Last Class

Capstone Progress Check, Course Reflection, and Wrap-Up

Byeong-Hak Choe
bchoe@geneseo.edu

SUNY Geneseo

May 6, 2026

✅ Today’s Goals

What we will do today

  • 🔎 Check your capstone project progress
  • 🛠️ Identify remaining issues before final submission
  • 📑 Review final paper and website expectations
  • 🤖 Reflect on data analytics in the era of AI and agentic AI
  • 📝 Complete the Student Course Experience (SCE) survey
  • 🙌 Wrap up the semester

Part 1: Capstone Progress Check

Where is your project now?

For each student, please briefly share your current progress.

Suggested update format:

  1. Project topic: What is your capstone about?
  2. Current status: What have you completed so far?
  3. Main result: What is one key finding or tentative finding?
  4. Remaining task: What still needs to be completed?
  5. Biggest challenge: What issue are you currently facing?

Quick progress categories

Which category best describes your current project?

  • 🟢 Mostly complete: Writing, polishing, and checking submission materials
  • 🟡 Almost there: Results are mostly ready, but interpretation or formatting still needs work
  • 🟠 Still developing: Analysis is running, but results are not fully finalized
  • 🔴 Need help soon: Major issue with data, code, model, Quarto, or GitHub website

Common last-stage issues

Near the end of a capstone project, common issues include:

  • broken file paths in Quarto
  • figures or tables not rendering correctly
  • unclear interpretation of model results
  • too much code and not enough explanation
  • weak connection between research question and results
  • missing limitations or future work section
  • GitHub website not updating properly
  • final files not organized for submission

Part 2: Final Submission Checklist

Final capstone paper checklist

Before submitting, check whether your paper clearly includes:

  • Introduction: What question are you studying, and why does it matter?
  • Background / motivation: What context helps readers understand your project?
  • Data: Where did the data come from? What variables are used?
  • Methods: What statistical, econometric, or machine learning methods are used?
  • Results: What did you find?
  • Discussion: What do the results mean?
  • Limitations: What should readers be careful about?
  • Conclusion: What is the main takeaway?

Final technical checklist

Please check:

  • The rendered HTML webpage opens correctly.
  • All figures and tables appear correctly.
  • Your .qmd file is submitted.
  • Your code is included either inside the .qmd file or in separate script files.
  • Your data file or data source link is included.
  • File paths are relative when possible.
  • Your GitHub website link works.
  • The project is understandable to someone outside this class.

What makes a strong capstone project?

A strong capstone project does not have to be perfect.

It should show that you can:

  • ask a meaningful question,
  • work with real data from a credible source,
  • make reasonable analytical decisions,
  • communicate results clearly,
  • explain limitations honestly,
  • and produce a reproducible project.

Part 3: Reflection on the Analytics Workflow

Real data work is messy

In real-world data analytics, the hardest part is often not the final model.

The hard parts are usually:

  • finding usable data,
  • defining a good question,
  • cleaning messy data,
  • deciding what comparison is meaningful,
  • interpreting imperfect results,
  • and clearly communicating the limitations and implications of the results.

What would you improve with more time?

Think about your own project.

If you had one more month, what would you improve?

  • Better data?
  • More observations?
  • A stronger research design?
  • A better model?
  • Better visualization?
  • More robustness checks?
  • A clearer webpage?
  • A more polished paper?

Portfolio mindset

Your capstone project can become part of your professional portfolio.

A good portfolio project should help you explain:

  • what problem you studied,
  • what data you used,
  • what tools you applied,
  • what result you found,
  • what decisions you made,
  • and what you learned.

Part 4: Data Analytics in the Era of AI

Does data analytics still matter when AI can write code?

Yes.

AI can help with coding, debugging, visualization, writing, and brainstorming.

But AI does not replace the need to understand the analytics workflow.

Why fundamentals still matter

Even in the era of generative AI and agentic AI, you still need to understand:

  • whether the data are appropriate for the question,
  • whether the model is reasonable,
  • whether the results make sense,
  • whether the interpretation is too strong,
  • whether the analysis is reproducible,
  • and whether the final conclusion is useful.

AI can assist, but you must judge

AI may produce code, plots, summaries, or model suggestions.

However, the analyst is still responsible for:

  • checking data quality,
  • verifying code output,
  • choosing appropriate methods,
  • interpreting results carefully,
  • identifying limitations,
  • and communicating findings.

The human role in analytics

The most valuable analysts are not just people who can run code.

They are people who can combine:

  • domain knowledge,
  • statistical reasoning,
  • technical skill,
  • communication,
  • curiosity,
  • and good judgment.

✅ Course Reflection & Wrap-Up

📘 What We Worked On This Semester

  • 🧠 Research Design and Analytical Thinking
    • Developing research questions, motivation, and empirical strategies
    • Applying econometric and machine learning methods to real-world data
  • 🧹 Data Collection, Cleaning, and Wrangling
    • Finding data sources, preparing variables, cleaning datasets, and organizing analysis-ready data
  • 🗣️ Professional Communication and Presentation
    • Presenting research through the Midterm Presentation and GREAT Day Presentation
  • 📑 Capstone Research Paper
    • Completing a full empirical research project from research question to data, analysis, interpretation, and final paper

🧭 What You Practiced as Data Analysts

  • Turning a broad topic into a specific analytical question
  • Making practical decisions with imperfect data
  • Using statistical, econometric, or machine learning tools
  • Creating tables, figures, and written explanations
  • Presenting results to an audience
  • Building a reproducible project website

💬 Reflection Questions

Take a moment to think about these questions:

  • What part of your capstone project are you most proud of?
  • What was the hardest part of the process?
  • What is one technical skill you improved?
  • What is one thing you would do differently next time?
  • How might this project help you in your future academic or career path?

📝 Student Course Experience (SCE) Survey

Your Feedback Matters

I have made every effort to create a meaningful and supportive learning experience in this course.

Your feedback is extremely valuable and helps improve future versions of the course.

I sincerely encourage you to complete the Student Course Experience (SCE) survey.

✅ Please Take About 10 Minutes Now

🔗 https://go.geneseo.edu/scesurveys

Steps

  1. Log in to the SCE Survey Portal
  2. Click on “Surveys”
  3. Select DANL 410: Data Analytics Capstone
  4. Complete the evaluation survey

🙌 Thank You

Thank you for being part of DANL 410 this semester.

I really appreciate the effort you put into developing your own research ideas, working with real-world data, and completing independent analytics projects.

I hope this course helped you become more confident in conducting empirical research from start to finish, including data collection, modeling, interpretation, and professional communication of results.

Thank you again for your participation, feedback, and hard work throughout the semester. It has been a pleasure.

— Byeong-Hak