Computing · Year 6

Active learning ideas

Identifying Outliers and Anomalies

Active learning works for this topic because students need to see data as more than numbers on a page. When they physically sort, plot, and debate data points, they build a feel for what ‘normal’ looks like and why some points stand apart. This hands-on work turns abstract rules into lived experience.

• National Curriculum Attainment Targets
• KS2: Computing - Data Handling
• KS2: Computing - Computational Thinking

• 25–45 min
• Pairs → Whole Class
• 4 activities

Activity 01

• Case Study Analysis
• 30 min · Pairs

Spreadsheet Hunt: Weather Data Outliers

Provide datasets of daily temperatures. Students sort data in spreadsheets, calculate averages, and highlight points more than 1.5 times the range from the mean. Pairs discuss and mark potential outliers with colours, then share findings.

Explain how to identify an outlier in a given dataset.

Facilitation Tip

During Spreadsheet Hunt, circulate and ask each pair to explain why they marked a certain day as an outlier so you can catch any ‘highest or lowest only’ reasoning early.

What to look for

Present students with a small spreadsheet of data, for example, daily rainfall amounts for a month. Ask them to identify any data points that seem unusually high or low and write down their reasons for choosing them.

• Analyze
• Evaluate
• Create
• Decision-Making
• Self-Management

Activity 02

• Case Study Analysis
• 45 min · Small Groups

Group Debate: Real-World Anomalies

Distribute printed datasets on sports scores or sales. Small groups identify outliers, brainstorm causes like equipment failure or promotions, and vote on inclusion. Present decisions to the class with evidence from graphs.

Assess the reasons why an outlier might occur in real-world data.

Facilitation Tip

In Group Debate, appoint one student to play devil’s advocate to ensure multiple causes are weighed before groups vote on which explanations are most plausible.

What to look for

Provide a scenario: 'A student's test score is much lower than all their other scores. What are three possible reasons for this outlier? Should this score be included when calculating the average class score? Why or why not?'

• Analyze
• Evaluate
• Create
• Decision-Making
• Self-Management

Activity 03

• Case Study Analysis
• 25 min · Individual

Individual Creation: Plant a Fake Outlier

Students enter their own class-generated data, such as step counts, into spreadsheets. They deliberately add one outlier, then swap with a partner to detect and explain it using box plots.

Justify whether an outlier should be included or excluded from a data analysis.

Facilitation Tip

For Individual Creation, remind students to keep the rest of the dataset tightly clustered so peers can clearly see the planted fake outlier’s effect.

What to look for

Give students a simple dataset, e.g., ages of people at a party. Ask them to calculate the range and identify any potential outliers. Then, ask them to write one sentence explaining why an outlier might occur in this specific context.

• Analyze
• Evaluate
• Create
• Decision-Making
• Self-Management

Activity 04

• Case Study Analysis
• 35 min · Whole Class

Whole Class Simulation: Sensor Faults

Project a live-updating spreadsheet of simulated sensor data. Class calls out anomalies as they appear, predicts causes, and tests exclusion effects on summary stats in real time.

Explain how to identify an outlier in a given dataset.

Facilitation Tip

During Whole Class Simulation, freeze the simulation at key points and ask students to predict what the sensor error will do to the next reading before it appears.

What to look for

Present students with a small spreadsheet of data, for example, daily rainfall amounts for a month. Ask them to identify any data points that seem unusually high or low and write down their reasons for choosing them.

• Analyze
• Evaluate
• Create
• Decision-Making
• Self-Management

A few notes on teaching this unit

Start with visual methods—scatter plots and box plots—before any formulas appear. Students first trust their eyes, then their calculations. Avoid rushing to the mean or standard deviation; these can hide subtle anomalies that simple plotting reveals. Research shows that early visual checks build the intuition students need before formal tests make sense.

Successful learning looks like students confidently pointing to a data point and explaining why it stands out, not just stating it is an outlier. They should connect the visual or calculated evidence to real-world causes and defend whether the point changes the story the data tells.

---

Watch Out for These Misconceptions

• During Spreadsheet Hunt, watch for students who only flag the highest or lowest values immediately.

  Ask them to circle the cluster of ‘normal’ temperatures first, then explain why any point outside that circle stands out, regardless of its rank.

• During Group Debate, watch for students who treat all outliers as mistakes to be removed.

  Remind groups to list at least two possible causes—one error-based and one real-world event—before deciding whether removal is justified.

• During Individual Creation, watch for students who create obvious outliers far from the rest of the data.

  Encourage them to make the fake outlier closer to the cluster so peers see how subtle changes can still skew results.

---

Methods used in this brief

• Case Study Analysis