Web page · The Carpentries
TrainingData Carpentry: R for Social Scientists
A two-day, hands-on lesson that takes someone from no R at all to loading a survey dataset, cleaning it with dplyr and producing publishable charts with ggplot2.
Published November 6, 2025
- Audience
- Analysts Program staff
- Access
- Freely available
What do these mean?
- Analysts
- Research, data and evaluation staff.
- Program staff
- The people running the service day to day.
- Freely available
- Open on the public web, no sign-in.
What it is
An open, self-paced lesson built for people whose day job is not programming. It uses one survey dataset the whole way through, so nothing is a toy example, and it covers the sequence that actually matters: getting data in, reshaping it, summarising by group, plotting, and exporting something you can hand to someone else.
Who it is for
Staff who currently do their analysis in a spreadsheet and have hit the wall where the same clean-up has to happen every month. It assumes no prior programming and does not pretend otherwise.
How to use it
Two full days works better than four half-days — the material builds, and the gap between sessions costs more than it saves. Running it internally with a facilitator, rather than sending people to it alone, roughly doubles the completion rate in our experience.
Notes from the library
The lesson materials are openly licensed, so you can fork the episodes and swap in one of your own datasets. That is the single change that has made the biggest difference when we have run it: people finish the day with a script that works on data they recognise.
About
- Year published
- 2025
- Contact email
- Fictional contact — demo only
Who it is for
- Topics
- R
- training
- data analysis
- reproducibility
- Language
- English