Mean vs standard deviation of a random variable

By Jude Wallis · Published

The mean of a random variable, μX\mu_X, is its long-run average value: the sum of each value times its probability. The standard deviation, σX\sigma_X, is the typical distance of the values from that mean: the square root of the probability-weighted squared deviations.

AP Statistics: Unit 2 (topics 2.9 Parameters of Random Variables). The mean and standard deviation of a discrete random variable are the two parameters of Unit 2 topic 2.9, Parameters of Random Variables, in the Fall 2026 AP Statistics course. The binomial shortcuts for both come in topic 2.10. Unit 2 is 15% to 25% of the multiple-choice section.

Mean vs standard deviation of a random variable: the short answer

A probability distribution is summarized by two parameters that answer two different questions.

  • The mean μX\mu_X ("mu of X"), also called the expected value E(X)E(X), measures center. It tells you the average outcome over a very large number of repetitions.
  • The standard deviation σX\sigma_X ("sigma of X") measures variability. It tells you how far individual outcomes typically fall from the mean.

The mean alone cannot tell you how much outcomes vary, and the standard deviation alone cannot tell you where they are centered. Two random variables can share a mean and have very different standard deviations, so a full description needs both.

What the mean of a random variable measures

For a discrete random variable XX with possible values xix_i and probabilities P(xi)P(x_i), the mean is

μX=∑xi P(xi)\mu_X = \sum x_i \, P(x_i)

Multiply each value by its probability and add the products. The result is a weighted average of the values, with the probabilities as weights, so likelier values pull the mean toward themselves.

Interpret it as a long-run average. If XX is the payout of a raffle ticket and μX=3\mu_X = 3 dollars, then over a very large number of tickets the average payout per ticket is about 3 dollars. The mean does not have to be a value the variable can take. A fair die has μX=3.5\mu_X = 3.5, and no roll shows 3.5. The computation is walked through in how to find expected value.

What the standard deviation of a random variable measures

The standard deviation uses the mean as its reference point:

σX=∑(xi−μX)2 P(xi)\sigma_X = \sqrt{\sum (x_i - \mu_X)^2 \, P(x_i)}

For each value, find its deviation from the mean, square it, and weight it by the probability of that value. Add those up to get the variance σX2\sigma_X^2, then take the square root.

Interpret it as a typical distance. If σX=6\sigma_X = 6 dollars, individual payouts typically differ from the mean payout by about 6 dollars. A standard deviation is never negative, and it equals 0 only when the variable takes a single value with probability 1.

The square root matters for units. The variance is in squared units, such as dollars squared, and the standard deviation returns to the original units, which is why it is the one reported in context. See why variance is in squared units.

The differences side by side

FeatureMean of a random variableStandard deviation of a random variable
SymbolμX\mu_X or E(X)E(X)σX\sigma_X or SD(X)SD(X)
MeasuresCenterVariability
Formula (discrete)∑xi P(xi)\sum x_i \, P(x_i)∑(xi−μX)2 P(xi)\sqrt{\sum (x_i - \mu_X)^2 \, P(x_i)}
InterpretationLong-run average valueTypical distance of values from the mean
UnitsSame as XXSame as XX
Can be negativeYesNo
Needs the other one firstNoYes, the mean is inside the formula
Binomial shortcutnpnpnp(1−p)\sqrt{np(1-p)}

Both are parameters: fixed numbers that describe the distribution. They are computed from the probability model with no data involved, which separates them from the sample mean xˉ\bar{x} and sample standard deviation ss computed from observations. That contrast is drawn in expected value vs sample mean.

When to use which

Use the mean when the question is about what happens on average: the average profit per sale, the average number of customers per hour, whether a game favors the player in the long run.

Use the standard deviation when the question is about consistency or risk: how much the outcome changes from one trial to the next, or which of two options is more predictable.

Many decisions need both. Suppose two games each pay 3 dollars on average. In one the payout is always close to 3 dollars, and in the other it is usually 0 with an occasional large prize. The means cannot separate them. The standard deviations can.

When the random variable is binomial, skip the general formulas and use the shortcuts μX=np\mu_X = np and σX=np(1−p)\sigma_X = \sqrt{np(1-p)}. For 50 trials with p=0.2p = 0.2, the mean is 50(0.2)=1050(0.2) = 10 successes and the standard deviation is 50(0.2)(0.8)=8≈2.828\sqrt{50(0.2)(0.8)} = \sqrt{8} \approx 2.828 successes.

The classic mix-up and how to avoid it

The first error is interpreting the mean as the most likely value or as what will happen on a single trial. For the raffle ticket in the worked example below, the mean is 3 dollars, yet no ticket ever pays 3 dollars and the most likely payout is 0. The mean describes an average over many trials.

The second error is stopping at the variance. The sum ∑(xi−μX)2P(xi)\sum (x_i - \mu_X)^2 P(x_i) is the variance. The standard deviation needs one more step, the square root. A variance of 36 means a standard deviation of 6.

The third error is leaving out the probabilities. Adding the squared deviations without multiplying each by P(xi)P(x_i), or averaging the values as if they were equally likely, treats the distribution as uniform when it is not.

The fourth error is an interpretation with no context. A complete statement names the variable and gives units: "Over many tickets, the average payout is about 3 dollars per ticket, and individual payouts typically differ from that by about 6 dollars."

Mean and standard deviation of a raffle payout

A raffle ticket pays 0 dollars with probability 0.7, 5 dollars with probability 0.2, and 20 dollars with probability 0.1. Let X be the payout. Find and interpret the mean and the standard deviation of X.

  1. Check that the probabilities form a distribution: 0.7+0.2+0.1=1.00.7 + 0.2 + 0.1 = 1.0.

  2. Mean: μX=0(0.7)+5(0.2)+20(0.1)=0+1+2=3\mu_X = 0(0.7) + 5(0.2) + 20(0.1) = 0 + 1 + 2 = 3 dollars.

  3. Deviations from the mean: 0−3=−30 - 3 = -3, 5−3=25 - 3 = 2, and 20−3=1720 - 3 = 17.

  4. Squared deviations: (−3)2=9(-3)^2 = 9, 22=42^2 = 4, and 172=28917^2 = 289.

  5. Weight each by its probability and add to get the variance: 9(0.7)+4(0.2)+289(0.1)=6.3+0.8+28.9=369(0.7) + 4(0.2) + 289(0.1) = 6.3 + 0.8 + 28.9 = 36 dollars squared.

  6. Standard deviation: σX=36=6\sigma_X = \sqrt{36} = 6 dollars.

  7. Interpret both in context. Over many tickets the average payout is about 3 dollars per ticket. Individual payouts typically differ from that average by about 6 dollars.

μX=3\mu_X = 3 dollars and σX=6\sigma_X = 6 dollars. The standard deviation is twice the mean here, which reflects a payout that is usually 0 and occasionally large.

Same mean, different standard deviation

A second game pays 2 dollars with probability 0.5 and 4 dollars with probability 0.5. Let Y be its payout. Find the mean and standard deviation of Y and compare them with the raffle ticket, which has mean 3 dollars and standard deviation 6 dollars.

  1. Mean: μY=2(0.5)+4(0.5)=1+2=3\mu_Y = 2(0.5) + 4(0.5) = 1 + 2 = 3 dollars.

  2. Deviations from the mean: 2−3=−12 - 3 = -1 and 4−3=14 - 3 = 1. Squared, both are 1.

  3. Variance: 1(0.5)+1(0.5)=11(0.5) + 1(0.5) = 1 dollar squared.

  4. Standard deviation: σY=1=1\sigma_Y = \sqrt{1} = 1 dollar.

  5. Compare the means: both games average 3 dollars per play in the long run, so neither pays more on average.

  6. Compare the standard deviations: 1 dollar against 6 dollars. The second game's payout is always within 1 dollar of its mean. The raffle's payouts range from 0 to 20 dollars.

μY=3\mu_Y = 3 dollars and σY=1\sigma_Y = 1 dollar. The two games have the same mean and standard deviations of 1 dollar and 6 dollars. They pay the same on average, and the raffle is far less predictable from play to play.

Frequently asked questions

Is the mean of a random variable the same as the expected value?

Yes. Expected value, E(X)E(X), and the mean of a random variable, μX\mu_X, are two names for the same number: the sum of each possible value times its probability.

Can the standard deviation of a random variable be larger than its mean?

Yes. The two measure different things, so there is no rule tying their sizes together. A raffle payout with mean 3 dollars and standard deviation 6 dollars is one example. The standard deviation only has to be zero or positive.

Do I need the mean to find the standard deviation?

Yes. The standard deviation formula measures each value's distance from the mean, so compute μX\mu_X first. Then square each deviation, weight it by its probability, add, and take the square root.