Interquartile Range (IQR) Statistics Example 4

Follow the full solution, then compare it with the other examples linked below.

Example 4

hard
Daily temperatures (ยฐC) for two weeks: 15, 16, 14, 18, 20, 22, 19, 35, 17, 16, 18, 21, 20, 19. Find the IQR and use the 1.5ร—IQR1.5 \times IQR rule to identify any outliers. Then recalculate the mean with and without the outlier(s).

Solution

  1. 1
    Step 1: Sort: 14,15,16,16,17,18,18,19,19,20,20,21,22,35. 14 values. Q2=18+192=18.5Q_2 = \frac{18+19}{2}=18.5. Lower half: {14,15,16,16,17,18,18}, Q1=16Q_1=16. Upper half: {19,19,20,20,21,22,35}, Q3=20Q_3=20. IQR=20โˆ’16=4IQR=20-16=4.
  2. 2
    Step 2: Fences: Lower=16โˆ’6=1016-6=10, Upper=20+6=2620+6=26. The value 35 > 26, so 35 is an outlier. Mean with outlier: 27014โ‰ˆ19.3\frac{270}{14}\approx 19.3. Mean without outlier: 23513โ‰ˆ18.1\frac{235}{13}\approx 18.1. The outlier raises the mean by about 1.2ยฐC.

Answer

IQR=4IQR = 4. The value 35ยฐC is an outlier (above the upper fence of 26ยฐC). Mean with outlier โ‰ˆ 19.3ยฐC; mean without โ‰ˆ 18.1ยฐC.
The IQR-based outlier detection method is practical for identifying unusual values in real data. Removing outliers and comparing the mean before and after quantifies their influence. This helps decide whether outliers represent errors (remove) or genuine extreme observations (keep but note).

About Interquartile Range (IQR)

The interquartile range (IQR) is the range of the middle 50% of data, calculated as Q3โˆ’Q1Q_3 - Q_1. It measures spread while ignoring the top and bottom 25% of values, making it resistant to outliers.

Learn more about Interquartile Range (IQR) โ†’

More Interquartile Range (IQR) Examples