EPID 785R

Data

The datasets used in the course lecture notes are available here. Data for the labs (the raw eICU demo tables and the Lab 1 analytic dataset) ships inside each lab's starter bundle on the Labs page.

Course datasets

kidneys.csv
Kidney length, age, and infection status for 350 kidneys in 180 young children; Example 1 of Carlin & Moreno-Betancur (2025), used throughout the regression weeks. The first line is a source URL, so read with skip = 1.
CC BY 4.0. Carlin (2024), University of Melbourne, doi:10.26188/26973436.v1.
nhefs.csv
NHANES I Epidemiologic Followup Study extract used in Hernán & Robins, Causal Inference: What If. Used in the Week 2 notes.
US public-use survey data; freely redistributable for teaching.
NHEFS_Codebook.xls
Variable codebook accompanying nhefs.csv.
As above.
cluster_trial_data_bmi.csv
Twenty patients from a cluster randomized trial example: BMI at one year (kg/m²), treatment arm (1/0), and clinical practice (10 practices, five per arm). The data are Table 1 of Mansournia et al. (2021), the Week 7 reading; the Week 7 notes use them for model-based, robust, and cluster-robust standard errors. The file begins with a byte-order mark, which readr::read_csv() ignores; base read.csv() needs fileEncoding = 'UTF-8-BOM'.
Transcribed from Table 1 of Mansournia MA et al. (2021), Int J Epidemiol 50(1):346–351, doi:10.1093/ije/dyaa260. Cite the article when you use the data.

kidneys.csv attribution. Carlin JB (2024). Dataset on kidney lengths in young children for teaching purposes. University of Melbourne. doi:10.26188/26973436.v1 (CC BY 4.0).

nhefs.csv and NHEFS_Codebook.xls attribution. The NHEFS extract and its codebook accompany Hernán MA and Robins JM, Causal Inference: What If (Boca Raton: Chapman & Hall/CRC, 2020), and were obtained from the book's website, which also hosts the full text and related materials.

cluster_trial_data_bmi.csv attribution. The twenty observations are Table 1 of Mansournia MA, Nazemipour M, Naimi AI, Collins GS, Campbell MJ (2021). Reflection on modern methods: demystifying robust standard errors for epidemiologists. International Journal of Epidemiology 50(1):346–351, doi:10.1093/ije/dyaa260. The article takes the example from Campbell MJ, Statistics at Square Two: Understanding Modern Statistical Applications in Medicine (Blackwell, 2006). Please cite the article when you use the data.