The RIEEE EnviroData Collaborative (EDC) is designed to meet a growing need at Appalachian State University: building environmental data science and modeling capacity through hands-on skill building, shared learning, and collaborative exchange. As the pace of innovation in areas like machine learning, remote sensing, and cloud computing accelerates, researchers across disciplines are seeking opportunities to update and expand their technical skill sets to continue learning beyond graduate training.
Through periodic, skill-focused workshops led by both campus experts and invited facilitators, this program will create space for faculty and research staff to gain practical experience with new tools, exchange knowledge, and grow their capacity to conduct cutting-edge, data-intensive research. Beyond individual skills, the EDC will foster a culture of collaboration and interdisciplinary connection. Our goal is to cultivate a vibrant, exchange-driven community that enhances the visibility of environmental data science at Appalachian and increases the impact of faculty research.
Our objectives for the RIEEE EnviroData Collaborative:
- Create space for faculty and research staff—regardless of discipline or career stage—to gain practical experience with new tools, exchange knowledge, and grow their capacity to conduct cutting-edge, data-intensive research
- Offer structured time, community support, and accountability to help participants finally engage with that tool or technique they’ve been meaning to learn “when they have time”
- Cultivate a vibrant, interdisciplinary community of Appalachian State faculty and research staff advancing environmental data science and modeling capacity
Upcoming Workshops
- Causal Inference in R: An introduction
- Date: Friday, February 20, 2026
- Instructor: Dennis Guignet, Associate Professor of Economics
- Description: We have all heard the maxim “correlation does not imply causation”, but causation is often what environmental researchers seek to find. In this workshop we will explore tools to infer causal relationships in real-world data, using an example of environmental pollution and home values. The two-hour workshop will feature a presentation on necessary background material and then segue into an applied R-based example, with plenty of time for discussion and questions throughout. Through this workshop, you will:
- Gain a basic understanding of some of the key tools for causal inference with observational data, including difference-in-differences, regression discontinuity, and instrumental variables. We will discuss the idea behind these different approaches, their strengths and weaknesses, and to what scenarios they can be applied.
- Work through an R-based example of the difference-in-difference approach
- Talk through ways in which these methods might be applied to your own research
- Meet other environmentally-interested faculty across campus who are seeking to “level up” their research skills and build community
- Please Note: This workshop is designed for faculty or staff with basic R familiarity and provides both conceptual understanding and applied skills relevant across disciplines. Participants should bring their own laptops with a pre-installed copy of RStudio (https://posit.co/download/rstudio-desktop/).
Past Workshops
Nov 12, 2025: Supervised Machine Learning with R
Instructors: Hasthika Rupasinghe, Associate Professor of Mathematical Sciences; and Lasanthi Watagoda, Assistant Professor of Mathematical Sciences
Description: This hands-on workshop introduces participants to R Markdown as a tool for conducting and documenting data analysis in a single, reproducible environment. Participants will use RMarkdown throughout the session to write code, view results, and generate their own workshop notebook. We will then explore the fundamentals of supervised machine learning using R, focusing on regression models for continuous outcomes. The session will progress from basic linear regression to modern machine learning methods, including shrinkage approaches (LASSO, Elastic Net, AHRLR) and tree-based ensembles (Bagging, Random Forest, Boosting). By the end of the workshop, participants will understand how different modeling strategies balance interpretability, flexibility, and predictive accuracy.
Sep 23, 2025: Getting Hands-On with Data: Interactive visualizations in R
Instructors: Michael Erb, Research Scientist, RIEEE and Research Associate Professor of Geological and Environmental Sciences; and William Armstrong, Associate Professor of Geological and Environmental Sciences
Description: Interactive visualizations provide an engaging way of exploring large datasets and browsing data with peers and students. In this workshop, you’ll learn to process data in R and produce interactive figures using several R packages: tidyverse, plotly, and leaflet. We’ll focus on two main examples: an interactive 3D scatterplot of penguin data—which can be zoomed, rotated, and filtered—and an interactive map of North Carolina population. Other visualizations will be included throughout. Participants will learn to:
- Browse these figures locally or share them online using App State webspace
- Load data in R and process it with tidyverse
- Create interactive figures with plotly
- Create an interactive map with leaflet
