Random Variables and Probability Distributions

Introduction

A random variable turns the outcomes of a random process into numbers. Once outcomes have numerical values, you can describe uncertainty with a probability distribution, calculate long-run averages, compare variability, and build mathematical models for real situations. This aiMOOC is designed for Grades 11–13 and develops the topic from core definitions to model selection, simulation, and interpretation.

You will learn how random variables, probability distributions, expected values, and variances work together. You will also compare important discrete and continuous distributions and decide when each model is appropriate.

Learning Goals

By the end of this course, you should be able to explain what a random variable represents, distinguish discrete from continuous random variables, construct and interpret probability distributions, calculate expected value and variance, work with cumulative probabilities, recognize common distribution families, and justify a suitable probability model for a real-world situation.

You should also be able to move between verbal descriptions, tables, formulas, graphs, and simulations. These representations are different ways of expressing the same probabilistic structure.

From Random Experiments to Random Variables

A random experiment is a process whose exact outcome is uncertain before it occurs. The set of all possible outcomes is the sample space. A random variable assigns a real number to every outcome in that sample space.

For example, toss two fair coins. The sample space is HH, HT, TH, TT. Define the random variable $X$ as the number of heads. Then $X$ can take the values 0, 1, and 2. The random variable does not describe a new experiment; it provides a numerical way to describe the outcomes of the original experiment.

This distinction matters: an outcome might be HT, while the corresponding value of the random variable is 1.

Discrete and Continuous Random Variables

A discrete random variable takes values that can be listed individually, such as 0, 1, 2, 3, and so on. Counts are usually discrete. Examples include the number of defective components in a batch, the number of goals in a match, or the number of customers arriving in ten minutes.

A continuous random variable can take any value in an interval. Measurements such as time, mass, temperature, and distance are commonly modeled as continuous. Between two possible values there are infinitely many others.

Probability Distributions

A probability distribution tells you how probability is allocated across the possible values of a random variable. It is a complete mathematical description of the variable's uncertainty.

Probability Mass Functions

For a discrete random variable $X$ , the probability mass function or PMF gives $p (x) = P (X = x)$ .

A valid PMF satisfies two rules:

Every probability is between 0 and 1.
The probabilities of all possible values add to 1.

Suppose $X$ is the number of heads in two fair coin tosses. Then $P (X = 0) = 1 / 4$ , $P (X = 1) = 1 / 2$ , and $P (X = 2) = 1 / 4$ . The values and probabilities together form the probability distribution of $X$ .

Probability Density Functions

For a continuous random variable, probability is described by a probability density function or PDF, written $f (x)$ . The density must be nonnegative and the total area under the curve must equal 1.

For continuous variables, probability comes from area. For example, $P (a \leq X \leq b) = \int_{a}^{b} f (x) d x$ .

A single exact value has probability zero in a continuous model: $P (X = c) = 0$ . This does not mean the value is impossible. It means that probability is assigned to intervals rather than isolated points.

Cumulative Distribution Functions

The cumulative distribution function or CDF works for both discrete and continuous random variables: $F (x) = P (X \leq x)$ .

A CDF never decreases. It approaches 0 far to the left and 1 far to the right. For a discrete distribution it rises in jumps; for many continuous distributions it increases smoothly.

You can use the CDF to find interval probabilities. For example, when the distribution is continuous, $P (a < X \leq b) = F (b) - F (a)$ .

Expected Value and Variability

A probability distribution is often summarized by its center and spread.

Expected Value

The expected value or mean of a random variable describes its long-run average over many repetitions of the same random process.

For a discrete random variable, $E (X) = \sum_{x} x P (X = x)$ .

For a continuous random variable with density $f (x)$ , $E (X) = \int_{- \infty}^{\infty} x f (x) d x$ .

The expected value does not have to be a possible outcome. If a game has expected winnings of 2.40 currency units, you should interpret this as a long-run average, not as a guarantee that one play pays exactly 2.40.

Variance and Standard Deviation

The variance measures how far values typically spread around the mean: $V a r (X) = E [(X - E (X))^{2}]$ .

An equivalent computational formula is $V a r (X) = E (X^{2}) - [E (X)]^{2}$ .

The standard deviation is $\sqrt{V a r (X)}$ . Because it has the same unit as the random variable, standard deviation is often easier to interpret than variance.

For a linear transformation $Y = a X + b$ , $E (Y) = a E (X) + b$ and $V a r (Y) = a^{2} V a r (X)$ .

Adding a constant shifts the distribution but does not change its variance. Multiplying by a constant changes both center and spread.

Important Discrete Distributions

Discrete distributions are especially useful for binary outcomes and counts.

Bernoulli Distribution

A Bernoulli random variable models one trial with two outcomes, commonly coded as 1 for success and 0 for failure. If the probability of success is $p$ , then $P (X = 1) = p$ and $P (X = 0) = 1 - p$ .

Its mean is $E (X) = p$ , and its variance is $p (1 - p)$ . Bernoulli variables are the building blocks of the binomial distribution.

Fehler beim Erstellen des Vorschaubildes:

Binomial Distribution

A binomial random variable counts successes in a fixed number $n$ of independent Bernoulli trials when each trial has the same success probability $p$ .

If $X \sim B i n (n, p)$ , then $P (X = k) = (\binom{n}{k}) p^{k} (1 - p)^{n - k}$ for $k = 0, 1, \dots, n$ .

The mean is $n p$ , and the variance is $n p (1 - p)$ . Before using a binomial model, check the key assumptions: a fixed number of trials, two outcomes per trial, constant success probability, and independence.

Datei:Binomial distribution pmf.svg

Poisson Distribution

A Poisson random variable models the number of events occurring in a fixed interval of time, distance, area, or volume when events occur independently at an approximately constant average rate.

If $X \sim P o i s s o n (λ)$ , then $P (X = k) = e^{- λ} \frac{λ^{k}}{k!}$ for nonnegative integers $k$ .

Its mean and variance are both $λ$ . Examples include calls arriving at a help desk per minute or imperfections occurring along a length of material, provided the Poisson assumptions are reasonable.

Datei:Poisson pmf.svg

Important Continuous Distributions

Continuous distributions are used when the random variable can vary across an interval.

Uniform Distribution

A uniform distribution on the interval from $a$ to $b$ assigns equal density across that interval: $f (x) = \frac{1}{b - a}$ for $a \leq x \leq b$ .

Its mean is $(a + b) / 2$ , and its variance is $(b - a)^{2} / 12$ . The uniform model is useful when every value in a bounded interval is treated as equally plausible.

Normal Distribution

The normal distribution is a symmetric, bell-shaped continuous distribution. It is determined by its mean $μ$ and standard deviation $σ$ .

Its density is $f (x) = \frac{1}{σ \sqrt{2 π}} e^{- \frac{(x - μ)^{2}}{2 σ^{2}}}$ .

A normal variable can be standardized using the z-score $z = \frac{x - μ}{σ}$ . This transformation shows how many standard deviations a value lies above or below the mean.

Normal models arise in many measurement settings and are also important because sums and averages often become approximately normal under suitable conditions, an idea formalized by the central limit theorem.

Exponential Distribution

An exponential random variable is often used to model waiting time until the next event in a Poisson process. With rate $λ > 0$ , its density is $f (x) = λ e^{- λ x}$ for $x \geq 0$ .

Its mean is $1 / λ$ , and its variance is $1 / λ^{2}$ . The exponential distribution has the memoryless property: given that you have already waited a certain amount of time, the additional waiting-time distribution does not depend on how long you have already waited.

Datei:Exponential probability density.svg

Choosing a Probability Model

A probability distribution is useful only when its assumptions match the situation closely enough for the purpose of the model. Ask what is being measured, what values are possible, whether the variable is a count or a measurement, and what mechanism could plausibly generate the data.

A Bernoulli model fits one yes-or-no trial. A binomial model fits a fixed number of similar independent trials. A Poisson model can fit event counts over an interval. A uniform model fits bounded outcomes treated as equally likely in density. A normal model fits many roughly symmetric measurements. An exponential model fits certain waiting times.

Real data rarely follow a theoretical model perfectly. Your task is therefore not simply to recognize a formula but to justify why the assumptions are reasonable, check whether the model agrees with data, and explain the limitations of your conclusion.

Simulation and the Law of Large Numbers

Simulation lets you imitate random processes repeatedly. If you simulate many trials, the empirical relative frequencies usually stabilize near their theoretical probabilities. Similarly, the sample average tends to move toward the expected value under appropriate conditions. This behavior is connected with the law of large numbers.

Simulation is useful when exact calculations are difficult, when you want to check an analytic result, or when you want to study the effect of assumptions. A spreadsheet, graphing calculator, or programming language can generate random values from common distributions and summarize the results.

A strong simulation study should state the model, parameters, number of repetitions, quantity recorded, and comparison between simulated and theoretical results.

Interactive Tasks

Quiz: Test Your Knowledge

What is a random variable? (A function assigning a numerical value to each outcome) (!A list containing only equally likely outcomes) (!A graph that must be bell shaped) (!A number that can never change)

Which type of random variable is most suitable for the number of emails received in one hour? (Discrete random variable) (!Continuous random variable) (!Deterministic constant) (!Standard score)

What must the probabilities in a discrete probability mass function sum to? (One) (!Zero) (!The expected value) (!The sample size)

How is probability obtained from a continuous probability density function? (By finding area under the density curve) (!By reading the height as the exact probability) (!By adding all possible x values) (!By counting only integer outcomes)

What does the cumulative distribution function F at x represent? (The probability that the random variable is at most x) (!The probability that the random variable equals its mean) (!The variance below x) (!The density only at x)

What does expected value describe? (The long run average value of a random variable) (!The largest possible outcome) (!The most recent observed value) (!The width of the sample space)

Which conditions characterize a binomial model? (A fixed number of independent trials with constant success probability) (!A continuous measurement with no upper bound) (!A changing number of trials with changing probabilities) (!A waiting time with memory)

For a Poisson distribution with parameter lambda, what are the mean and variance? (Both are lambda) (!Both are one) (!The mean is zero and the variance is one) (!The mean is lambda squared and the variance is lambda)

What happens to variance when a random variable is transformed from X to 3X plus 5? (It is multiplied by nine) (!It is increased by five) (!It is multiplied by three) (!It stays unchanged)

What does a z score of negative two mean? (The value is two standard deviations below the mean) (!The value is two standard deviations above the mean) (!The probability of the value is negative) (!The variance equals two)

Memory Game

Random variable	Numerical assignment to outcomes of a random experiment
PMF	Function giving probabilities for discrete values
PDF	Density whose interval areas give continuous probabilities
CDF	Function giving the probability of being at or below a value
Expectation	Long-run average described by a probability distribution
Variance	Average squared distance from the mean
Binomial	Distribution for successes in a fixed number of suitable trials
Exponential	Distribution often used for waiting time to the next event

Drag and Drop

Match the correct terms.	Topic
Bernoulli distribution	One trial with two possible outcomes
Binomial distribution	Count of successes in a fixed number of independent trials
Poisson distribution	Count of events in an interval at an approximately constant rate
Normal distribution	Symmetric bell-shaped continuous model
Exponential distribution	Waiting time until the next event in a Poisson process

...

Crossword Puzzle

Bernoulli	Which distribution models one binary trial?
Binomial	Which distribution counts successes in a fixed number of suitable trials?
Poisson	Which distribution often models event counts in a fixed interval?
Density	What function type gives continuous probability through area?
Variance	What measure describes squared spread around the mean?
Expectation	What term names the long-run average of a random variable?

LearningApps

Cloze Text

A random variable assigns a numerical value to each

of a random experiment. A variable that takes countable values is called

. A discrete probability mass function must have probabilities that sum to

. For a continuous model, probability is represented by

under the density curve. The function that gives the probability of being at or below a value is the

. The long-run average of a random variable is its

. The measure based on squared distance from the mean is the

. A count of successes in a fixed number of suitable independent trials is modeled by the

distribution. A symmetric bell-shaped model determined by a mean and standard deviation is the

distribution. A waiting time to the next event in a Poisson process can be modeled by the

distribution.

Open-Ended Tasks

Easy

Random variable diary: Identify five uncertain quantities from everyday life, define a random variable for each one, and classify each as discrete or continuous.
Coin experiment: Toss two coins at least 40 times, record the number of heads, create an empirical distribution, and compare it with the theoretical probabilities.
Distribution sketching: Draw one discrete and one continuous probability distribution, label the axes, and explain what the heights or areas mean.
Expected value explanation: Write a short explanation for a younger student showing why expected value is a long-run average rather than a guaranteed outcome.

Standard

Binomial investigation: Choose a realistic repeated-trial situation, justify whether the binomial assumptions are reasonable, calculate at least three probabilities, and interpret them in context.
Simulation comparison: Use a spreadsheet, calculator, or program to simulate a probability distribution, compare empirical frequencies with theoretical probabilities, and discuss why they differ.
Normal data study: Collect or find at least 30 measurements, calculate their mean and standard deviation, create a graph, and evaluate whether a normal model is plausible.
Probability interview: Interview someone who uses uncertain data in work or daily decisions, identify the random quantities involved, and present how probability affects their choices.

Advanced

Model selection project: Analyze one real dataset and compare at least two candidate probability distributions, explaining which model fits better and why.
Transformation investigation: Start with a random variable X, study several transformations of the form aX plus b, and verify experimentally how the mean and variance change.
Poisson and exponential connection: Design a simulation that generates event counts and waiting times from the same rate parameter, then explain the relationship between the two models.
Probability explainer video: Produce a three-to-five-minute video that teaches PMF, PDF, CDF, expectation, and variance through one coherent real-world example and includes at least one original visual.

Learning Assessment

Model justification assessment: Given three real-world scenarios, select an appropriate distribution for each, state the assumptions, and explain one reason each model could fail.
Representation assessment: Convert a probability problem between a verbal description, a table or formula, and a graph, then explain what information remains unchanged across the representations.
Expectation and risk assessment: Compare two games or investments using expected value and standard deviation, then defend a decision while discussing why expectation alone is insufficient.
CDF reasoning assessment: Use a supplied cumulative distribution graph to estimate several probabilities and explain how the graph would differ for discrete and continuous variables.
Simulation validation assessment: Design a simulation to test one theoretical probability result, justify the number of repetitions, summarize the results, and explain the remaining random variation.
Transfer assessment: Choose a new context from science, economics, engineering, or social research and build a probability model that includes a random variable, distribution, parameters, interpretation, and limitations.

Evidence of Learning

Knowledge: You can define random variables, distinguish PMFs, PDFs, and CDFs, and explain the role of expectation, variance, and common distribution families.

Skills: You can calculate and interpret probabilities, expected values, variances, standard scores, and interval probabilities; move between formulas, tables, graphs, and simulations; and check modeling assumptions.

Products: Strong evidence may include a correctly analyzed experiment, a simulation report, a distribution graph, a model comparison, a dataset study, or an explanatory video.

Transfer: You can recognize probabilistic structure in unfamiliar contexts, select a defensible model, justify its assumptions, communicate uncertainty clearly, and explain limitations instead of treating a model as exact reality.

OERs on the Topic

The embedded English Wikipedia article provides an open reference on probability distributions and links to related distribution families and concepts.

Linked Learning Areas

Random Variables and Probability Distributions

The topic connects directly with mathematics, statistics, data science, economics, physics, computer science, engineering, risk analysis, and quantitative decision-making. It is especially suitable for upper-secondary mathematics and introductory university statistics.