The 3-sigma rule is applied to identify outliers in which type of distribution?

• A. Uniform distribution
• B. Normal distribution
• C. Binomial distribution
• D. Exponential distribution

Answer: (B)

The idea here is that the 3-sigma rule relies on the bell-shaped, symmetrical normal distribution, where distances from the mean measured in standard deviation units behave in a predictable way. In a normal distribution, about 68% of the data lie within one standard deviation of the mean, about 95% within two, and about 99.7% within three. So values that fall more than three standard deviations away from the mean are extremely rare and are typically labeled as outliers.

Why this matters: the standard deviation is a meaningful measure of spread only when the data are roughly symmetric around a center and the distribution follows the normal shape. That’s why the rule is associated with identifying unusual observations in normally distributed data.

Why the other distributions aren’t a good fit for this rule: a uniform distribution has a finite, flat range with all values equally likely inside the bounds, so there isn’t a meaningful tail beyond the mean to flag as outliers using a 3-sigma criterion. The binomial distribution is discrete and its shape depends on the parameters; it isn’t guaranteed to be symmetric or to have its tail probabilities aligned with a fixed 3-sigma threshold in the same way. The exponential distribution is skewed, with a long tail to the right, so the distance from the mean in standard deviation units doesn’t translate to the same probability cutoff as in a normal distribution.

In short, values beyond mu plus or minus three times sigma are treated as outliers specifically because the normal distribution’s symmetry and standard deviation-based spread produce that well-known 99.7% interval.