Mathematics · Year 13

Active learning ideas

Expectation and Variance of Discrete Random Variables

Active learning sticks for expectation and variance because students need to see how theoretical formulas emerge from real data. Repeated simulations and hands-on calculations show why E(X) and Var(X) matter beyond the page, turning abstract sums into quantities they can visualize and critique.

• National Curriculum Attainment Targets
• A-Level: Mathematics - Statistical Distributions

• 30–50 min
• Pairs → Whole Class
• 4 activities

Activity 01

• Collaborative Problem-Solving
• 45 min · Small Groups

Simulation Rotations: Dice and Coins

Prepare stations with dice, coins, and spinners marked for custom probabilities. Groups run 200 trials at each, tally outcomes, calculate empirical expectation and variance from frequency tables. Compare results to theoretical values and discuss convergence.

Explain the meaning of the expected value in the context of a discrete random variable.

Facilitation Tip

During Simulation Rotations, have each group record 50 trials before calculating E(X) to highlight how sample size affects proximity to the theoretical mean.

What to look for

Provide students with a probability distribution table for a discrete random variable (e.g., number of heads in 3 coin flips). Ask them to calculate E(X) and Var(X) on a mini-whiteboard and hold it up for review.

• Apply
• Analyze
• Evaluate
• Create
• Relationship Skills
• Decision-Making
• Self-Management

Activity 02

• Collaborative Problem-Solving
• 35 min · Pairs

Spreadsheet Builds: Distribution Tweaks

Students input probability tables into spreadsheets with formulas for E(X) and Var(X). They adjust probabilities to model scenarios like biased games, graph outcomes, and predict changes in spread. Pairs share and critique each other's models.

Analyze how the variance quantifies the spread of a discrete distribution.

Facilitation Tip

While students build Spreadsheet Builds, circulate and ask them to explain how changing a probability in their distribution affects both E(X) and Var(X) numerically.

What to look for

Present two different probability distributions for the same discrete random variable. Ask students: 'Which distribution has a higher expected value and why? Which distribution is more spread out, indicated by its variance, and what might this mean in a practical scenario?'

• Apply
• Analyze
• Evaluate
• Create
• Relationship Skills
• Decision-Making
• Self-Management

Activity 03

• Collaborative Problem-Solving
• 50 min · Whole Class

Game Design Challenge: Whole Class

Class brainstorms a probability game with payouts and odds. Compute house expectation and variance collectively on board. Teams propose fairer variants, vote, and simulate 100 plays to verify calculations.

Construct the expectation and variance for a given probability distribution.

Facilitation Tip

For the Game Design Challenge, require students to write a short rationale linking their game’s scoring rules to its expected payout and variance, tying mechanics to mathematical outcomes.

What to look for

Give students a scenario, such as the number of customers arriving at a shop in an hour, with a defined probability distribution. Ask them to write one sentence explaining the meaning of the expected value and one sentence explaining what the variance tells us about customer arrival patterns.

• Apply
• Analyze
• Evaluate
• Create
• Relationship Skills
• Decision-Making
• Self-Management

Activity 04

• Collaborative Problem-Solving
• 30 min · Individual

Binomial Trials: Individual Logs

Each student flips a coin n times for binomial setup, repeats for multiple trials. Log observed means and variances, then derive theoretical using formulas. Reflect on sample size effects in journals.

Explain the meaning of the expected value in the context of a discrete random variable.

Facilitation Tip

In Binomial Trials, ask students to reflect in their logs on how the number of trials influences the gap between their sample mean and the theoretical expectation.

What to look for

Provide students with a probability distribution table for a discrete random variable (e.g., number of heads in 3 coin flips). Ask them to calculate E(X) and Var(X) on a mini-whiteboard and hold it up for review.

• Apply
• Analyze
• Evaluate
• Create
• Relationship Skills
• Decision-Making
• Self-Management

A few notes on teaching this unit

Teach this topic through layered experiences: start with physical simulations to build intuition, move to spreadsheets for precision, then apply concepts in open-ended design. Avoid rushing to formulas; instead, let students derive E(X²) from raw data first. Research shows that students grasp variance better when they calculate it both ways (deviation method and E(X²) - [E(X)]²) and compare results side by side.

By the end of these activities, students will calculate expectations and variances correctly, explain why variance uses squared deviations, and connect both measures to real contexts through simulations or designed scenarios. They will also recognize when empirical results align with or diverge from theory.

---

Watch Out for These Misconceptions

• During Simulation Rotations, watch for students who assume the most frequent outcome in their trials equals the expectation.

  During Simulation Rotations, have students graph their empirical frequencies and overlay the theoretical expectation as a vertical line, prompting them to compare the mode, median, and mean visually.

• During Spreadsheet Builds, watch for students who treat variance as the average of absolute deviations.

  During Spreadsheet Builds, ask students to calculate both the sum of absolute deviations and the squared deviations for the same data set, then discuss why squaring is necessary to match the formula Var(X) = E(X²) - [E(X)]².

• During Binomial Trials, watch for students who expect their sample mean from ten trials to exactly match the theoretical expectation.

  During Binomial Trials, collect class data on the whiteboard and have students plot all sample means to observe the distribution of results, reinforcing that convergence to E(X) requires many trials.

---

Methods used in this brief

• Collaborative Problem-Solving