ELIGIBILITY: This project is for a student taking the MSc in Statistics and Sustainability.
Increasing the number of species sown in an agricultural grassland system provides a potential way to improve the sustainability of the overall system. Experiments designed to test this typically have species diversity manipulated across field-scale plots, with a plot-level response such as harvested yield measured. The relationship between species diversity and the response is then modelled. There are multiple statistical methods that can be used to analyse this relationship. Some rely on using the number of species sown in the plot as a predictor in a regression model, while others use the identity of the species sown and their proportions, in addition to the number of species. This project will investigate data from a published paper; in the paper, the statistical method used relied on the number of species as the measure of species diversity. The work in this project, will be to apply and compare alternative statistical methods of assessing the species diversity – response relationships to determine the impact of the various statistical approaches on inference.