Exposure data in many occupational settings are best described by which distribution?

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

Answer: (A)

Exposure measurements in workplaces are typically right-skewed and bound at zero, with many observations at lower values and a long tail of higher concentrations. This pattern arises because the factors that determine exposure—emission rate, duration, ventilation, work practices, and controls—tend to interact multiplicatively rather than additively. When you take the logarithm of the data, those multiplicative effects become additive, and the log-transformed values often resemble a normal distribution. This makes the lognormal model a natural fit for occupational exposure data.

Using a lognormal distribution also helps with interpretation and statistics. Central tendency and spread are described with the geometric mean and geometric standard deviation, and percentiles of interest (like the higher-exposure tails used in risk assessments) are more stable and meaningful on the log scale. In contrast, a normal distribution assumes symmetry around a mean and can imply negative concentrations, a uniform distribution treats all values as equally likely across a range, and a binomial distribution applies to counts of two outcomes rather than continuous concentration data. Those options don’t capture the typical skew and nonnegative nature of real-world exposure data as well as the lognormal model does.