One of the biggest advantages of using PROC GLIMMIX, is that the data coming into the analysis no longer needs to be normally distributed. It can have a number of distributions and SAS can handle it. Our job now is to be able to recognize when a normal distribution is NOT appropriate and which distribution is an appropriate starting place. Non-Gaussian distributions are what these are referred to. Remember Gaussian is the same as calling it Normal.
Where do we start? Think about your data – what is it?
- A percentage?
- A count?
- A score?
How do we know that our data is not from a normal distribution?
- Always check your residuals!
- Remember the assumptions of your analyses?
- Normally distributed residuals is one of them!
Let’s work with the following example. We have another RCBD trial with 4 blocks and 4 treatments randomly assigned to each block. There were 2 outcome measures taken: proportion of the plot that flowered, and the number of plants in each plot at the end of the trial.
Please copy and paste the attached code to create the SAS dataset on your computer.
We will work through the output and how/when you need to add the DIST= option to your MODEL statement. We will also talk about the LINK= function and what it does.