## Can one factor overpower another when analyzing full factorial RCBD?

Solved
Occasional Contributor
Posts: 6

# Can one factor overpower another when analyzing full factorial RCBD?

[ Edited ]

My apologies for the somewhat convoluted title. Let me explain:

We are conducting a greenhouse study in which we are looking at plant genotype (factor A) and irrigation (factor B). Plants are grown in pots to which we randomly assigned both genotype and irrigation. The study is a full factorial RCBD with five blocks. In analyzing the data, I notice that the effect of irrigation is very strong, and I wonder if the variability associated with it overshadows potential interactions? Here is an example of analyzed data for stomatal conductance in which we took two subsamples per plant.

SAS Output

SAS Output
Type III Tests of Fixed Effects

Effect    Num DF    Den DF     F Value       Pr > F
Genotype   2       50               3.06         0.0559

Irrigation    1       50      1168.62         <.0001
Interaction   2      50              1.97           0.1506

Any input would be much appreciated!

David

Accepted Solutions
Solution
‎08-08-2017 11:12 AM
Posts: 1,917

## Re: Can one factor overpower another when analyzing full factorial RCBD?

If you truly have a full factorial, then the answer to your question is NO. Each factor and interaction explains a certain amount of variability of the response, independently of the others, and so the big value you get is because the effect of irrigation is huge, and the effect of genotype and interaction is quite small.

All Replies
Solution
‎08-08-2017 11:12 AM
Posts: 1,917

## Re: Can one factor overpower another when analyzing full factorial RCBD?

If you truly have a full factorial, then the answer to your question is NO. Each factor and interaction explains a certain amount of variability of the response, independently of the others, and so the big value you get is because the effect of irrigation is huge, and the effect of genotype and interaction is quite small.

Occasional Contributor
Posts: 6

## Re: Can one factor overpower another when analyzing full factorial RCBD?

Thank you, Paige. That's what I was thinking, but wanted more expert advice.
Thanks again.
SAS Super FREQ
Posts: 3,753