Statistics Symbols: Greek Letters in Statistics

Statistics symbols are a set of letters and marks that name the parts of a formula, and they follow one simple rule: Greek letters usually describe a whole population, while Latin letters describe a sample. So μ is a population mean, and x̄ is the mean of a sample. Once you know this pattern, most formulas become much easier to read.

This guide covers the Greek letters you will meet in an intro statistics course. It also pairs each one with its sample version. For the broader picture, see our hub on Greek letters in math.

Population vs Sample: The Core Rule

A population is every member of the group you care about. A sample, by contrast, is the smaller group you actually measure. Numbers that describe a population are called parameters, and numbers from a sample are called statistics.

Parameters usually get Greek letters because they are often unknown. Sample statistics then get Latin letters because you calculate them from data. For example, OpenStax Introductory Statistics uses s for a sample standard deviation and σ for a population standard deviation.

Memory trick: Greek for the group you cannot fully see, Latin for the sample in your hands. Therefore, μ and σ describe the population, while x̄ and s describe your data.

Statistics Symbols Table: Greek Letters and Their Meanings

The table below lists the most common Greek statistics symbols. In addition, the right-hand column shows the usual sample counterpart. Textbooks vary a little, so always check your course notes.

SymbolNameMeaningSample version
μmuPopulation meanx̄ (x-bar)
σsigmaPopulation standard deviations
σ²sigma squaredPopulation variances²
Σcapital sigmaSum of valuesSame symbol
ρrhoPopulation correlation coefficientr
βbetaRegression coefficient; Type II error rateb or β̂
αalphaSignificance level (Type I error rate)—
χ²chi-squaredChi-squared statistic and distribution—
λlambdaRate in a Poisson distributionλ̂
θthetaA general unknown parameterθ̂ (theta-hat)
εepsilonRandom error term in a modele (residual)
νnuDegrees of freedom (in some texts)df
Φ φphiStandard normal CDF (Φ) and density (φ)—

Mean and Spread: μ, σ and σ²

The population mean is written μ (mu). It is the sum of all values divided by the population size, N. In contrast, the sample mean x̄ divides by the sample size, n.

Spread uses sigma. Lowercase σ is the population standard deviation, and σ² is the population variance.

Their sample versions are s and s². The key difference is the denominator: samples divide by n − 1, while populations divide by N.

  • Population variance: σ² = Σ(x − μ)² / N
  • Sample variance: s² = Σ(x − x̄)² / (n − 1)
  • Standard deviation: the square root of the variance in each case

These formulas come from OpenStax Introductory Business Statistics. For a full walk-through with numbers, read our guide to the standard deviation symbol (σ and s). Our page on the variance symbol then explains σ² and s² step by step.

Proportions: p, p̂ and π

Proportions follow the same pattern, with one twist. Many books write the population proportion as p and the sample proportion as p̂ (p-hat).

However, some texts use π for the population proportion instead. In that case, π is not 3.14159, so read the definitions first.

Software and reference guides may use other letters again. For instance, the NIST Engineering Statistics Handbook writes the sample variance as s² with N − 1 in the denominator. The idea is the same, even though the letters differ.

The Summation Sign Σ

Capital sigma (Σ) means “add up.” So Σx is the total of all x values in a data set. You will see it inside nearly every formula for the mean, variance and regression.

Statistics often writes Σ without limits, which simply means “sum over all the data.” However, a full version shows an index, such as Σ from i = 1 to n of xᵢ. Our guide to sigma notation explains the index and limits with worked examples.

Hypothesis Testing Symbols

Hypothesis tests use α and β for the two kinds of error. First, α is the significance level, the chance of rejecting a true null hypothesis. Common choices are 0.05 and 0.01.

Next, β is the chance of missing a real effect, called a Type II error. Power is then 1 − β, the chance of catching that effect. Meanwhile, the null hypothesis itself is written H₀, with a Latin H.

Test statistics come next. The chi-squared statistic, χ², tests counts in categories. Similarly, z and t are Latin-letter statistics that compare a result with a normal or t distribution.

Regression and Correlation Statistics Symbols

Correlation uses ρ (rho) for the population and r for a sample. Both run from −1 to 1. A value near 0 means little straight-line relationship.

In regression, β₀ and β₁ are the true intercept and slope. Their estimates are written b₀ and b₁, or β̂₀ and β̂₁ with a hat.

The hat symbol always means “estimated from data.” Finally, ε is the random error in the model, the part the line cannot explain.

Probability Distribution Symbols

Several distributions are named or described by Greek letters. For example, a normal distribution is written N(μ, σ²), with its mean and variance inside. Also, the standard normal CDF is Φ(z), which gives the area to the left of z.

The Poisson distribution uses λ for its average rate. Likewise, the chi-squared and t distributions depend on degrees of freedom. Some books write those as ν, while most intro courses use df.

Typing tip: On Windows, Alt+229 gives σ and Alt+228 gives Σ on the numeric keypad. For μ, type 03BC in Word and press Alt+X. Alt+230 gives the micro sign µ, a lookalike.

How to Type Statistics Symbols

In LaTeX, use \mu, \sigma, \rho and \chi. For x̄, type \bar{x}, and for a hat, type \hat{\beta}. Our list of LaTeX Greek letter commands covers every letter.

In Word or Google Docs, the equation editor accepts the same backslash names. For example, typing \sigma and a space turns into σ in Word’s equation mode. Our step-by-step guide to inserting Greek letters in Word shows the menus as well.

Common Mix-Ups

Most mix-ups in statistics come from the population vs sample rule above.

  • μ vs x̄: μ is the population mean, while x̄ is the mean of a sample.
  • σ vs s: σ describes a whole population, while s is the sample standard deviation.
  • ρ vs r: ρ is the population correlation, and r is the sample correlation you calculate from data.
  • χ² vs x²: chi-squared uses the Greek letter χ, not a Latin x, even though they look alike.

Frequently Asked Questions

What is the symbol for mean in statistics?

The population mean is μ (mu), and the sample mean is x̄ (x-bar). Use μ when you have every value, and x̄ when you have only a sample.

What does σ mean in statistics?

Lowercase sigma (σ) is the population standard deviation. The sample standard deviation is written s, and both measure how spread out the data are.

What does α mean in statistics?

Alpha (α) is the significance level of a hypothesis test. It is the chance of rejecting the null hypothesis when it is actually true, often set at 0.05.

Why do statistics use Greek letters for populations?

Greek letters mark parameters, which describe a whole population and are usually unknown. Latin letters then mark the sample statistics you calculate to estimate them.

What does a hat over a symbol mean?

A hat, as in β̂ or p̂, means the value is an estimate from sample data. It separates the estimate from the true population value.

Similar Posts