Workshop Overview
Modern operations move fast; information moves faster. This workshop helps U.S. and NATO personnel turn raw data into clear, defensible insights using R and RStudio—without heavy programming. We focus on the parts of R that speed up decision support: clean data quickly, visualize clearly, summarize reliably, and produce repeatable reports.
No prior coding required. We use sanitized, unclassified defense-style datasets (logistics, readiness, geospatial context) and emphasize practical judgment, not buzzwords. Each day ends with Q&A and an optional office-hour block for extra help.
Learning Objectives
- Use R and RStudio to import, clean, join, and validate mission-relevant data.
- Apply the tidyverse to summarize, reshape, and visualize information quickly.
- Build clear charts and maps that support briefings and after-action reviews.
- Produce repeatable, parameterized R Markdown/Quarto reports.
- Run and interpret light-weight models (trend, regression, simple classification) for planning.
- Follow good practices for reproducibility, OPSEC, and handoff to teammates.
Prerequisites
- Familiarity with defense or government operations helpful
- Comfortable with spreadsheets; no prior coding required
Target Audience
- Analysts, planners, logisticians, intelligence personnel
- Defense contractors and federal program managers
- Technical staff seeking a fast on-ramp to R
Development Environment
- R (≥ 4.3) and RStudio (Posit Desktop)
- Packages:
tidyverse,readxl,janitor,lubridate,arrow,sf,ggplot2,tmaporleaflet,broom,knitr,rmarkdown,quarto - Offline-friendly docs; sanitized CSV/Parquet datasets provided
Agenda (5 Days)
Day 1 – Foundations & Setup
- Orientation (30m): What “decision support with R” means in practice; success criteria for your role.
- R & RStudio Tour (60m): Console, scripts, projects, file hygiene, working directories.
- Data In, Clean, and Ready (2h): Import CSV/Excel/Parquet; fix headers with
janitor; parse dates withlubridate; handle missingness. - Tidyverse Basics I (90m):
dplyrverbs, grouping/summarizing; quick quality checks to catch bad data early. - Mini-Lab (30m): Clean a logistics dataset and produce a 1-page EDA summary.
Day 2 – Wrangling Deep Dive & Visualization
- Tidyverse Basics II (90m): Joins, keys, cross-checks;
rowwisevs. vectorized ops; pitfalls and fixes. - Reshape & Compare (60m):
pivot_longer/wider; building tidy tables for side-by-side readiness comparisons. - Visual Basics with ggplot2 (90m): Lines, bars, densities; faceting; labeling and theme choices for briefs.
- Design for Decision Makers (60m): Clarity vs. complexity; encoding choices; avoiding chartjunk.
- Lab (30m): Turn wrangled tables into 2–3 briefing-quality charts.
Day 3 – Time, Trend, and Geospatial
- Time & Trend (75m): Time-based summaries, rolling means, simple change detection; communicating uncertainty.
- Geospatial Foundations (90m):
sfbasics, CRS/projections, simple choropleths; routes/areas (unclassified). - Operational Maps (60m):
tmapvs.leaflet; static vs. interactive choices for exercises/briefs. - Reporting I (45m): R Markdown/Quarto basics, structure, chunk options, inline values,
kabletables. - Mini-Exercise (30m): Add a time-series panel and a simple map to your report.
Day 4 – Reporting, Parameters & Modeling Lite
- Reporting II (75m): Parameters, child docs, reusable templates; producing HTML/PDF variants.
- Reproducibility & Handoff (45m): Projects, file structure, relative paths, minimal versioning (Git optional).
- Modeling Basics (75m): Linear/logistic regression; when simple beats complex; feature hygiene.
- Fit, Check, Explain (60m):
broomtidiers; confidence vs. prediction intervals; practical interpretation. - Scenario Workshop (45m): Build a parameterized brief that toggles between units/regions/time windows.
Day 5 – Capstone Build & Brief
- Predict then Plan Lab (60m): Example: predict delay risk; translate to an action option; sensitivity checks.
- Capstone Build (120m): Team project: data → wrangle → visuals (incl. map) → small model → parameterized report.
- Brief & Peer Review (60m): 5-minute briefs; peers stress-test assumptions and clarity.
- Next Steps (30m): Packaging your work, small pilots, sustainment.
Materials Provided
- Slides, step-by-step notebooks, sanitized datasets (CSV/Parquet)
- Cheat sheets (tidyverse, ggplot2, geospatial quickstart)
- Report templates (R Markdown/Quarto) and a small style guide
Security & Classification
All materials are unclassified; scenarios and data are sanitized for training.
Contact
To request dates or a private cohort: [email protected]