Random effects are a critical part of modern (and even not-so-modern) statistical analysis. A random effect lets you account for non-independence in your experimental structure (nested plots, repeated observations, etc.). Consequently, using random effects not only increases the predictive power of your models, it also lets you accomplish one of the most universal statistical assumptions: independence. This workshop was presented alongside a "chalk talk" for a lab meeting at the Cornell University Sustainable Cropping Systems Lab and then again at Virginia Tech's Whitehead lab.

Never stop coding! 🌱 + 🤔 = 📊 --> 💡