Use a main effects plot to visualize the effect of the process variables on the response and to compare the relative strength of the effects. Minitab draws a single main effects plot for one process variable, or a series of plots for two or more process variables.

Minitab plots the response means for each process variable level, then connects the points for each process variable. A reference line is drawn at the overall (grand) mean. Look at the line connecting the process variable levels to determine whether or not a main effect is present. You should only view the main effects plots for factors that are significant according to the estimated regression coefficients table from Analyze Mixture Design. A main effect is present when the change in the mean response across the levels of a factor is significant.

·    If the line is horizontal (parallel to the x-axis), there is no main effect present. The response mean does not change depending on the process variable level.

·    If the line is not horizontal (parallel to the x-axis), there may be a main effect present. The response mean does change depending on the process variable level. The greater the degree of departure from horizontal, the stronger the effect. Make sure to determine whether the factor is significant.

Remember that plots just indicate patterns. To determine if a pattern is statistically significant, you must do an appropriate test.

For the fondue data, the main effects plot indicates the following:

·    Temperature: High serving temperature (90°) produced fondue with better flavor than low serving temperature (80°).

·    ServeTime: When the fondue set 1 hour before serving the flavor was better than when the fondue was only allowed to set for 0.5 hours.

·    The overall mean (about 62.1) is plotted across each panel.

By comparing the slopes of the lines, you can compare the relative magnitude of the process variable effects. These plots show that serving temperature had a stronger effect than serving time on the flavor of the fondue.