|
A normal distribution is a frequency distribution which has a symmetrical, bell shape when drawn as a graph showing the value on the horizontal axis and the number of times that each value occurs on the vertical axis. This means that examples in a set of data are clustered around the average, while relatively few examples tend to one extreme or the other. The normal distribution accurately represents variations in a wide range of attributes, and thus it is an extremely useful concept in statistics and other fields. Examples of normal distributions include the height or weight of any population, employee performance and the returns of a diversified asset portfolio. Although the bell curve for a normal distribution will always be symmetrical, it will vary in steepness according to the particular population or characteristic that it is describing. That is, if the population is relatively uniform (for example, most students in a class have nearly the same height), then the curve will be relatively steep. Likewise, if their is a wide variety in the characteristic being measured, then the curve will be flatter. Not all phenomena have a normal distribution. Also, even when taking samples of some attribute that is normally distributed within a population, the data may not appear to represent a normal distribution unless the sample size is sufficiently large. Created October 23, 2006. |