Python for Data Analysis

Learn to answer real questions with data in Python: load spreadsheets, CSV and JSON files into pandas tables, clean messy data, filter, group, join and…

Updated

Module 1: Notebooks and Tables

Set up a data workspace with pandas and notebooks in VS Code, brush up on the Python that data work uses, then load a farmers market's sales, its Excel workbook and its app's JSON into pandas tables and find out what's in them.

Module 2: Cleaning Data

Clean real, messy data: find and handle missing values honestly, fix wrong types, remove duplicates, tidy text and parse dates, catch values that can't be right, and put every step into one cleaning function that saves a clean file, then clean a year of bike-share rides.

Module 3: Answering Questions

Answer real questions with data: pick rows and columns, filter and sort, add calculated columns, group and summarize, build pivot tables, total by week and month, join tables and reshape them, then answer five questions a coffee chain's manager asked.

Module 4: Charts

Turn answers into charts people can read: draw line, bar, scatter and histogram charts with matplotlib and pandas, choose the chart that fits the question, write titles that state the finding, keep axes honest and colors readable, and lay several charts out on one saved page.

Module 5: A Report Someone Can Use

Finish an analysis so someone else can use it: a notebook that tells a story with its caveats, shared as a web page; logic moved into a module and a script that rebuilds the report from new data; and tests that prove the numbers and charts are right, then build a monthly report that rebuilds itself.