BookmarkSubscribeRSS Feed
tswong
Calcite | Level 5

Hi,

 

I am running a GH experiment for 16 different plants or "treatments" in batches. They are all inoculated with the same nematode and I measured the nematode eggs and root weight after 45 days. There is a total of 4 "batches" with different treatments in each batch set up as a RCBD. Treatments had 5 reps (pots) per batch and each treatment was replicated twice. A "control" treatment was used in each batch. I'm trying to figure out how to combine the repeated measures of each type of plant and then rank the combined means. However, there are significant variations between the controls of each batch. 

 

Would I be able to use PROC GLIMMIX to set "batch" as a random effect to account for the differences across batches? Or should I use mean-center the controls and adjust the measurements from each plants to the batch? I've tried looking online for batch effects, but couldn't find any in this forum. In the end, I'm trying to rank which plants were good hosts or non-hosts for the nematode. 

 

Thanks!

 

 
4 REPLIES 4
PaigeMiller
Diamond | Level 26

There is a total of 4 "batches" with different treatments in each batch set up as a RCBD.

 

I'd like to see the design matrix for the above, as the words don't clearly (to me) indicate what the design is.

 

Also, for a randomized complete block design, what are the blocks?

--
Paige Miller
tswong
Calcite | Level 5

Sorry about that! Here's an example of how I set up each run or batch. The letters would be the different treatments or cultivars. 

 

Screen Shot 2019-10-16 at 6.08.13 PM.png

Then I would set up 5 blocks on a bench. Here's an example of the first 2 blocks, but blocks 3-5 would be randomized. 

 

Screen Shot 2019-10-16 at 6.13.22 PM.png

PaigeMiller
Diamond | Level 26

I'm still lost. I don't see how the batches are connected to the blocks, or how either of those are connected to treatments A through P.


How is the "control" identified here?

--
Paige Miller
tswong
Calcite | Level 5

The plant or treatment "A" is the control and is run in each of the 4 batches. Each batch is run at different times in the greenhouse. When setting up the batch, I arrange the treatments in 5 randomized blocks on a bench because each treatment has 5 reps in a batch. 

sas-innovate-2024.png

Available on demand!

Missed SAS Innovate Las Vegas? Watch all the action for free! View the keynotes, general sessions and 22 breakouts on demand.

 

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
  • 4 replies
  • 511 views
  • 0 likes
  • 2 in conversation