Standard Deviation
The standard deviation is the most frequently used measure of variability.
The formula for calculating the standard deviation is:
\( s=\sqrt{\frac{\sum\left(x-\bar{x}\right)^2}{n-1}} \) (6.6)
\( x \) - each
individual value in the data set
\( \bar{x} \) - the
mean of all values in the data set
\( n \) - sample size
Procedure
- List each value. When calculating standard deviation, it is not necessary the values to be in any order.
- The mean of the data set is calculated.
- The mean is subtracted from each individual value.
- Each individual difference is squared.
- The sum all the squared deviations about the mean is calculated.
- The sum is divided by \( n-1 \) .
- Finally the square root is calculated.
Example
The data set [71, 50, 48, 67, 53] is given.
Each column of Table 6.3 represents a step from the procedure:
Table 6.3 Calculation of the standard deviation
\( x \) |
\( \bar{x} \) |
\( \left(x-\bar{x}\right) \) |
\( \left(x-\bar{x}\right)^2 \) |
71 |
57.8 |
13.2 |
174.24 |
50 |
57.8 |
-7.8 |
60.84 |
48 |
57.8 |
-9.8 |
96.04 |
67 |
57.8 |
9.2 |
84.64 |
53 |
57.8 |
-4.8 |
23.04 |

The sum of the deviations from the mean is always equal to zero.
\( \sum\left(x-\bar{x}\right)=0 \)
The squared deviations are used for escaping the negative sign, and when the sum is calculated it is no more 0:
\( \sum\left(x-\bar{x}\right)^2=438.8 \)
\( \frac{\sum\left(x-\bar{x}\right)^2}{n-1}=\frac{438.8}{4}=109.7 \)
And finally, the square root of the entire value is calculated, because it is necessary to return to the same units with which the calculation originally started.
\( s=\sqrt{\frac{\sum\left(x-\bar{x}\right)^2}{n-1}}=\sqrt{109.7}=10.47 \)