Why are exposure data often analyzed as lognormal distributions?

• A. Exposures are additive and left-skewed.
• B. Exposures are multiplicative and right-skewed.
• C. Data are always normally distributed.
• D. To simplify calculations using arithmetic mean.

Answer: (B)

Exposures often arise from many factors that multiply together—like emission rate, duration, breathing rate, and transfer efficiencies. When you multiply several positive factors, you tend to get many small values with a long tail toward larger ones, which is a right-skewed pattern. Taking logs converts multiplication into addition, and the sum of many independent log-values tends to be approximately normal. So the log of exposure is near normally distributed, making the original data well described by a lognormal distribution. This approach also supports using geometric means and regression on the log scale, with results interpretable back on the original scale as multipliers. The other options don’t fit because exposures are not typically left-skewed or normally distributed in the original scale, and simply using the arithmetic mean isn’t the reason lognormal modeling is used.