Git & GitHub for Data Analysts

1. Introduction

Version control records how analytical work changes over time. It makes experiments reviewable, supports collaboration, and provides a path back when a change damages a report or pipeline.

2. Core concepts

A repository contains tracked files and commit history. The working tree is your current edit, the staging area is the proposed next snapshot, and a commit records that snapshot. A branch supports isolated work; a remote such as GitHub synchronizes shared history but should never be used to publish secrets or private raw datasets.

3. Topics covered in this module

• Repository initialization, status, add, diff, commit, and log
• Branches, merges, pull requests, review comments, and conflict resolution
• .gitignore, remotes, push, pull, clone, stash, and GitHub Desktop
• Notebook diffs, data artifacts, reproducible environments, and secret hygiene

4. Syntax and how to read it

git status
git diff
git add analysis.py
git commit -m "Add monthly retention calculation"
git push

These commands inspect, stage, commit, and publish a change. Push only to a repository where you have permission, and check the remote before sending work.

5. Practical workflow

Check status, inspect the diff, stage only intended files, and write a commit message describing the analytical change. Before sharing, inspect tracked files for credentials, exports, and large generated outputs.

6. Common mistakes

• Committing a secret and assuming a later deletion removes it from history.
• Resolving a merge conflict by deleting code without understanding both changes.

7. Best practices

• Commit small, coherent changes and review the diff before each commit.
• Use synthetic or approved sample data; keep sensitive exports outside version control.

8. Practice task

You edited a notebook and accidentally created a large CSV export. What should you inspect before committing?

9. One possible solution

Run git status and git diff, add only intended source files, and add an appropriate ignore rule for generated output. Check whether the CSV was already tracked.

10. Summary

Track analytical code and documentation with commits, branches, reviews, remotes, and ignore rules. Commit small, coherent changes and review the diff before each commit. Use synthetic or approved sample data; keep sensitive exports outside version control.

11. What comes next

The next module, Web Development for Data Analysts (UI, Dashboards & Visualization), builds on this foundation and applies it to a new part of the analytics workflow.