Measures of Variability

Fundamentals of Social Statistics by Adam J. McKee

Knowing the center of a distribution gives us a partial picture of the characteristics of a set of scores, but that picture is often not in clear focus. In other words, a measure of central tendency, standing alone, can lead us to an inaccurate view of the characteristics of a distribution.

Let us take the weather report as an example. Suppose the weather report tells you that the average temperature tomorrow will be 70 degrees. That sounds comfortable, so you decide to wear jeans and a t-shirt. Little do you know that the low will be a chilly 45 degrees and that the high will be a sunny 95 degrees. The problem is that you did not take into account how much the temperature at any given time would deviate from the mean. This potential confusion is why weather reports usually report the high and low temperature. When it comes to weather, we care about extreme scores more than we do the middle.

For most social science variables, we also want to consider how far scores are spread out from the middle.
Social science researchers often refer to the high and low scores as the minimum and maximum. Reporting the minimum score and the maximum score is an elementary approach to reporting the variability of a distribution of scores.

Four other measures of spread will be discussed in this text:

1. The range
2. The interquartile range (IQR)
3. Standard deviation
4. Variance

Synonyms of Variability

The idea of variability among a set of scores is very important in statistics. There are several terms that researchers use to talk about variability. Spread, Dispersion, Variability, and Deviation are commonly used. 

Your author likes the homemade word Spreadoutness.


[ Back | Contents | Next ]

Last Modified:  02/03/2021

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.