BookmarkSubscribeRSS Feed
🔒 This topic is solved and locked. Need further help from the community? Please sign in and ask a new question.
Mutua
Calcite | Level 5

Dear SAS community,

I did an experiment in climate chambers (different temperatures) with plants infested with two different fungal species. To investigate the development and severity of disease, i sampled 15 plants from each treatment (per climate chamber/ temperature). The plants were used for destructive analysis i.e they were not replaced. The sampling was done per month for a total of four months.

The objective of the study was to investigate the influence of temperature on disease severity and growth of the plants.

Statistical question:

1. Should i perform a repeated measurement analysis on these data?

2. If Yes to question 1, is Proc mixed model the best statistical procedure for this type of analysis?

Your support will be highly appreciated.

Regards,

Peter.

1 ACCEPTED SOLUTION

Accepted Solutions
SteveDenham
Jade | Level 19

I don't see how this can be considered a true repeated measures design, as the subjects were destroyed for measurement.  However, you might wish to model the correlation over time using a spline or some other semi-parametric approach.

Choosing the right mixed model procedure can be difficult, but I would tend toward GLIMMIX, as you can do (almost) everything that MIXED does, plus many more things.  I am reasonably certain that your response variable for severity would not meet the assumptions of NID(0, sigma^2) for the residuals.  In GLIMMIX, you can work around this, depending on the endpoint definition.

Steve Denham

View solution in original post

2 REPLIES 2
SteveDenham
Jade | Level 19

I don't see how this can be considered a true repeated measures design, as the subjects were destroyed for measurement.  However, you might wish to model the correlation over time using a spline or some other semi-parametric approach.

Choosing the right mixed model procedure can be difficult, but I would tend toward GLIMMIX, as you can do (almost) everything that MIXED does, plus many more things.  I am reasonably certain that your response variable for severity would not meet the assumptions of NID(0, sigma^2) for the residuals.  In GLIMMIX, you can work around this, depending on the endpoint definition.

Steve Denham

Mutua
Calcite | Level 5

Thanks Steve for your suggestions and answer to my question. I agree with the argument since the subject were destroyed and not replaced during the analysis.

True, choosing the right model was difficult, because i choose Proc Mixed. But with the additional information, and the preference for GLIMMIX, then, i will work around and find how to carry out the analysis. Could i have more questions, i will post them out again to the SAS community.

I thank you so much for your continued support in my quest to learn more Statistical procedure. Thumbs up.

Peter

sas-innovate-2024.png

Join us for SAS Innovate April 16-19 at the Aria in Las Vegas. Bring the team and save big with our group pricing for a limited time only.

Pre-conference courses and tutorials are filling up fast and are always a sellout. Register today to reserve your seat.

 

Register now!

What is ANOVA?

ANOVA, or Analysis Of Variance, is used to compare the averages or means of two or more populations to better understand how they differ. Watch this tutorial for more.

Find more tutorials on the SAS Users YouTube channel.

Discussion stats
  • 2 replies
  • 1361 views
  • 0 likes
  • 2 in conversation