Cleaning and Summarizing Data with pandas (Python 2)

Move beyond the basics and start analyzing real-world research data.

This 2-hour, hands-on workshop is designed for researchers, students, and faculty who have a basic grasp of Python and are ready to tackle the "messy" side of data science. Transitioning from basic scripts & notebooks (Python 1) to a professional local development workflow, we will use Positron—the new data science IDE from the creators of RStudio—to manage a complete analysis pipeline.

Using the OASIS-1 neuroscience dataset, we will work through the practical steps of transforming raw MRI demographics into research insights. By the end of the session, we will answer a specific clinical question: Does normalized brain volume differ by dementia rating?

By the end of this session, learners will be able to:

1. Establish a reproducible research workflow by setting up a local environment with Positron and virtual environments
2. Load tabular data into a pandas DataFrame and inspect its structure using info(), head(), describe(), and related methods
3. Select specific rows, columns, and subsets of data using bracket notation and .loc[]
4. Clean a real-world dataset by handling missing values (dropna(), fillna()), renaming columns, dropping unnecessary columns, and converting data types
5. Summarize data by grouping and aggregating with groupby() to answer a specific research question
6. Visualize data distributions and relationships using seaborn (if time permits)