What is an outlier?
Big Idea: An outlier is a data value that is much larger or much smaller than the other values in the dataset.
Unusual or extreme.
| Example data | Outlier? | Why? |
|---|---|---|
| 85, 87, 89, 91, 93 | No | All close together |
| 85, 87, 89, 91, 150 | Yes | 150 much larger |
| 2, 3, 4, 5, 6 | No | All consecutive |
| 2, 3, 4, 5, −50 | Yes | −50 much smaller |
Free preview
This is the free notes preview
You're reading the free notes. Aimnova Pro unlocks the full study experience — and you can try it with your first topic free to keep:
- FlashcardsLock in vocabulary and key terms with spaced repetition.
- Practice questionsAnswer exam-style questions and get instant AI marking.
- Mock exams & past-paper vaultSit full mocks and see exactly how examiners award marks.
- Personalised study planA daily plan built around your exam date and weak areas.
Causes of outliers
Legitimate: Real extreme values: World records, rare events.
Keep.
Error: Measurement mistakes, wrong units.
Investigate and consider removing.
Key point: Never blindly remove outliers.
Investigate first.
Worked example — investigate before removing
A class of 25 students records their daily screen-time hours: most values are between 2 and 6, but one student records 50 hours.
Is this an outlier, and what should you do?
Step by step
- Identify: 50 is far above the other values (2-6), so it IS an outlier numerically.
- Investigate the cause. A real day has only 24 hours, so 50 is impossible — this is a measurement / data-entry ERROR, not a true extreme value.
- Decide: remove the impossible value (or correct it if the true entry can be recovered). Do NOT silently drop true extreme values.
- Report: state in your write-up that the value 50 was excluded as an obvious data-entry error (>24 hours).
Final answer
Yes, 50 is an outlier. It is a data-entry error (impossible value) and should be removed or corrected, with a note in the report explaining the exclusion.
Study smarter, not longer
Most students waste 40% of study time on topics they already know. Our AI tracks your progress and optimizes every minute.
IQR method for identifying outliers
Worked example
Data: 12, 14, 15, 16, 18, 19, 20, 21, 25, 45 Identify outliers.
Solution
- Q1 = 15, Q3 = 21 (the median of each half: 12, 14, 15, 16, 18 and 19, 20, 21, 25, 45), so IQR = 6
- Lower bound = 15 − 1.5(6) = 6
- Upper bound = 21 + 1.5(6) = 30
- 45 > 30, so 45 is an outlier
Final answer
Outlier: 45
IB-style question — largest non-outlier
A data set has Q1 = 20 and Q3 = 32.
Find the largest value that would NOT be classed as an outlier.
Step by step
- IQR first, then the upper fence Q3 + 1.5·IQR — anything at or below it is not an outlier.
Final answer
50 — any value above 50 is an outlier.
With Q1 = 20, Q3 = 32 (IQR = 12), the upper outlier fence is Q3 + 1.5·IQR = 50 — the largest value that isn't an outlier.
Interactive diagram
Explore the labelled diagram, charts and maps for this topic in full study mode.
How outliers affect statistics
| Statistic | Outlier effect |
|---|---|
| Mean | Highly affected |
| Median | Resistant |
| Range | Highly affected |
| IQR | Resistant |
| SD | Highly affected |
Strategy: With outliers: Use median and IQR. Without: Can use mean and SD.
Practice with exam-style questions
Answer exam-style questions and get AI feedback that shows you exactly what examiners want to see in a full-marks response.
Three or four marks, three or four lines: Outlier questions are marked line by line: the fence with its substitution, its value, the comparison, and the conclusion. A correct verdict with no fence earns almost nothing.
See it drawn, step by step
In My Learning, handwritten diagrams build one step at a time. Play, pause and replay each explanation until it clicks. Create your free account to try it.
Write all of these
- The fence formula with the numbers in it.
- The value of the fence.
- The comparison with < or >.
- The conclusion: "is / is not an outlier" (and "so the claim is correct" if someone made a claim).
Valid or error?: After finding an outlier, say whether it is a real value (a very fast swimmer) or a likely mistake (a height of 250 cm). Real outliers stay in the data.
Feeling unprepared for exams?
Get a clear study plan, practice with real questions, and know exactly where you stand before exam day. No more guessing.
Exam-style question — delivery times [6 marks]
The times, in minutes, of 12 food deliveries are:
52, 61, 64, 66, 67, 70, 72, 73, 75, 78, 80, 104
(a) Find the lower quartile and the upper quartile. [2]
(b) The manager believes that the delivery of 104 minutes is an outlier. Determine whether the manager is correct. [3]
(c) The fastest delivery took 52 minutes. State whether it is an outlier. [1]
Step by step
- (a) GDC 1-Var Stats (the median of each half of six values).
- (b) The upper fence, then compare.
- (b) Conclude.
- (c) The lower fence.
Final answer
(a) Q₁ = 65, Q₃ = 76.5 (b) fence 93.75; 104 > 93.75, so the manager is correct (c) no (52 > 47.75)