Normal distributions can assume a variety of shapes as seen on the right. Note that all three curves have the same mean or average value. Therefore, knowing the average for a normal data distribution will not tell you very much about the curve. However, the standard deviation will give you an idea of how broad the spread is... broad as in the bottom diagram, narrow as in the middle or intermediate as in the top diagram. The standard deviation is a value which if applied to either side of a normal distribution's mean will encompass 68% of the curve's area. 1.96 standard deviation values to either side of the mean will contain 95% of the curve's area. This observation provides the geneticist with an idea of how close a particular value is to the normal distribution's mean. If the value is more than two standard deviations to either side of the mean, then it lies outside 95% of the values in the data spread and represents a very small class of individuals in that spread. In addition to analyzing the amount of spread in a normal distribution, the standard deviation can be used to determine how close the data from your sample fits 95% of the expected data from an ideal curve. To see this in operation, flip to the next card.
