Analysis of variance requires the following assumptions:
To compare multiple samples, you need two columns. In the first column you need all the datasets. In the second column you need figures that act as a label to allow the samples to be split, for example, say we have three groups, we could use the numbers 1 to 3 as the labels...
| Data | Label |
|---|---|
| 23 | 1 |
| 25 | 1 |
| 18 | 1 |
| 36 | 1 |
| 27 | 1 |
| 26 | 1 |
| 23 | 1 |
| 33 | 1 |
| 25 | 1 |
| 24 | 1 |
| 36 | 2 |
| 40 | 2 |
| 42 | 2 |
| 44 | 2 |
| 33 | 2 |
| 47 | 2 |
| 50 | 2 |
| 49 | 2 |
| 43 | 2 |
| 44 | 2 |
| 37 | 3 |
| 41 | 3 |
| 41 | 3 |
| 43 | 3 |
| 34 | 3 |
| 46 | 3 |
| 51 | 3 |
| 48 | 3 |
| 44 | 3 |
| 44 | 3 |
Once the data is in, you need to pick the following menu item...
This will bring up the following dialog box...
This allows you to pick the column you want to test and the factor/label column. To pick columns, select them in the left side of the dialog box (example hightlighted red above), and click the arrow button in the middle of the dialog box to shift them into the relevant boxes. This will give the follow output table:
From this table it will be possible to tell whether there is enough evidence to reject the null hythothesis. If the null hypothesis is true, then the F value, should be close to 1. Large values for the F ratio indicate that the sample means vary more than you would expect if the null hypothesis were true.
You can tell if your F ratio is large enough to reject the null hypothesis by looking at the observed significance level, which is labelled sig. The results show that the probability of obtaining an F ratio of 35.271 is 0.000. This means that the probablilty of this occuring by chance is less than 0.0005. So you can reject the null hypothesis. It's unlikely that the three groups come from the same population.