Resources for Teaching and Design

“Writing Data Stories” is an NSF-funded research project that integrates computational data analysis into middle school science classrooms. The project teaches students to construct “syncretic data stories”—multimodal projects that blend academic statistical analysis of scientific datasets with personal narrative and social reflection.

“Show Your Work!” is a suite of free, web-based introductory Jupyter Notebooks designed for K-12 educators and curriculum designers with little to no prior programming experience. Built on learning sciences principles, the project introduces computational notebooks as epistemic tools, letting teachers experience firsthand what it feels like to conduct notebook-based computational data investigations in specific subject domains.

The Rivulet project provides teachers, curriculum designers, and educational researchers with Python notebooks that streamline querying and fetching environmental datasets from public API endpoints. These notebooks act as automated wrangling pipelines, translating complex federal database APIs into clean, structured data ready for science classrooms.

The MoDa (Modeling and Data) project combines block-based programming, agent-based simulation modeling, and data visualization tools to support students’ reasoning about complex environmental phenomena. By running simulations side-by-side with real-world datasets, students can test, validate, and refine their scientific theories.

“How to be Choosy” provides pedagogical strategies, technical methods, and interactive Jupyter Notebooks and CODAP templates to help educators and curriculum designers make large datasets manageable for classroom instruction. The collection accompanies the Teaching Statistics publication by Wilkerson et al. (2025).

Exploration Units are comprehensive, 2-3 week long curriculum units designed to build deep data literacy and statistical inquiry skills in middle school classrooms. Centered around authentic socioscientific datasets, these units provide bilingual (English/Spanish) student handouts, slides, and teacher lesson guides.

Data Launchpads are highly interactive documents built inside the Common Online Data Analysis Platform (CODAP) that act as scaffolded “on-ramps” for students exploring complex public datasets. Each launchpad features built-in background information, multimedia context-setters, data activators, and guided tutorials on graphs, maps, and filtering.

DataBytes are quick, bite-sized classroom activities (designed to take 30 minutes or less) that encourage students to interpret and analyze data visualizations related to everyday scientific issues. Sourced from news media and scientific agency reports, each lesson provides classroom slides, discussion prompts, and teacher guides.

This dedicated data launchpad provides a curated ecological dataset focusing on Yellowstone National Park’s trophic levels. It allows students to investigate how the reintroduction of gray wolves triggered a trophic cascade, impacting elk populations, willow and aspen growth, and beaver dams.

This launchpad integrates Billboard Hot 100 chart histories (1958–2021) with Spotify’s audio analysis parameters (such as energy, tempo, and danceability). It provides a high-interest dataset for students to examine how popular music has changed over time.

This launchpad explores high-altitude lake water chemistry and ecosystem indicators using long-term monitoring data from Emerald Lake in the Sierra Nevada. It allows students to examine temperature, pH, and nutrient variables over several decades.

This launchpad provides students with structured epidemiological data tracking cases, vaccination rates, and mortality statistics from the COVID-19 pandemic. It introduces key concepts in public health data representation and disease transmission modeling.

This launchpad utilizes California’s CalEnviroScreen dataset to explore cumulative environmental pollution burdens. It allows students to map and analyze various health, environmental, and demographic metrics across different California census tracts.

This module introduces educators to GIS and spatial mapping in Python and R. It guides teachers through plotting geographic coordinates, layering data points, and analyzing location-based datasets.

This module walks educators through basic statistical analysis of datasets, illustrating concepts like mean, median, and data spread. It provides interactive visualizations to help teachers explore mathematical distribution shapes and trends.

This module focuses on longitudinal data and time series analysis, showing educators how to plot chronological data and measure change over time. It provides code examples for analyzing historical sensor or environmental data.

This introductory module guides educators through the basic mechanics of Jupyter Notebooks. It covers how code cells are executed, how comments are read, and how variables are modified, serving as a gentle entry point into coding in Python and R.

This Jupyter notebook connects to the Water Quality Portal (WQP) to query the EPA and USGS Water Quality Exchange (WQX) database. It automates the extraction of water chemistry and ecological indicators—such as dissolved oxygen, salinity, pH, nitrates, heavy metals, and bacterial levels—for local monitoring sites.

Rivulet Next is an experimental collection of interactive browser plugins designed for the Common Online Data Analysis Platform (CODAP). These plugins bypass the need for python programming or server hosting by allowing users to fetch, map, and analyze real-time environmental data directly inside CODAP’s graphical interface.

This Jupyter notebook queries oceanographic datasets hosted on the National Oceanic and Atmospheric Administration’s (NOAA) CoastWatch ERDDAP servers. It streamlines the retrieval of sea surface temperatures (SST), sea level height deviations, and thermal expansion telemetry across custom coordinates and historical time spans.

This Jupyter notebook connects directly to the US Environmental Protection Agency’s Air Quality System (AQS) API. It allows users to programmatically extract air quality telemetry—including particulate matter (PM2.5 and PM10), ozone, nitrogen dioxide, and overall Air Quality Index (AQI) values—for any US county or monitoring station across custom time ranges.

This resource utilizes the EPA Toxic Release Inventory (TRI) database to illustrate data reduction strategies, showing how to filter spatial-environmental data down to localized neighborhoods or target chemicals. It provides structured examples for managing large, multi-attribute datasets without losing analytical value.

This single-page reference guide outlines six primary pedagogical strategies for preparing large datasets for classroom use: three case-reduction strategies (Random, Purposeful, and Build-Your-Own) and three attribute-reduction strategies (Thematic, Mathematical, and Question-Driven). For each strategy, the guide defines the method, evaluates its pedagogical benefits and drawbacks, and maps it to GAISE II and IDSSP educational standards.

This interactive data wrangling resource features a curated Billboard Hot 100 dataset along with guided workflows demonstrating how to filter, sort, and slice large, case-heavy data to fit specific lesson objectives. It helps educators learn and teach data wrangling techniques by focusing on sub-selections of music charts.