Outlier Calculator
Find outliers using the 1.5 IQR rule. Enter your data set and the calculator sorts it, finds Q1, Q3, and the interquartile range, then works out the lower and upper outlier boundaries. Any value outside those boundaries is an outlier.
Enter any data set and this outlier calculator flags the values that sit too far from the rest. It sorts your numbers, finds Q1 and Q3, works out the interquartile range, then applies the 1.5 IQR rule to give you the lower and upper boundaries. Anything past those boundaries is an outlier. You also get the full step-by-step math, so it works whether you are checking homework or cleaning a real data set.
What counts as an outlier?
An outlier is a value that sits unusually far from the middle of your data. In a set like 21, 23, 25, 29, 30, 35, 60, the number 60 is the outlier because it is far above where the rest of the numbers cluster.
Outliers matter because they pull the mean and standard deviation off course. One extreme value can drag an average up or down and make a data set look different from what it really is. That is why finding them is usually the first step before you trust any summary statistic.
Outliers show up for a few reasons: a typing mistake during data entry, a measurement error, or a genuine rare event that really did happen. The calculator finds them either way. Deciding what to do with them is your call.
How does the 1.5 IQR rule find outliers?
The 1.5 IQR rule is the standard method for spotting outliers, and it is what this calculator uses. It builds a “normal” range from the middle half of your data, then flags anything outside it.
Here is the full process:
- Sort the data from smallest to largest.
- Find the first quartile (Q1) and third quartile (Q3).
- Subtract to get the interquartile range: IQR = Q3 − Q1.
- Multiply the IQR by 1.5.
- Set the lower boundary at Q1 − (1.5 × IQR) and the upper boundary at Q3 + (1.5 × IQR).
- Any value below the lower boundary or above the upper boundary is an outlier.
The IQR covers the middle 50% of your numbers, so it ignores the extremes when it builds the range. That is what makes it more reliable than using the mean, which the outliers themselves would distort.
What are the outlier boundary formulas?
Two formulas do the work. The lower one catches values that are too small, the upper one catches values that are too large.
| Boundary | Formula | What it flags |
|---|---|---|
| Lower outlier boundary | Q1 − (1.5 × IQR) | Any value below this is a low outlier |
| Upper outlier boundary | Q3 + (1.5 × IQR) | Any value above this is a high outlier |
Say Q1 is 23 and Q3 is 35, so the IQR is 12. Multiply by 1.5 to get 18. The lower boundary is 23 − 18 = 5, and the upper boundary is 35 + 18 = 53. In that data set, any number under 5 or over 53 is an outlier.
How to find an outlier by hand
Find Outliers in a Data set manually, you can do the whole thing on paper. Take the data set 23, 60, 30, 21, 25, 35, 29 and follow the steps.
Step 1: Sort the data
21, 23, 25, 29, 30, 35, 60 (seven values)
Step 2: Find Q1 and Q3
The median is the middle value, 29. The lower half is 21, 23, 25, so Q1 = 23. The upper half is 30, 35, 60, so Q3 = 35.
Step 3: Find the IQR
IQR = Q3 − Q1 = 35 − 23 = 12
Step 4: Build the boundaries
Lower = 23 − (1.5 × 12) = 23 − 18 = 5
Upper = 35 + (1.5 × 12) = 35 + 18 = 53
Step 5: Check every value
Everything from 21 to 35 sits inside 5 to 53. Only 60 falls above the upper boundary, so 60 is the outlier.
Doing it by hand is fine for small sets. For anything larger, the calculator above removes the arithmetic slips that creep in when you sort and split quartiles manually. If you want the quartiles broken out on their own, the 5 number summary calculator gives you the minimum, Q1, median, Q3, and maximum in one view.
Why do some calculators disagree on outliers?
This trips up a lot of students, so it is worth knowing. Two calculators can look at the same data and report different outliers. Neither one is broken. They just use different rules to find Q1 and Q3.
Take the set 8, 9, 2, 5, 46, 9, 50. Sorted, that is 2, 5, 8, 9, 9, 46, 50. There are two common ways to split the quartiles:
| Method | How it splits | Q1 | Q3 | Outliers found |
|---|---|---|---|---|
| Inclusive (TI-84, Moore) | Keeps the median in both halves | 5 | 46 | None |
| Exclusive (Tukey) | Splits around the median | 5 | 46 | None |
With this particular set the two methods land in the same place, but on many data sets they do not. The gap widens when the count of values is small or when the numbers are skewed. If your textbook answer key says one thing and an online tool says another, check which quartile method each one uses before assuming a mistake was made. This calculator uses the inclusive method, the same one the TI-84 and most US stats courses teach.
If you want to see exactly how the quartiles are being cut for your own numbers, the step-by-step output above shows the lower half and upper half it used.
How do you find outliers in Excel?
Excel has no single outlier button, so you build the 1.5 IQR rule from its quartile functions.
- Put your data in a column, say A1 to A20.
- Q1:
=QUARTILE(A1:A20,1) - Q3:
=QUARTILE(A1:A20,3) - IQR: subtract Q1 from Q3 in a new cell.
- Lower boundary:
=Q1cell-1.5*IQRcell - Upper boundary:
=Q3cell+1.5*IQRcell - Flag outliers with
=IF(OR(A1<lower,A1>upper),"Outlier","OK")dragged down the column.
One thing to note: QUARTILE in Excel uses the inclusive method, so it matches this calculator. If your class uses the exclusive method, use QUARTILE.EXC instead and your boundaries may shift.
What should you do with an outlier once you find it?
Finding the outlier is the easy part. Deciding its fate takes judgment.
If the value came from a typo or a broken sensor, correcting or removing it is reasonable. If it is a real observation, a genuinely tall person in a height study, for example, throwing it out hides something true about your data. Removing outliers just to make the numbers look tidy is how analysis goes wrong.
A safer habit is to run your summary both ways, once with the outlier and once without, and see how much it moves the result. If the standard deviation barely changes, the outlier is not doing much harm. If it swings wildly, that value is driving your conclusions and deserves a closer look before you report anything.