Overview

Here’s what we’ve covered so far:

  • Coding
  • Data exploration
  • (Generalized) Linear models
  • Bayesian vs. Frequentist stats
  • Model selection

In this 2-part lab, we’re going to put everything we’ve learned so far together! We’ll use some familiar data collected in 2016 at Fort Robinson State Park, Nebraska, USA:

Roberts, Caleb P.; Donovan, Victoria M.; Wonkka, Carissa L. et al. (2019). Data from: Fire legacies in eastern ponderosa pine forests [Dataset]. Dryad. https://doi.org/10.5061/dryad.3sp331p

The goal of these labs will be to start with the data and a hypothesis and end with a fully executed, reproducible analysis. Our hypothesis:

Bird species richness will be best predicted by burn status and fire severity. Current habitat structures will not be as predictive.

And here’s a pretty picture of Fort Robinson to motivate you!

Scripts

In the first part, we will download, organize, explore, and wrangle the data. Here’s the for the first part: Rscript.

In the second part, we will execute our analysis and produce results (tables, figures). Here’s the Rscript for the second part: Rscript.