MATH/COSC 3570 Final Project Guidelines

Author

Dr. Cheng-Han Yu

Modified

April 23, 2026

This project is the final team assignment for the course. The goal is to bring together the skills you have practiced throughout the semester in one clear, thoughtful, and well organized written report.

A report template qmd file can be download here

A report template PDF file can be download here

NoteQuick Start
  • Work in teams of 3.
  • Submit one final written report per team.
  • Submit one single PDF that includes the main report, references, and appendix.
  • Use a real dataset and investigate one focused question.
  • Include data preparation, exploratory analysis, at least one course method, results, and limitations.
  • No presentation is required.

Learning Objectives

By completing this project, you should be able to:

  1. pose a focused question that can be investigated with data
  2. describe, clean, and prepare a real dataset for analysis
  3. use visualizations and summaries to explore patterns in data
  4. apply at least one method from the course appropriately
  5. interpret results carefully and communicate conclusions clearly
  6. discuss limitations, uncertainty, and possible improvements

Task Overview

Your team will:

  1. choose a dataset and a focused project question
  2. describe the data source, observational units, and key variables
  3. clean and prepare the data for analysis
  4. carry out exploratory analysis using tables and visualizations
  5. apply at least one course method
  6. interpret results in context
  7. write a clear final report

Project Scope

A strong project is usually focused and well executed, not overly ambitious.

Your project should:

  • investigate one main question
  • use one main dataset or two datasets that can be combined in a simple and well explained way
  • include at least one method from the course
  • remain at a level your team can explain clearly and honestly

You may compare more than one method, but this is optional.

WarningAvoid Overly Ambitious Projects

Avoid projects that depend on very advanced methods, very messy multi-source data collection, or a broad question that your team cannot answer well in the time available.

Data Requirements

You may use a public dataset or another dataset approved by the instructor.

Important

  • You may not use the dataset from your mini-projects.

  • Your dataset should be large enough and rich enough to support meaningful analysis.

In your report, clearly explain:

  • where the data came from
  • what the observational units are
  • what the important variables mean
  • any important limitations or quality concerns in the data

Required Report Components

Your final report should include all of the following.

1. Introduction

State the main question clearly and explain why it is worth investigating.

2. Data Description

Describe the data source, observational units, key variables, and any important data quality issues.

3. Data Preparation

Explain how the data were prepared for analysis. Depending on the project, this may include:

  • handling missing values
  • filtering observations
  • recoding variables
  • reshaping or joining data
  • creating new variables

4. Exploratory Analysis

Include informative summary statistics and visualizations that help the reader understand the data and the project question.

Each figure or table should be discussed in words. Do not include plots only to fill space.

5. Method

Use at least one method from the course in a meaningful way. Possible choices include:

  • multiple linear regression
  • logistic regression
  • decision trees
  • k-nearest neighbors
  • principal component analysis
  • k-means clustering

You are welcome to use another method, but please get approval from Dr. Yu first.

6. Results

Explain what your analysis found. Interpret results in context using clear language.

Your report should show that you understand the difference between:

  • prediction and explanation
  • association and causation
  • statistical patterns and strong conclusions

7. Limitations

Discuss important limitations of your analysis. Possible issues include:

  • data quality
  • missing information
  • possible bias
  • modeling assumptions
  • limited scope of conclusions
  • what could be improved with more time or better data

8. Conclusion

Summarize the main takeaway of the project in plain language.

9. References

List the dataset source and any other references you used.

10. Appendix

Include the following in an appendix:

  • AI use statement
  • team contribution statement

Report Expectations

Your report should:

  • be clearly written and well organized
  • include a title and all team member names
  • include at least 3 informative figures
  • include at least 1 summary table
  • use at least 1 course method or a method approved by Dr. Yu
  • explain results in complete sentences, not only code or output
  • be approximately 5 to 8 pages, not counting references and appendix

A shorter report is acceptable if it is complete, clear, and substantive. A longer report is not automatically better.

AI Use Statement

AI use is allowed, but your team remains responsible for the accuracy, quality, and clarity of the final report.

Include a short AI use statement in the appendix. This statement should do one of the following:

  • state that your team did not use AI tools for the project, or
  • briefly explain which tool or tools were used, what they were used for, and how your team checked or revised the output

This is included for transparency and professional practice. It is not a major separate grading category, but missing or unclear AI disclosure may lower the completeness portion of the project grade.

Team Contribution Statement

Include a short team contribution statement in the appendix.

This statement should:

  • list the name of each team member
  • briefly describe each member’s main contributions
  • mention how the team divided the work and reviewed the final report

A simple format is fine. For example:

  • Student A: data cleaning, exploratory plots, part of the writing
  • Student B: modeling, results interpretation, revision
  • Student C: data description, conclusion, final editing and formatting

If contributions were not fully equal, state that honestly and professionally. In rare cases, individual grades may be adjusted if the contribution statement or other evidence shows major differences in participation.

Submission

  • Submit one single PDF per team.
  • The PDF should include the main report, references, and appendix.
  • No presentation is required.
  • Submit the report by the posted course deadline.

Evaluation Criteria

Your project will be evaluated based on:

  • clarity and focus of the project question
  • quality of data understanding and preparation
  • usefulness and interpretation of exploratory analysis
  • appropriateness and correctness of the method
  • quality of interpretation and discussion of limitations
  • organization, completeness, and writing quality of the final report

Rubric Summary

Criterion Points
Project question, data understanding, and preparation 25
Exploratory analysis and methods 35
Results, interpretation, and limitations 30
Writing, completeness, and professional practice 10
Total 100

Grading Rubric

Criterion Excellent Good Needs Improvement
Project question, data understanding, and preparation (25 points) The question is clear and focused. The data are well described. Preparation steps are appropriate and clearly explained. The question and data are mostly clear, with some minor gaps in focus, explanation, or preparation. The question is vague, the data are poorly explained, or preparation is incomplete or unclear.
Exploratory analysis and methods (35 points) Visualizations and summaries are informative and clearly connected to the question. The method is appropriate and correctly used. Exploratory work and methods are generally reasonable, but some choices or explanations could be stronger. Exploratory analysis is weak, limited, or poorly interpreted, or the method is inappropriate or incorrectly used.
Results, interpretation, and limitations (30 points) Results are interpreted carefully and accurately. Conclusions are well supported. Limitations are honest and thoughtful. Results are mostly interpreted correctly, though some explanations may be brief or somewhat overstated. Limitations are present. Results are unclear, overstated, or weakly supported. Limitations are missing or very limited.
Writing, completeness, and professional practice (10 points) Report is well organized, clear, and polished. Required components are included, including appendix materials such as the AI use statement and team contribution statement. Report is generally clear and complete, with minor problems in organization, writing, or required components. Report is difficult to follow, missing important required components, or lacks clear appendix documentation.

Submission Checklist

Before submitting, make sure your report:

  • states a clear question
  • explains the dataset clearly
  • shows meaningful data cleaning and preparation
  • includes at least 3 informative figures
  • includes at least 1 summary table
  • uses at least 1 course method
  • interprets results carefully
  • discusses limitations honestly
  • includes an AI use statement
  • includes a team contribution statement
  • is submitted as one single PDF