There are three main kinds of averages – the mean, the median, and the mode. These are collectively known as the measures of the central tendency of the data. Given a particular data set it might be difficult to decide which average should be used, In this article, we explain how to make the choice of average depending upon the data.

**How to Choose an appropriate Average?**

It is obvious that no single average is suitable for all practical problems. Each of the

averages has its own merits and demerits and consequently its own field of importance and utility. For example, arithmetic mean is not to be recommended while dealing with frequency distribution with extreme observations or open end classes.

Median and mode are the averages to be used while dealing with open-end classes. In the case of qualitative data which cannot be measured quantitatively (e.g., for finding average intelligence, honesty, beauty, etc.), the median is the only average to be used.

Mode is particularly used in business and geometric mean is to be used while dealing with rates and ratios. The harmonic mean is to be used in computing special types of average rates or ratios where the time factor is variable and the act being performed e.g., distance, is constant.

Hence, the averages cannot be used indiscriminately. For sound statistical analysis, a judicious selection of the average depends upon:

- the nature and availability of the data,
- the nature of the variable involved,
- the purpose of the inquiry,
- the system of classification adopted, and
- the use of the average for further statistical computations required for the inquiry in mind.

However, since arithmetic mean satisfies almost all the properties of an ideal average, is quite familiar to a layman, and has very wide applications in statistical theory at large, it may be regarded as the best of all the averages. Thus in case of doubt, it may be advisable to use the arithmetic mean rather than any other average.

**References:**

Business Statistics – S.C Gupta