Network Meta-Analysis Framework for Plant Pathology
Project collaborators
Damon Smith, Department of Plant Pathology
Maria Oros, DSI
Project summary
We designed and built an interactive dashboard to simplify the complex process of meta-analysis, a statistical method that combines results from multiple studies to identify trends or draw general conclusions. Meta-analysis can be technically challenging and inaccessible to those without a statistical background, limiting its use to specialists. This project aimed to democratize access to meta-analysis, allowing more people to benefit from the insights that can be derived from combining multiple research studies. The tool provides a single solution for plant pathologists that is generalizable to other crops and fields.
Our work included developing a user-friendly interface, integrating robust statistical tools, and ensuring the dashboard could be used with a variety of data types. The result: an easy-to-use dashboard that allows users to input data and receive clear, visual interpretations and descriptive statistics. This work builds on the DSI-supported meta-analysis of the effects of fungicide application timing on corn yield and heath.
The primary beneficiaries of this work are researchers, policy makers, and educators who need to conduct meta-analyses to inform decisions, but who may lack the technical skills to do so. Additionally, students and non-professionals interested in understanding research trends can benefit from this tool.
We measured the impact of our dashboard by collecting user feedback on its accessibility and effectiveness, tracking the increase in meta-analyses performed by non-experts, and evaluating case studies where our tool influenced decision-making processes. These metrics have shown significant enhancement in user engagement with statistical research and an increase in data-driven decision-making among our users. It also allows the user to input their data and reduce the time they spend on data cleaning and standardization.
Project deliverables
The Streamlit dashboard for visualization serves as the primary interface for visualizing model results and key statistics. Users can select datasets for meta-analysis, both from predefined cases for publication and their own datasets, through a secure, password-protected interface.
Project start and end dates
10/1/23-2/29/24