0.275625+0.00075+0.1805+0.570375=1.44754.2025(0.10) + 1.1025(0.25) + 0.0025(0.30) + 0.9025(0.20) + 3.8025(0.15) = 0.42025 + 0.275625 + 0.00075 + 0.1805 + 0.570375 = 1.4475. This is the variance.

  5. Standard deviation: σX=1.4475=1.2031\sigma_X = \sqrt{1.4475} = 1.2031.

μX=2.05\mu_X = 2.05 pets and σX≈1.20\sigma_X \approx 1.20 pets. Over many households the average number of pets is about 2.05, and the number of pets in a household typically differs from that by about 1.2.

Binomial mean, standard deviation, and probabilities

A spinner lands on red with probability 0.3 on each spin, and spins are independent. Let XX be the number of reds in 12 spins. Find the mean and standard deviation of XX, P(X=4)P(X = 4), and P(X≥3)P(X \geq 3).

  1. XX is binomial with n=12n = 12 and p=0.3p = 0.3: two outcomes, independent spins, a fixed number of spins, and the same probability each time.

  2. Mean: μX=np=12(0.3)=3.6\mu_X = np = 12(0.3) = 3.6.

  3. Standard deviation: σX=12(0.3)(0.7)=2.52=1.587\sigma_X = \sqrt{12(0.3)(0.7)} = \sqrt{2.52} = 1.587.

  4. Exactly 4: P(X=4)=(124)(0.3)4(0.7)8=495(0.0081)(0.057648)=0.2311P(X = 4) = \binom{12}{4}(0.3)^4(0.7)^8 = 495(0.0081)(0.057648) = 0.2311.

  5. At least 3 uses the complement: P(X≥3)=1−P(X≤2)P(X \geq 3) = 1 - P(X \leq 2).

  6. P(X≤2)=P(X=0)+P(X=1)+P(X=2)=0.0138+0.0712+0.1678=0.2528P(X \leq 2) = P(X = 0) + P(X = 1) + P(X = 2) = 0.0138 + 0.0712 + 0.1678 = 0.2528.

  7. P(X≥3)=1−0.2528=0.7472P(X \geq 3) = 1 - 0.2528 = 0.7472.

μX=3.6\mu_X = 3.6 reds, σX≈1.59\sigma_X \approx 1.59 reds, P(X=4)≈0.231P(X = 4) \approx 0.231, and P(X≥3)≈0.747P(X \geq 3) \approx 0.747.

Frequently asked questions

Is expected value the same as the mean?

Yes. For a random variable, "expected value" and "mean" name the same number, μX=∑xi⋅P(xi)\mu_X = \sum x_i \cdot P(x_i). It is the long-run average over many repetitions and may be a value the variable never takes.

How do I find an "at least" binomial probability?

Use the complement. P(X≥k)=1−P(X≤k−1)P(X \geq k) = 1 - P(X \leq k - 1). For "at least 3", subtract the cumulative probability of 2 or fewer from 1. See binompdf vs binomcdf.

Is the geometric distribution part of the course?

No. The College Board lists the geometric distribution among the topics removed for the Fall 2026 course. The binomial distribution remains as topic 2.10.