RELS 309 Lecture Notes - Lecture 14: Quartile, Papyrus 75, Interquartile Range
Document Summary
A measure of dispersion indicates how varied the observations are in a collection. A small value indicates that the observations are not too different from each other, while a large value indicates that the observations are very different from each other. Meanwhile, a value of zero indicates the absence of variation in the observations. There are two general classifications of the measures of dispersion. Measures of absolute dispersion have the same unit as the unit of the observations, while measures of relative dispersion have no unit and can be used in comparing the variability of one distribution with another distribution. It is the simplest and easiest-to-use measure of dispersion. However, it can be misleading when there are outliers as it only uses the extreme values. The population variance, denoted by 2, for a finite population with n elements is: The sample variance, denoted by s2, for a sample of size n is: These may be used for ease of computation: