Generalized Full Factorial Design Questions

[weisusu]

I am working with a Generalized Full Factorial Design with factor A and B. Factor A has three levels and factor B has three levels as well. Three replicates were used for a totoal of n=27 runs. The ANOVA model only shows that factor A is significant with a P-value of 0.009. I believed the model is a General Liner Model. Questions:

There is a missing data in one of the runs (remember I have three replicates). Minitab was able to generate the analysis anyway. I wondered if the missing data in one of the runs would create any bias in the analysis?

No pure error lack of fit test was shown in the ANOVA output. Is it because of the design was a generalized full factorial (instead of two-level full factorial design)?

I re-did the ANOVA analysis using only factor A as the main effect since it is the only factor that is significant. How do I get the regression model ( Y = alpha + beta*X) in Minitab? I would like to use the model for prediction.

Thanks

 

[magoo2]

If you go to

stat > regression > regression

your predictor is A and your response is Y

MiniTab should give you a linear model from that.