Back to search

Michigan State University

Bayesian Statistical Methods : Advancing Field-Level Risk Assessment in Agriculture, Accessible Statistical Training, and Inclusive Global Education

Abstract

dc:description

Bayesian statistical methods have gained widespread recognition across disciplines due to their intuitive probabilistic nature, incorporation of prior domain knowledge through prior distributions, robust uncertainty quantification, and suitability for working with relatively small datasets. However, the successful implementation, interpretation, and communication of Bayesian methods require a solid understanding of both probability theory and computational techniques. As a Bayesian statistician, I have developed and employed Bayesian methodologies to tackle applied problems across disciplines, collaborating with experts from different fields. Additionally, as a statistics educator, I have designed curriculum to share fundamental skills necessary for comprehending and performing Bayesian analysis.In this dissertation, I present three projects that illustrate the complexities of utilizing Bayesian methodology in applied problems as a statistician and effectively communicating and teaching the fundamentals of Bayesian theory and application to diverse audiences as a statistics educator. Firstly, I introduce a project that develops Bayesian linear regression and prediction methodologies to quantify the field-level risk mitigation associated with regenerative soil practices in agriculture at a regional scale. Secondly, I discuss the development and execution of an inclusive and accessible workshop aimed at teaching research professionals how to learn the statistical programming language R, as mastering such a language is crucial for practical Bayesian analysis. Finally, I relay how the work from the preceding projects helped build the foundation of a five-day novel training experience to teach the fundamentals of Bayesian statistics to agronomy professionals in Africa. These projects collectively highlight the multifaceted nature of Bayesian analysis, from its application in addressing real-world challenges to the importance of statistical education and knowledge transfer. By sharing the insights gained through these projects, I aim to contribute to the advancement of Bayesian methodology and facilitate the adoption of Bayesian statistics across disciplines.

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Manski, Sarah
Contributors dc:contributor
  • Viens, Frederi
  • Green, Jennifer
  • Snapp, Sieglinde
  • Sakhanenko, Lyudmila

Subjects

dc:subject × 3

Rights

dc:rights
Statement dc:rights
  • Attribution-NonCommercial-ShareAlike 4.0 International
Language dc:language
English

Identifiers

dc:identifier.*
Identifier
etd:51058
local:Manski_grad.msu_0128D_20172

Chain of custody

source
Harvested from
Michigan State University
Base URL
d.lib.msu.edu/oai
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Manski, Sarah. Bayesian Statistical Methods : Advancing Field-Level Risk Assessment in Agriculture, Accessible Statistical Training, and Inclusive Global Education. 2023. https://doi.org/doi:10.25335/aj8v-nz88