Final Project
The final project is an open-ended applied data analysis project. Students choose a practical question, find a real dataset, and use statistical tools from INF 659 to produce an evidence-based conclusion. The emphasis is on asking one good question and answering it well.
What every project needs
- One focused primary question
- One documented real-world dataset
- At least one substantial inferential method from the second half of the course
- At least two labeled visualizations
- A clear limitations discussion
Good applied directions
- Comparing groups
- Explaining or predicting an outcome
- Working with categories and proportions
- Operations, risk, and uncertainty decisions
Deliverables
- Proposal
- Checkpoint
- Final notebook plus report or annotated slides
- Presentation of about 5 minutes plus questions
Submission method
- Email project materials directly to rongyu.lin@quinnipiac.edu
- Use a clear subject line such as INF 659 Final Project - Your Name
- If files are too large, include a shareable cloud link in the email
Applied focus: most projects should use a real dataset and aim at a practical interpretation. Kaggle is a good place to start looking for real datasets, and you may also use other reliable public sources. A narrow, well-executed analysis is better than an ambitious but unfinished project.
Spring 2026 Timeline
| Milestone | Purpose | Target Date |
|---|---|---|
| Proposal | Confirm the question, dataset, and planned methods before the project grows too broad. Submit by email. | Fri, Apr 24 |
| Checkpoint | Show cleaned data, one draft figure, and one preliminary result. Submit by email. | Mon, Apr 27 |
| Presentation | Share the question, evidence, conclusion, and limitation with the class. | Wed, Apr 29 |
| Final Submission | Email the polished notebook plus report or annotated slides to the instructor. | Wed, May 6 |
Grading Emphasis
| Criterion | Points | What strong work looks like |
|---|---|---|
| Question and scope | 10 | A focused, meaningful, course-appropriate question. |
| Data quality and documentation | 15 | Clear source, understandable variables, and a dataset that supports the analysis. |
| Method choice and design | 25 | Methods fit the question and are explained rather than only applied. |
| Accuracy and interpretation | 20 | Careful reasoning, correct calculations, and conclusions that do not overclaim. |
| Communication and organization | 15 | A coherent notebook, readable figures, and a clear story. |
| Reproducibility and professionalism | 5 | Reasonable notebook structure and complete references. |
| Presentation | 10 | A concise talk centered on evidence and conclusions, not just code. |