I have a set of sample data of ages of customers of an arbitrary business. I need to filter and display this data in the most meaningful way possible, to the business.
I already am graphing the data and overlaying normal distribution and standard deviation, but what other information should be parsed out?
It seems that it would be useful to collect any major groups of outliers. For example, I've included an annotated graph below that represents a business that serves alcohol in a college town. The data pool is too small to identify trends by seeing a lack or excess of customers of a specific age. However, I can easily identify that there is a heavy grouping of statistical outliers in the 21-26 age range. How would I go about isolating and identifying such outlier groupings, mathematically?