CMU S&DS Data Repository

The Data Repository curates interesting datasets for use in statistics and data science education. Each dataset is supported by a story describing its origin and application, and a set of interesting questions that can be answered using the data. This means:

• Every dataset has context in a scientific field, pop culture, or daily life.
• Beyond context, datasets are interesting. They feature more than just a dozen observations from an antiquated scientific study — many feature thousands of observations of dozens of variables, and answer questions interesting to a wide audience.
• Just like in science, some datasets give null results.
• Instructors can easily build lessons and assignments from the suggested questions.

Datasets are organized by broad subject areas on the left, or you can browse a sortable list of all datasets.