ABSTRACT
The variance is given by SS (the sum of squares) and the number of degrees of freedom by df. The mean square (MS) is calculated by dividing SS by df. The ratio F is calculated by dividing the between-group MS by the within-group MS. If there is little difference between the individual samples, F is close to unity. The probability of the result occurring is derived from tables a in which F is read against df. An analysis of variance test will indicate that the difference between the five means did not occur by chance at the 95% level; it will not indicate which mean varies from which. It is therefore necessary to perform a further test to show which means are different from which others. An example of such a test is Scheffé’s test.
