Replicated samples will be collected at two soil depthes. Two factors were tested in the experiment. When doing ANOVA, is it necessary to consider soil depth as a factor?
Are samples at the two depths taken from the same core? That is do you core and then take a sample from two depths or do you do randomly assign each core to one of two depth treatments and then take a single sample from the assigned depth? I'm guessing the former is the case.
If the samples taken from the same core they are not independent of each other. Therefore, 'core' must be a factor in the ANOVA model otherwise you will be pseudoreplicating. Likewise, samples taken from the same depth are also not independent of each other and must be included in the model. Here are the factors you'll need in the final model:
1) Factor A fully crossed with Factor B
2) Depth
3) Core (nested within Factor A crossed with Factor B)
Depending on the full design and your research question, you may want to do a three-way fully crossed model with Factor A, Factor B, and depth.
If each core only has one sample taken from it from a single depth then you do not need to include Core as a factor.
Are samples at the two depths taken from the same core? That is do you core and then take a sample from two depths or do you do randomly assign each core to one of two depth treatments and then take a single sample from the assigned depth? I'm guessing the former is the case.
If the samples taken from the same core they are not independent of each other. Therefore, 'core' must be a factor in the ANOVA model otherwise you will be pseudoreplicating. Likewise, samples taken from the same depth are also not independent of each other and must be included in the model. Here are the factors you'll need in the final model:
1) Factor A fully crossed with Factor B
2) Depth
3) Core (nested within Factor A crossed with Factor B)
Depending on the full design and your research question, you may want to do a three-way fully crossed model with Factor A, Factor B, and depth.
If each core only has one sample taken from it from a single depth then you do not need to include Core as a factor.
Yes, if the factors evaluated are expected to differ by depth you would need to consider this in the anova as an extra factor. If non signifcant, you can carry out the other analysis ignoring depth.
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Modeling
- Analyses of Variance
- ANOVA