What is an outlier in math?
An outlier is a data value that is far away from most of the other values in a set. In 4, 5, 6, 7, 50, the value 50 is an outlier. A common test flags any value more than 1.5 times the interquartile range below the first quartile or above the third. Outliers pull the mean strongly.
Outlier in a picture
Formal definition and notation
Informally, an outlier is an observation that differs markedly from the rest of the data. The standard 1.5 × IQR rule calls a value x a potential outlier if x < Q1 − 1.5 × IQR or x > Q3 + 1.5 × IQR, where Q1 and Q3 are the first and third quartiles and IQR = Q3 − Q1. In a scatter plot, an outlier is a point far from the overall pattern. An outlier may be a recording error, a rare but real value, or a sign of a different group, so it calls for investigation.
| Symbol | Read it as | Meaning |
|---|---|---|
| Q1, Q3 | Q one, Q three | the first and third quartiles: the medians of the lower and upper halves of the ordered data |
| IQR | interquartile range | Q3 − Q1, the spread of the middle half of the data |
| 1.5 × IQR | one and a half I Q R | how far past a quartile a value must be to count as an outlier |
Key facts
- Outliers move the mean and the range a lot but barely move the median: for 4, 5, 6, 7, 50 the mean is 14.4 while the median is 6.
- On a box plot, outliers are drawn as separate dots beyond the whiskers, and the whiskers stop at the last value inside the fences.
- Outliers can be high or low: in 2, 29, 30, 31, 32, 33 the outlier is 2.
- A data set can have no outliers, one, or several.
- In a scatter plot, a point far from the trend of the others is an outlier, even if its x- and y-values are each ordinary on their own.
The 1.5 × IQR rule (outlier fences)
- Q1
- first quartile (median of the lower half)
- Q3
- third quartile (median of the upper half)
- IQR
- Q3 − Q1
Any value below the lower fence or above the upper fence is an outlier. Values exactly on a fence are not. Some books find quartiles slightly differently, which can move the fences a little.
Try it: add an outlier
Press Add an outlier, type your own extreme value, or drag a dot far to the right. Values outside the fences turn orange; compare how far the mean moves with the median.
Outlier examples, step by step
Example 1: A low outlier
Which value is an outlier: 31, 29, 33, 30, 2, 32?
- Order the data: 2, 29, 30, 31, 32, 33.
- Five of the values are between 29 and 33. The value 2 is far below them.
- Check with fences: Q1 = 29, Q3 = 32, IQR = 3, and the lower fence is 29 − 4.5 = 24.5.
- 2 is below 24.5.
Answer: The outlier is 2, a low value far below the rest.
Example 2: Use the 1.5 × IQR rule
Use the 1.5 × IQR rule to test for outliers in 10, 12, 13, 13, 14, 15, 16, 26.
- The data are in order and n = 8. Lower half: 10, 12, 13, 13, so Q1 = 12.5. Upper half: 14, 15, 16, 26, so Q3 = 15.5.
- IQR = 15.5 − 12.5 = 3, and 1.5 × 3 = 4.5.
- Lower fence: 12.5 − 4.5 = 8. Upper fence: 15.5 + 4.5 = 20.
- Nothing is below 8. Only 26 is above 20.
Answer: The fences are 8 and 20, so 26 is an outlier and no other value is.
Example 3: How an outlier moves the mean and median
Five test scores are 82, 85, 88, 90 and 35. The 35 is an outlier. If it is left out, which changes more, the mean or the median?
- With 35: the mean is 3805 = 76; ordered, 35, 82, 85, 88, 90 has median 85.
- Without 35: the mean is 3454 = 86.25; 82, 85, 88, 90 has median 85 + 882 = 86.5.
- The mean changes by 86.25 − 76 = 10.25. The median changes by 86.5 − 85 = 1.5.
Answer: The mean changes more (by 10.25 points, against 1.5 for the median).
Common mistakes
Do not remove an outlier unless there is a valid reason, such as a recording error.
Every data set has a largest and a smallest value. An outlier must be far from the rest, not just at the end.
Not to be confused with
The range is greatest minus least, so it always includes any outlier: for 4, 5, 6, 7, 50 the range is 46. An outlier is one value; the range is a measure of spread.
Every data set has a maximum, its largest value. The maximum is an outlier only when it is far from the rest, as 22 is in 4, 5, 6, 7, 7, 8, 9, 22.
Quick check: 3 questions
Which value is most likely an outlier in 52, 55, 49, 51, 90, 53?
Answer: C. Ordered: 49, 51, 52, 53, 55, 90. Q1 = 51, Q3 = 55, IQR = 4, so the upper fence is 55 + 6 = 61, and only 90 is beyond it. 49 and 55 are just the smallest and second-largest values.Show answer and explanation
A data set has Q1 = 20 and Q3 = 30. Which value is an outlier?
Answer: B. IQR = 10, so the fences are 20 − 15 = 5 and 30 + 15 = 45. Only 46 is outside. 44 and 6 are close to the fences but still inside, and 35 is an ordinary value.Show answer and explanation
For 3, 4, 4, 5, 5, 6, 30, which measure of center changes most if the 30 is removed?
Answer: A. The mean drops from about 8.14 to 4.5. The median only moves from 5 to 4.5, and the modes (4 and 5) stay the same. Outliers pull the mean far more than the median or the mode.Show answer and explanation
Outlier: frequently asked questions
How do you find an outlier?
First look: a value far from the cluster on a dot plot or box plot is a candidate. To decide, use the 1.5 × IQR rule. Find Q1 and Q3, subtract to get the IQR, multiply it by 1.5, then subtract that from Q1 and add it to Q3. Any value outside those two fences is an outlier.
Should you remove outliers?
Only with a reason. If an outlier comes from a typing or measuring error, correct it or remove it and say so. If it is a real value, keep it, because it is part of what happened. You can report results both with and without it, or use the median and IQR, which outliers barely affect, to describe the data.
Can a data set have more than one outlier?
Yes. Any value beyond either fence counts, so a set can have several high outliers, several low ones, or both. It can also have none. In 1, 20, 21, 22, 23, 24, 60 both 1 and 60 are outliers, while in 3, 4, 5, 6, 7 there are none.
Practice and tools
Sources
- Common Core State Standards, Grade 6 Statistics and Probability (6.SP.B.5.c)
- Common Core State Standards, Grade 8 Statistics and Probability (8.SP.A.1)
- OpenStax, Introductory Statistics 2e, 2.3 Measures of the Location of the Data (outliers, 1.5 × IQR)
Updated November 2026·Reviewed by the Effortless Math team·Standards: CCSS 6.SP.B.5.c, CCSS 8.SP.A.1
Cite this entry
Effortless Math Team. “Outlier.” Effortless Math Dictionary. Updated November 2026. https://www.effortlessmath.com/math-dictionary/outlier/
Flashcard
Browse all terms in the Math Dictionary · More statistics terms · Random term