Upper and Lower Fence Calculator
Calculate upper and lower fence values to identify outliers in your dataset. Uses Tukey's method with IQR to detect mild and extreme outliers for robust statistical analysis.
Enter up to 50 numeric values separated by commas
What are upper and lower fences in statistics?
Fences are cutoff points for identifying outliers. Lower Fence = Q1 - 1.5×IQR identifies low outliers. Upper Fence = Q3 + 1.5×IQR identifies high outliers. Values outside these fences are considered potential outliers that may warrant further investigation.
How are fence values calculated?
Calculate: (1) Q1 and Q3 (first and third quartiles), (2) IQR = Q3 - Q1, (3) Lower Fence = Q1 - 1.5×IQR, (4) Upper Fence = Q3 + 1.5×IQR. The 1.5 multiplier is Tukey's rule, balancing sensitivity and specificity for outlier detection.
What is the difference between mild and extreme outliers?
Mild outliers: Between inner fences (1.5×IQR) and outer fences (3×IQR). Extreme outliers: Beyond outer fences (3×IQR). Extreme outliers are more unusual and often indicate data errors or truly exceptional cases. In box plots, mild outliers shown as circles, extreme as asterisks.
Should I always remove outliers identified by fences?
No! Investigate first. Outliers may be: (1) Data errors (measurement/recording mistakes) - remove these, (2) Valid extreme values (important findings) - keep these, (3) Special cases worth separate analysis. Never automatically delete outliers without understanding why they occurred.
Why use 1.5×IQR instead of other multipliers?
Tukey chose 1.5×IQR as a balance: Smaller multipliers (e.g., 1×IQR) flag too many points as outliers. Larger multipliers (e.g., 2×IQR) miss real outliers. 1.5×IQR works well for normally distributed data and is the standard in box plots and exploratory data analysis.