Mode

The mode is the most commonly observed score in a distribution. It is the only measure of central tendency that can be applied to nominal data, and can also be used on ordinal, interval, and ratio data.

Data at the nominal level of measurement are most often presented as a frequency distribution, in a table with two columns, where the first column contains the scores and the second column contains the frequencies, i.e. a number indicating the number of times a score occurs in the distribution.

In addition to table display of data, it is also possible to display them graphically. For graphic display of frequency distributions, a bar chart is most often used, where the tallest bar represents the mode of the distribution. Depending on the type of distribution, there may be multiple peaks on the frequency polygon. The reason for displaying data graphically, in addition to tabular display, is to make it easier to see on the graph the characteristics of the data that are important for determining the appropriate statistical technique.

The mode, like the median, is not affected by extreme values. If samples were taken from a population and the arithmetic mean, median, and mode were calculated for each, then the mode would vary more from sample to sample than the median and arithmetic mean.

If there is only one result with the highest frequency in a set of results, then such a distribution is called a unimodal distribution.

It is possible for more than one outcome to appear in a distribution with the highest frequency, and such a distribution is then called a multimodal distribution. A special subtype of a multimodal distribution is a bimodal distribution where two outcomes have the highest frequency.

If all scores appear the same number of times in a set of results, then all scores are modes and such a distribution is called uniform. In research on humans and animals, uniform distributions are rarely encountered, because most traits in humans and animals follow a normal distribution.