Upper Fence Calculator — Outlier Detection
Enter Q1 and Q3 from your data set (and optionally the k multiplier) to find the upper Tukey fence — the cutoff above which values are flagged as potential outliers.
Upper Fence = Q3 + k × IQR — values above this are potential outliers
- 1
IQR
Q3 − Q1 = 75 − 25 = 50 - 2
k × IQR
1.5 × 50 = 75 - 3
Upper Fence
Q3 + k × IQR = 75 + 75 = 150Values above this threshold are potential outliers.
How does this calculator work?
Upper Fence = Q3 + k × IQR, where IQR = Q3 − Q1 and k = 1.5 (mild outliers, standard box-plot whisker) or k = 3 (extreme outliers). Any observation above this threshold is a potential outlier to investigate.
Formula
How this is calculated
The upper fence is a threshold used in Tukey's outlier-detection rule, which is the statistical basis of the "whisker" in a box-and-whisker (box plot) chart. It is constructed from the interquartile range: IQR = Q3 − Q1. The upper fence equals Q3 + k × IQR, where k is the fence multiplier. The conventional choices are k = 1.5 for mild outliers (the standard box-plot whisker end) and k = 3 for extreme outliers.
Any data point whose value exceeds the upper fence is labelled a potential outlier. On a box plot, the whisker extends to the largest observed value that is still ≤ the upper fence; observations beyond the fence are plotted as individual dots. The fence itself is not plotted, but it determines where the whisker stops.
This rule is distribution-free: it does not assume normality, relying only on the quartiles. Under a normal distribution, the mild fence (k = 1.5) excludes roughly 0.7% of values on each tail, while the extreme fence (k = 3) excludes fewer than 0.01%. For skewed or heavy-tailed distributions the proportion of flagged values will differ, so flagged points should always be investigated rather than automatically removed.
Frequently asked questions
A value beyond the upper fence is flagged as unusual relative to the bulk of the data, but not automatically erroneous. Investigate first: is it a data-entry mistake, a measurement error, or a genuine extreme but valid observation? Only remove an outlier if you have a substantive reason — trimming valid extremes biases your estimates.
The mild fence uses k = 1.5 (Q3 + 1.5 × IQR) and is the default for box plots — it catches moderately unusual values. The extreme fence uses k = 3 (Q3 + 3 × IQR) and only flags values that are very far from the central mass. Anything between the two fences is a "mild outlier"; anything beyond the extreme fence is an "extreme outlier".
The lower fence formula is Q1 − k × IQR — simply mirror the upper fence formula around Q1 instead of Q3. Use the upper-lower-fence calculator to compute both fences simultaneously.
Also known as
TG we-Calculate Editorial Team. (2026). Upper Fence Calculator — Outlier Detection [Online calculator]. TG we-Calculate. https://we-calculate.com/calculator/upper-fence-calculator
TG we-Calculate Editorial Team. "Upper Fence Calculator — Outlier Detection." TG we-Calculate. 2026. https://we-calculate.com/calculator/upper-fence-calculator.
TG we-Calculate Editorial Team, "Upper Fence Calculator — Outlier Detection," TG we-Calculate, 2026. [Online]. Available: https://we-calculate.com/calculator/upper-fence-calculator
@misc{wecalculate_upper_fence_calculator, title = {Upper Fence Calculator — Outlier Detection}, author = {{TG we-Calculate Editorial Team}}, howpublished = {\url{https://we-calculate.com/calculator/upper-fence-calculator}}, year = {2026}, note = {TG we-Calculate} }
Did this calculator help you?