Data
The datasets used in the workshop
All datasets are fictional and were created for teaching. They come with the workshop folder (see Setup) — you don’t need to download them separately.
The workshop dataset: an outbreak linelist
A linelist is a table with one row per case, the standard format for outbreak data. The instructor uses this dataset in the 👀 Follow along parts.
| File | What it is | Used in |
|---|---|---|
raw_data/linelist_raw.xlsx |
6,604 cases with dates, demographics, hospital, and outcome | Sessions 1–2 |
raw_data/case_info.csv |
Symptoms and temperature for each case | Session 3 (joined to the linelist) |
raw_data/hosp_info.txt |
Look-up table of the correct hospital names | Session 2 |
Problems built into the raw data (you find them in Session 1 and fix them in Sessions 2–3):
- Column names with spaces, dots, and capitals
- Dates stored as different types; ages stored as text, some in months
- Misspelled and inconsistent hospital names (“Hopital”, “Regional MC”)
- Impossible weights and heights
- 130 rows with no case ID, and 25 duplicate rows
The practice dataset: a measles outbreak
You use this dataset in every ✏️ TRY IT. It has the same kinds of problems as the workshop dataset, so every skill transfers directly.
| File | What it is | Used in |
|---|---|---|
raw_data/measles_raw.csv |
1,215 measles cases in 2024 across five districts | Sessions 1–8 |
raw_data/district_info.csv |
Population and number of clinics for each district | Session 3 (joined), Sessions 5 and 7 (case rates) |
Problems built into the practice data:
- Column names with spaces and symbols (
Temp (C),Vaccinated?) - Age stored as text, with “unknown” and the code 999
- District names misspelled (“Nroth”, “Cetnral”) and capitalized inconsistently
- Sex written six ways and vaccination status written eight ways
- Some temperatures entered in Fahrenheit instead of Celsius
- 37 rows with no patient ID, and 15 duplicate rows
Files made during the sessions
As you work through the lessons, you save cleaned versions in processed_data/. Ready-made copies are included, so you can start any session even if you missed the one before.
| File | Made in | Used in |
|---|---|---|
processed_data/linelist_clean.rds, measles_clean.rds |
Session 2 | Session 3 |
processed_data/combined_linelist.rds, measles_final.rds |
Session 3 | Sessions 4–8 |
.rds?
.rds is R’s own file format. It keeps dates as dates and missing values as NA, so it’s the best choice for files you will open in R again. .csv files are better for sharing with other software, but missing text comes back as blank "" instead of NA. Session 3 explains this in detail.
Bring your own data
Every lesson ends with a ⭐ BONUS for your own data. Put your file in raw_data/ and work through the same steps. Please only use data you are allowed to share with others in the room, and never bring identifiable patient data to a workshop.