Hi, I’m working with big data sets very often and normally every data set has several independent variables. Sometimes I need to do ANOVA and get main effects and interaction effects as well as post hoc grouping such as Duncan’s grouping for them. Regarding the output Proc GLM, I’m facing with 2 big problems: after doing analysis and getting e.g. Duncan’s grouping, the tables from GLM procedure are arranged with a combination of variables from biggest values to lowest one, and also the alphabets representing each category is in the same row with the Mean, but not the same cell. These two problems cost me to spend so much time and sometimes get frustrated to rearrange them again and put those alphabets on each mean value as superscript, appropriate for publication. I do some arrangement with Proc sort, but that is not exactly what I really need. So I would like to know: 1. Is there any way to change this output format in an another way based on the arrangement of treatments and not base on biggest value to smallest one, or any other desire format (still containing Duncan’s grouping)? 2. Is there any way to change the location of those alphabets that represent Duncan’s grouping from being in a separate cell to the same cell with Mean values as superscript, as we publish in journals? for example how can I change the following table : A 16.6 30 Rapid A A 16.1 28 Slow B 10.8 30 CO_slow B B 10.6 30 CO_ rapid To this: 10.6 B 30 CO_rapid 10.8 B 30 CO_slow 16.1 A 28 slow 16.6 A 30 rapid I would greatly appreciate any help you may give me, especially if you can write the needed program. Thank you.
... View more
