Home » User manual » ANOVA » Tests of Multiple Comparisons

Dunnett

The Dunnett strategy is used to compare all levels with a reference level.

Example: We consider a data set with one factor and several levels, as in the following table.

Factor Fiber Strenght
15 7
15 7
15 15
15 11
15 9
20 12
20 17
20 12
20 18
20 18
25 14
25 18
25 18
25 19
25 19
30 19
30 25
30 22
30 19
30 23
35 7
35 10
35 11
35 15
35 11

 

click here to download the data

To use the tool Dunnett the user must follow these steps: 

1. Access the menu:

Action $ \blacktriangleright $ ANOVA $ \blacktriangleright $ Multiple Comparisons $ \blacktriangleright $ Dunnett

2. The following screen will be shown;

  

3. With the cursor in the field Factor select the column that refers to the Factor (without the name of the column).

4. With the cursor in the field Response Variable select the column that refers to the Response Variable (without the name of the column);

 

5. Choose the Control Group, with which the test will compare all the means. If none is chosen, it will use as Control Group the smallest of them;

6. Choose the Significance Level α. In this case, we will use α = 0.05;

7. Choose the Alternative Hypothesis: Not Equal, Lower than 0 or Greater than 0.

8. In Show Results, choose one of the options. We suggest the option New Sheet, because Action does not have the undo command;

9. To finish, click Ok.


 

Results and Interpretation

Once the process is finished, the following results will be shown:

The interpretation of Dunnet test is analogous to Tukey test, that is, the equality between the levels will be rejected if the p-value is < α, or if the confidence level does not contain the value "zero". Thus, in the example, we reject the hypothesis of equality between all levels, except for 15 and 35.