Share your ideas for the JMP Scripting Unsession at Discovery Summit by September 17th. We hope to see you there!
Choose Language Hide Translation Bar
Highlighted
Shelsi
Level I

Two-Way ANOVA of Unbalanced Data Containing Levels?

I have SCFA data from 4 types of fiber at 3 inclusion levels (25, 50, 75%) that were mixed with protein and fermented for 24 and 48 hours (in triplicate for each time point). Additionally, there was a control containing only protein (also in triplicate for each time pt), and fiber only controls for each of the fiber types (also triplicates).

Here is a small example, I have time and Fiber Inclusion set as nominal btw.

Screen Shot 2019-07-01 at 4.41.58 PM.png

 

When I try to do a full factorial by time, fiber type, and inclusion level I get dropped DFs because, of course, the all protein sample cannot have data for containing some level of fiber.

 

Firstly, I've been operating under the assumption I cannot simply change the name of the all protein control and insert the same data in for each fiber type.

For example: 

The 24 hour "(Control) All Protein" 0% fiber sample with 33.95 mM SCFAs would become:

  • 24 hour "FOS" 0% fiber = 33.95 mM SCFA
  • 24 hour "Pectin" 0% fiber = 33.95 mM SCFA
  • 24 hour "WB+PS" 0% fiber = 33.95 mM SCFA
  • 24 hour "Even Mixture" 0% fiber = 33.95 mM SCFA

I haven't done it because I assume the duplicated data would weigh the analysis with more observations than actual (all told: n=24 instead of the actual n=6).

 

I know I can combine the "Fiber Type" and "Fiber Inclusion" columns but then I'm unable to see interactions. Is there some way around this? I've been staring at this for far too long and can no longer see the forest for the trees! Please help!  Also, I don't know anything about scripting (yet).

Article Labels

    There are no labels assigned to this post.