ANOVA with Minitab: Using General Linear Model

[@Emmaibarralugo]

Great tutorial! It was really good to find this video to know how use this tool in Minitab 17 due the big changes versus Minitab 16. Thank you!

[@brownmsoe]

Thank you, this helped with the chapter 4 and 5 problems a lot. Translating the HW problems from Montgomery's book to minitab was not explained well.

[@mahyar7126]

It was useful a lot, thank you so much teacher

[@howellistance2534]

I agree, really, a very helpful tutorial. Thanks very much

[@seraon]

thank you!

[@6400loser]

This was a very helpful tutorial! I can't seem to get the "factor x *factor y" in my output, however. Are there constraints on the number of levels that need to be present for each or something?? Thanks!!!

[@lauriet498ify]

Any thought on why trying to use the comparisons function to do a Tukey test doesn't work. I keep trying it but it fails to produce any results as in nothing pops up.

[@americanbluediamonds]

How to do Latin Square in Minitab? Thanks

[@lawyewka4726]

Hello, may I know how to do this in minitab macbook version?

[@ProfNoria]

Unfortunately I don’t use MacBook, so I’m unable to accommodate your request.

[@kariiamba7324]

Thankyou

[@shambelawgchew9110]

Thank you

[@cesur1149]

should the data be normal distributed?

[@christinenoria1214]

Yes. ANOVA is a parametric test and assumes normality. There are nonparametric tests that do not require a normal distribution, but their power is much lower.

[@cesur1149]

thanks!! i executed a DOE and the results failed the normality test ......what statistical test should i use?

[@christinenoria1214]

Ces Ur, there isn't a simple answer if you want to test a factorial design. For one-way ANOVA, the comparable test is the Kruskal-Wallace H. I have found Tabachnik & Fidell's multivariate analysis text to be a useful reference in terms of assessing robustness to violations of assumptions. The appropriate course of action really depends on the design of your study and the data itself.