Minitab provides three graphs that help you identify the terms that influence the response: a Pareto chart, a normal plot, and a half-normal plot. These graphs allow you to compare the relative magnitude of the effects and evaluate their statistical significance.
The threshold for statistical significance depends on the significance level (denoted by α or alpha). Unless you use a stepwise selection method, the significance level is 1 minus the confidence level for the analysis. For more information on how to change the confidence level, go to
Specify the options for Analyze Factorial Design. If you use backwards selection or stepwise selection, the significance level is the significance level where Minitab removes a term from the model, known as
Alpha to remove. If you use forward selection, the significance level is the significance level where Minitab adds a term to the model, known as
Alpha to enter. For more information on the choices for the stepwise methods, go to
Perform stepwise regression for Analyze Factorial Design.
Note
- For a 2-level design, if the number of terms in the model equals the number of runs, the standardized effects cannot be calculated. Minitab shows the unstandardized effects and uses Lenth's method to draw a reference line for statistical significance. For more information on Lenth's method, go to Methods and formulas for the effects plots in Analyze Factorial Design and click "Lenth's pseudo standard error (PSE)."
- For a general full factorial design, if the number of terms in the model equals the number of runs, the standardized effects cannot be calculated. Minitab does not produce a Pareto chart in this case.
- Pareto
- Select to determine the magnitude and the importance of an effect. The chart displays the absolute value of the effects and draws a reference line on the chart. Any effect that extends beyond this reference line is statistically significant.
- Normal
- Select to compare the magnitude and statistical significance of main and interaction effects from a 2-level factorial design. The fitted line indicates where you would expect the points to fall if the effects were zero. Significant effects have a label and fall toward the left or right side of the graph.
- The normal probability plot displays negative effects on the left side of the graph and positive effects on the right side of the graph.
- Half Normal
- Select or to compare the magnitude and statistical significance of main and interaction effects from a 2-level factorial design. The fitted line indicates where you would expect the points to fall if the effects were zero. Significant effects have a label and fall toward the right side of the graph.
- The half normal plot displays the absolute value of all effects, positive and negative. Instead of putting negative effects to the left and positive effects to the right, all the significant effects are on the right side, which emphasizes their relative magnitudes.
For 2-level factorial and Plackett-Burman designs, select Display only model terms to display only the terms that are in the model or select Display
all terms to display all of the terms in the graphs.
For split-plot designs, you can choose to:
- Display all terms, subplot effects only
- Display all terms, subplot and whole-plot
effects
- Display only model terms, subplot effects
only
- Display only model terms, subplot and whole-plot
effects
