Hello,
I am new to coding in R and have come up with the following code to perform a nested 2-way ANOVA (with Tukey post-hoc) to be able to account for individual animal variability within each group. I am wondering if someone can confirm this is correct or provide alternative methods? I am assessing the effects of diet and stress on certain cellular outcomes, with n=3-5 animals/group. Thank you!
# Load required packages
library(lme4)
library(emmeans)
library(ggplot2)
# Convert factors
Data_For_R$Diet
No this looks completely fine (just make sure your categorical variables are coded as factors if you're using numerical identifiers for them, been there done that), the only thing I would add is a check for residual violations for the mixed effects model,
plot(resid(model_lmer))
qqnorm(resid(model_lmer))
qqline(resid(model_lmer))
Jochen Wilhelm
that is a good point, I suppose I do not need both. To my understanding, the aov models do not take into account the fact that measurements within individual animals will likely be more similar than from other animals, so the lmer model is needed for the nested design to account for this? I originally put separate aov models for each factor to gain information on each factor independently, as well as their interaction, for curiosity sakes. Any suggestions to alter are much appreciated!
The Error(AnimalID/Diet/Stress) term in the model accounts for the nested structure, where animals are nested within Diet and Stress conditions. This is crucial when you have repeated measures on the same animals.
Two-Way ANOVA: The model includes both Diet and Stress as factors and their interaction (Diet*Stress) to assess their main effects and any potential interaction between them.
TukeyHSD: This function performs Tukey's Honestly Significant Difference post-hoc tests to identify which specific groups (Diet or Stress levels) significantly differ from each other.
Alternative Approach (using lme4 package):library(lme4) # Fit the model using lmer() function model_lmer
- Badges
- Science topic
- Similar topics
- Higher Education
- Universities