topic Re: block and split-plot analysis by "fit model" in Discussions

block and split-plot analysis by "fit model"

mahroo — Fri, 26 Jul 2013 04:23:45 GMT

Hi,

I am having a data set with blocking factor, whole plot, and split plot factors. I want to use fit model in JMP. I read some stuff online and figured out for whole plot I need to specify it as "random effect", but couldn't find anything regarding the blocking factor. How I tell fit model in JMP that a factor is a blocking factor?

Re: block and split-plot analysis by "fit model"

David_Burnham — Fri, 26 Jul 2013 15:35:16 GMT

I'm not sure if you are saying that you have a blocking factor and a whole plot factor, or just a whole plot factor. Either way, it's most common that the blocking factor acts as a random effect - so you select the term in your model specification then choose random effect from the attributes option. You'll find more details under Help>Books>Modeling and Multivatiate Methods ... just do a search for split-plot

Re: block and split-plot analysis by "fit model"

mahroo — Fri, 26 Jul 2013 15:45:59 GMT

I actually have both.

So, does this mean if the blocking factor is not random, we don't need to specify anything for the blocking factor different from other factors. so what is the difference between a blocking factor and a normal factor when fitting a model?