Code for Human activity drives establishment, but not invasion, of non-native plants on islands

This software release contains 7 R markdown files, the contents of which are described below. The raw data analyzed in these files is available at the following DOI: https://scholarworks.umass.edu/data/175/

1. SEMs_Native.Rmd

- Code for the piecewise structural equation model (pSEM) that predicts native plant richness on islands using the covariates described in the manuscript.
- Results of this code are reported in Figure 3A and Table 2.
- The a priori hypothesis used to start the modeling process can be found in Figure S1A.

2. SEMs_Established.Rmd

- Code for the pSEM that predicts established plant richness on islands.
- Results of this code are reported in Figure 3B and Table 2.
- The a priori hypothesis used to start the modeling process can be found in Figure S1B.

3. SEMs_Invasive.Rmd

- Code for the pSEM that predicts invasive plant richness on islands.
- Results of this code are reported in Figure 3C and Table 2.
- The a priori hypothesis used to start the modeling process can be found in Figure S1B.

4. Extra_Native_SEMs.Rmd

- Our dataset had three pairs of correlated variables (see "Model Building" subsection under "Methods" in manuscript).
- This file shows the steps we took to decide which variable (from each of the three pairs) to use in the SEMs_Native.Rmd file above.

5. Extra_Established_SEMs.Rmd

- Our dataset had three pairs of correlated variables (see "Model Building" subsection under "Methods" in manuscript).
- This file shows the steps we took to decide which variable (from each of the three pairs) to use in the SEMs_Established.Rmd file above.

6. Extra_Invasive_SEMs.Rmd

- Our dataset had three pairs of correlated variables (see "Model Building" subsection under "Methods" in manuscript).
- This file shows the steps we took to decide which variable (from each of the three pairs) to use in the SEMs_Invasive.Rmd file above.

7. Simulations.Rmd

- Some of the generalized linear models (GLMs) within the pSEMs (created in the files above) seemed to slightly violate some of the usual assumptions.
- In this file, we run some data simulations to make sure that the slighly violated assumptions don't invalidate our results.
- See "Supplemental Methods" in Appendix 1 for more details.