Greetings,
I have a set of data that we are analyzing using a mixed linear model. The
response variable is a 100 mm visual analog scale (VAS) for certain clinical
signs of disease. Measurements were taken four times over a period of 8
weeks. Baseline score at the start of the study is included as a covariate
in the model and is statistically significant. WE also tested for an
interaction effect between the covariate and treatment to see if animals who
were affected more or less severely with the symptom responded differently
to diet. For some of the response variables, we are seeing a significant
interaction; specifically that animals who had more severe signs at the
start of the study were more likely to respond to the treatment diets. My
client is interested in looking at estimated least square treatment effects
(mean VAS scores) for three different estimated covariates that reflect mild
(25), moderate (50) or severe (75) symptoms at the start of the study.
While I can obtain estimated means for all of the factors in the model, I am
not sure how to obtain these.
Thanks for any help you can provide!
Linda Case
Linda P. Case
AutumnGold Consulting
(217) 586-4864
www.autumngoldconsulting.com
[hidden email] or
[hidden email] <mailto:
[hidden email]>
=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD
Locked
