What can be inferred if kurtosis is less than 3?

When kurtosis is less than 3, it indicates that the data distribution is flatter than a normal distribution. This characteristic is often referred to as "platykurtic." In a platykurtic distribution, there are fewer outliers in the tails, suggesting that the data is more evenly distributed without the extreme values that would cause the tails to be heavier.

Understanding kurtosis helps in associating the shape of the distribution to its propensity for outliers; thus, a kurtosis value below 3 conveys that the distribution is light-tailed, meaning it has a lower likelihood of producing extreme outlier values compared to a normal distribution. This aspect is crucial in data analysis as it influences statistical testing and assumptions regarding the nature of the data.