Interpreting the Standard Deviation: Empirical Rule and Chebyshev’s Rule

Interpreting the Standard Deviation

Empirical Rule (Rule of Thumb)

The Empirical Rule provides a quick way to estimate the spread of data in a mound-shaped (or bell-shaped, i.e., normal) distribution using the mean and standard deviation. It is especially useful for understanding the proportion of data within certain intervals around the mean.

• Definition: The Empirical Rule states that for a normal distribution:

  • Approximately 68% of the data falls within one standard deviation of the mean: $(\bar{x} - s, \bar{x} + s)$

  • Approximately 95% of the data falls within two standard deviations of the mean: $(\bar{x} - 2s, \bar{x} + 2s)$

  • Approximately 99.7% of the data falls within three standard deviations of the mean: $(\bar{x} - 3s, \bar{x} + 3s)$

• Approximation for Standard Deviation: For mound-shaped data, a rough estimate for the standard deviation is: $s \approx \frac{\text{sample range}}{4}$

Example: Battery Life

Scenario: A manufacturer claims the average life of a battery is 60 months, with a standard deviation of 10 months. The distribution is mound-shaped.

• (i) Percentage lasting more than 50 months: 50 months is 1 standard deviation below the mean ($60 - 10 = 50$). By the Empirical Rule, about 84% of batteries last more than 50 months (since 68% are within one standard deviation, so 34% are between 50 and 60, and 50% are above 60; thus, 50% + 34% = 84%).

• (ii) Percentage lasting less than 40 months: 40 months is 2 standard deviations below the mean ($60 - 2 \times 10 = 40$). By the Empirical Rule, about 2.5% of batteries last less than 40 months (since 95% are within two standard deviations, so 2.5% are below $\bar{x} - 2s$).

• (iii) If a battery lasts only 37 months: 37 months is 2.3 standard deviations below the mean. This is quite rare in a normal distribution, suggesting the manufacturer's claim may be overstated.

Additional info: The percentages above are based on properties of the normal distribution. For exact probabilities, use the standard normal table.

Chebyshev’s Rule

Chebyshev’s Rule provides a minimum proportion of data within a certain number of standard deviations from the mean, regardless of the distribution's shape. It is more general than the Empirical Rule and applies to all distributions.

• Definition: For any dataset (not necessarily normal), at least $1 - \frac{1}{k^2}$ of the data falls within $k$ standard deviations of the mean, for any $k > 1$.

• Key Points:

  • Within 2 standard deviations: at least $1 - \frac{1}{2^2} = 0.75$ (75%) of the data is within $(\bar{x} - 2s, \bar{x} + 2s)$.

  • Within 3 standard deviations: at least $1 - \frac{1}{3^2} = 0.89$ (89%) of the data is within $(\bar{x} - 3s, \bar{x} + 3s)$.

  • Within $k$ standard deviations: at least $1 - \frac{1}{k^2}$ of the data is within $(\bar{x} - ks, \bar{x} + ks)$ for $k > 1$.

• Note: Chebyshev’s Rule gives a lower bound; the actual percentage may be higher.

Example: Traffic at an Intersection

Scenario: The mean number of vehicles per day is 375, with a standard deviation of 25. We want to know the percentage of days with more than 425 vehicles.

• (a) Unknown distribution shape: 425 is 2 standard deviations above the mean ($375 + 2 \times 25 = 425$). By Chebyshev’s Rule, at most 25% of the data can be outside two standard deviations (since at least 75% is within), so at most 12.5% can be above 425.

• (b) If the distribution is mound-shaped: By the Empirical Rule, about 2.5% of the data is above two standard deviations from the mean (i.e., above 425 vehicles).

Comparison Table: Empirical Rule vs. Chebyshev’s Rule

Rule	Distribution Requirement	Within 1 SD	Within 2 SD	Within 3 SD
Empirical Rule	Mound-shaped (normal)	68%	95%	99.7%
Chebyshev’s Rule	Any distribution	--	≥ 75%	≥ 89%

Additional info: The Empirical Rule provides more precise estimates but only applies to normal distributions. Chebyshev’s Rule is more general but less precise.