# Three options to describe the characteristics of data.

Many of you may use “average” to describe the characteristics of a single set of data.

As a result, you may compare two or more averages and conclude which is the highest or which is the lowest.

But it is a too much simplistic way to show the characteristics of the data.

I strongly suggest you that you should now graduate from such elementary approach to handle data.

In my program/class, I propose to have three options to show the data characteristics.

(1) Amount/ratio — many of you may already use these!

(2) Trend

(3) Distribution/variance

According to your goal (i.e. what you would like to know eventually), you may choose the right combination(s) of the three to show what information the data has.

This is an exactly “multiple-angle” approach, rather than simplistic approach.

Those three elements supplement one another with its own characteristic. For example, “average” does not show “trend” nor “distribution” vice versa.

If you want to illustrate/visualize your findings from a single data set, you may show these in a variety of graphs to fit your conclusion/message:

This is the most common example that many people frequently use:

Also some people use “pie chart” to show the ratio and “line chart” to show trend, which you can never know only with “Average” or ” total” charts.

I know that only a few people are aware of its importance of data distribution/ varience as it shows risk(s) especially in a business case.

Finally, if you can find a right combination of the three elements, then you may be able to enrich the features you can get from the data as the following examples:

(CV:Coefficient of variance — % of variance against the average )

This is a very fundamental knowledge to handle any kinds of data, which is required even for more advanced data analysis techniques. But not many people acquire the skills yet.

You may learn how to “see” data and to apply it to practical issues in my programs.

Please let me know if you are interested!