An outlier can move the arithmetic mean substantially because the mean includes the full size of every value. A very high outlier pulls the mean upward, while a very low outlier pulls it downward. The median is usually affected less because it depends mainly on the ordered middle position.
Enter a list of numbers to calculate the mean, sum, count, median, mode, range, minimum, and maximum.
Compare a balanced set with the same data after one ordinary value is replaced by an extreme value.
What Is an Outlier?
An outlier is an observation that sits unusually far from most of the other values.
It may be genuinely unusual, caused by measurement error, or created by incorrect data entry.
Use the Average Calculator to see how a suspected outlier changes the mean, median, and range.
Why Outliers Affect the Mean
The arithmetic mean includes the full numerical size of every observation.
An extreme value can add or subtract a large amount from the total while increasing the count by only one.
The resulting mean moves toward the outlier.
Worked Example without an Outlier
Consider 10, 11, 12, 13, and 14.
Their sum is 60 and their mean is 12.
The median is also 12, and the range is four.
Worked Example with a High Outlier
Replace 14 with 100. The new sum is 146.
Dividing 146 by five gives a mean of 29.2.
The median remains 12, while the range increases from four to 90.
How a Low Outlier Changes the Mean
A very small value pulls the mean downward.
For minus 50, 10, 11, 12, and 13, the sum is minus four and the mean is minus 0.8.
The median remains 11 because the central ordered position changes much less.
Why the Median Is More Resistant
The median depends on position rather than the full magnitude of each observation.
Moving the largest value from 14 to 100 does not change which observation is in the middle.
This resistance makes the median useful for some skewed data.
How Outliers Affect the Range
The range is the maximum minus the minimum.
Because an outlier often becomes the new maximum or minimum, it can increase the range dramatically.
The range is therefore even more directly sensitive to extreme endpoints than the mean.
Should an Outlier Be Removed?
An outlier should not be removed merely because it changes the average.
First investigate whether it is a valid observation, an entry error, a measurement problem, or evidence of a separate group.
Removing valid data without a defensible reason can distort the analysis.
Compare Results with and without the Outlier
A useful approach is to calculate the mean both with and without the suspected outlier.
Report the median as well, and explain why the two summaries differ.
This makes the effect transparent instead of hiding the unusual observation.
Outliers in Everyday Data
A small number of very expensive houses can raise the mean house price above what most buyers encounter.
A few exceptionally high salaries can raise a company's mean salary.
Extreme completion times can also affect the mean waiting time for a service.
Mean vs Median When Outliers Exist
The mean describes equal sharing of the complete total, including the extreme observation.
The median describes the middle ordered position and may better reflect a typical observation in skewed data.
Read Mean vs Median: What Is the Difference? for a detailed comparison.
Common Mistakes
Do not assume that every unusual value is an error.
Do not quietly remove an outlier without recording the reason and recalculating the result.
Do not rely on the mean alone when the distribution is strongly skewed.
Conclusion
An outlier can pull the arithmetic mean upward or downward because the mean uses the size of every value.
The median is usually less sensitive, while the range may change dramatically.
Use the Average Calculator to compare the results before and after including a suspected outlier.
FAQs
What is an outlier?
It is an observation that lies unusually far from most of the other values.
Do high outliers raise the average?
Yes. A sufficiently high observation pulls the arithmetic mean upward.
Are medians affected by outliers?
They can be affected, but usually less strongly than arithmetic means.
Should I always remove an outlier?
No. Investigate whether it is valid before deciding how to handle it.
How can I show the outlier's effect?
Calculate and report results both with and without the suspected outlier.