A
standard deviation (SD) is a score that we can use for the spread of data away from an average, although
only in the case of continuous (interval or ratio) data. There is also a related measure
called a
variance. Take a look at
Figure 7.1 Example of a final descriptive statistics table on page 68 which gives standard deviations of some variables.
The standard deviation and variance are very important, forming the basis for a great proportion of the world of statistics. Therefore,
understanding the concept is important.
The standard deviation gives a measure of the average dispersion of a variable around
the average. Precisely, if the variable is normally distributed in the complete population,
the standard deviation tells us the distance away from the mean within which approximately 66% (two-thirds) of the population would be expected to
lie.
-
If the population is normally distributed, then about 66% of the data lies 2.35 points
above and below 5.5 respectively.
-
The crucial thing to remember about a standard deviation is that it captures a range
of data within which most of the population (two-thirds) is expected to lie. Therefore,
it reflects the representative spread of the data away from the mean.
The variance of a variable is the standard deviation squared. As stated, you need
to understand this without necessarily having to know how it is calculated. The important
thing is to remember that, like the standard deviation, the variance is a measure
of variable spread.
Further reading on the meaning and use of the SD and variance can be done in any introductory
statistics text.