Choosing Opening Hours from Data is a practical pandas and Matplotlib project for learners who can already read, filter and group data. Help a community venue compare its opening slots fairly and produce reusable Python analysis, complete CSV tables, a real chart and an evidence-qualified recommendation.
Two Guided Labs move from a hand-calculated equal-opportunity example to missing observations, comparable rates and an explicit evidence rule. One independent Challenge transfers the same methods to another venue. Each unit starts with its own prepared data and workspace.
What You Will Learn
After completing this course, you will be able to:
- match an opening plan to observations using complete, unique date-and-slot keys;
- distinguish closed hours, missing recordings and genuine zero visitors;
- calculate planned hours, observed hours, recording coverage and visitors per observed hour;
- apply public eligibility boundaries, keep all tied leaders and defer when evidence is insufficient;
- deliver reproducible CSV tables and total-versus-rate charts with visible observation counts and coverage;
- explain why descriptive historical rates do not establish the effect of extending opening hours.
Who This Course Is For
This course is for beginning data analysts who want to turn familiar grouping and plotting tools into a careful practical decision.
Prerequisites: Basic Python, terminal input, pandas CSV reading, filtering and grouping, and Matplotlib bar charts. The course teaches the required composite-key merge, missing-data distinctions and decision rules.
Learning environment: A prepared browser-based WebIDE with Python, pandas and Matplotlib. Data is synthetic and the runtime labels and outputs are in English. No local installation is required.
Frequently Asked Questions
Do I need machine learning or statistics?
No. You will make a descriptive comparison, not train a model or perform a significance test.
Must the result recommend one winning slot?
No. Correct results retain all ties and may defer when no slot meets the stated evidence rule. The thresholds are exercise rules, not universal confidence guarantees.
What will I keep?
Reusable scripts, complete hourly and slot tables, a real PNG chart and short advice, with original inputs preserved. Each unit is independently initialized; previous VM files are unnecessary.
