JMP 13 - where did Dunnett go?I use Dunnett in greenhouse trials when we are screening new strains for effects over control. It used to 'live' in the scroll down box where you choose the type of comparisons (Tukey, Student's T, LSMeans Contrast, etc.). It is no longer an option. Does anyone know where to find the Dunnett option or do you know how to compare one treatment to the control with the options now available?
Re: JMP 13 - where did Dunnett go?Could we script Dunnett in the model type we are using?
Re: JMP 13 - where did Dunnett go?The attachment did not work, and Dunnett is not available for that platform. We will just have to wait for them to get it done!
Jesper
Re: JMP 13 - where did Dunnett go?Sorry I have not followed up in a long time; I was out on maternity leave. Am I missing your attachment?
Re: JMP 13 - where did Dunnett go?Please see attachment.
Re: JMP 13 - where did Dunnett go?Dunnett is not implemented for REML.
If there is some computational reason why it is not possible or particularly cumbersome, I don't know.
Maybe it just hasn't been implemented - I will add it as a wish to JMP development.
Re: JMP 13 - where did Dunnett go?I think if you still want to include random effect in the model and do Dunnett test. maybe you can change the method from REML to EMS. Then Dunnett can be run under the Model with Random effect.
Br juna
Re: JMP 13 - where did Dunnett go?I have a theory, and maybe MTOF can tell me if I'm off-target or not.
My guess is you have Treatment in your model, and you've assigned it the Random attribute. To JMP, the Random attribute tells the modeling platform that you've sampled from a larger set of Treatments, but you're not actually interested in how Treatment choice affects the outcome, you just recognize that Treatment choice WILL affect the outcome and you want to estimate how much. In other words, there's not a baseline control "Treatment" because Treatment isn't something you controlled, just something that happened. So Dunnett's test, from JMP's perspective, isn't reasonable, because there's no such thing as a control random sample.
My suggestion would be to try removing the Random attribute from Treatment. I'm assuming you control what the treatments are (Product A v Product 
and you want to compare them, so you can safely treat it as a Fixed Effect.
For further explanation, I found this blurb in the JMP Help that might explain it better than I did:
----
Random Effects
A random effect is a factor whose levels are considered a random sample from some population. Often, the precise levels of the random effect are not of interest, rather it is the variation reflected by the levels that is of interest (the variance components). However, there are also situations where you want to predict the response for a given level of the random effect. Technically, a random effect is considered to have a normal distribution with mean zero and nonzero variance.
Suppose that you are interested in whether two specific ovens differ in their effect on mold shrinkage. An oven can process only one batch of 50 molds at a time. You design a study where three randomly selected batches of 50 molds are consecutively placed in each of the two ovens. Once the batches are processed, shrinkage is measured for five parts randomly selected from each batch.
Note that Batch is a factor with six levels, once for each batch. So, in your model, you include two factors, Oven and Batch. Because you are specifically interested in comparing the effect of each oven on shrinkage, Oven is a fixed effect. But you are not interested in the effect of these specific six batches on the mean shrinkage. These batches are representative of a whole population of batches that could have been chosen for this experiment and to which the results of the analysis must generalize. Batch is considered a random effect. In this experiment, the Batch factor is of interest in terms of the variation in shrinkage among all possible batches. Your interest is in estimating the amount of variation in shrinkage that it explains. (Note that Batch is also nested within Oven, because only one batch can be processed once in one oven.)
Now suppose that you are interested in the weight of eggs for hens subjected to two feed regimes. Ten hens are randomly assigned to feed regimes: Five are given Feed regime A and five are given Feed regime B. However, these ten hens have some genetic differences that are not accounted for in the design of the study. In this case, you are interested the predicted weight of the eggs from certain specific hens as well as in the variance of the weights of eggs among hens.
Re: JMP 13 - where did Dunnett go?That was 13.1, but I also still have JMP 12 installed for comparison. Looking at your report, it looks like you've given the Random attribute to the Treatment ID number, and you're using REML to analyze it. When I do that in 12 or 13, Dunnett's is not available as an option. I don't understand why, yet, but I'm digging into the statistical details to see if there's a reason or if it's just an oversight.
JMP12JMP12
JMP13JMP13
Re: JMP 13 - where did Dunnett go?What verison do you have, 12 or 13? This is what my effects detail looks like:
Re: JMP 13 - where did Dunnett go?Which platform do you typically use? In the Fit Y by X platform, Dunnett's is under Compare Means in the hotspot menu, while in Fit Model, it's under the hotspot menu for Effect Details.
