Computational and AI-Assisted Methods

Experiential Learning (EL) designation: approval pending. Build one sustained, mentored computational research project on a topic you choose, with regular feedback, critical reflection, and a final product for readers outside your field. How the course aligns with the EL framework →

Turn a real question into usable evidence

How does a field frame a policy problem? Who works with whom on it? Which published claims rest on measures that hold up? In P539, you take one question like these through a complete computational pipeline, from gathering the records to sharing an analysis someone else can inspect and rerun.

You will collect and organize data, translate a concept into measures, apply network analysis or machine learning for a stated purpose, and communicate what the results support. AI can help you write code, classify documents, and critique a design. You remain responsible for deciding whether the method fits, checking its output, and explaining the result.

The course is designed for MPA, MPP, and doctoral students. College-level statistics is required. Basic familiarity with R or Python is recommended. AI assistance and setup support are provided for the practical work. Bring a laptop and a policy, management, or research topic you would like to investigate.

One project, built in four stages

Everyone builds the same kind of project, on a topic of their own choosing. You gather a corpus of records about your topic, by default the scholarly literature indexed in OpenAlex, and carry it through the four stages of research design that organize the required text: data management, concept representation, data analysis, and scientific communication. Each stage ends in a milestone, and each milestone is made of the deliverables from the book’s exercises.

By the end you have a documented corpus, measures you have audited, an analysis with a stated purpose, and an artifact that a reader outside your field can use, all in a project folder a classmate can run. How to choose a topic and plan the project →

What you will learn

By the end of the course, you should be able to:

1. Frame a feasible question and set up a project environment that an AI agent and a human reader can both follow.
2. Gather, structure, and document data so another person can understand where it came from and what it cannot show.
3. Turn a concept into measures, including measures produced by automated text coding, and audit their validity and reliability.
4. Choose a computational method for a stated purpose, assess it against a baseline, and explain what the result supports.
5. Communicate evidence and its uncertainty to a non-specialist audience, with materials that make the work inspectable and reproducible.
6. Explain how feedback, ethical choices, and the experience of building the project changed your research or professional judgment.

How the course uses the book

The required text is Ji Ma’s Computational & AI-Assisted Methods for Social Sciences: A Research Design Primer (SAGE, forthcoming). Assigned chapters are posted in Canvas. You do not need to buy anything. The book’s free companion website has the setup guide, technique primers, tool guides, worked notebooks, and a database of published studies that the exercises draw on.

The book pairs each theory chapter with an exercise chapter, and the course follows that pairing. Chapters 3, 5, 7, and 9 build the reasoning for a stage and close with a Practice section on published studies. Chapters 4, 6, 8, and 10 put the same reasoning to work on your own data, one deliverable at a time. Chapters 1 and 2 set up the project, Chapter 11 reviews its ethics, and Chapter 12 draws the stages together. Chapters, companion materials, and setup →

The book has more exercises than one semester of six-hour weeks can hold. P539 requires a core set, gives you a choice within Chapter 8, and treats the rest as optional extensions. The assignments page lists exactly which exercises count and what each one produces.

How the course fulfills experiential learning

Experiential Learning (EL) designation: approval pending. The course is designed for Applied experiential learning through a sustained, mentored project. Its activity type is project-based learning: you investigate a question whose answer is not known in advance and revise your work as the evidence comes in. You demonstrate the three outcomes in IU’s EL framework: Authenticity, Career Focus, and Meaning Making.

The instructor gives feedback on every milestone. Four in-class exercise reviews pair you with a classmate who tries to follow or reproduce your work before you submit it. Three individual reflections explain what you learned and how it affects your future practice, and the final presentation demonstrates the result. EL outcomes, activities, and assessment evidence →

How class works

Each stage runs for about three weeks. We begin with the theory chapter and a short student-led discussion of one published study, using the chapter’s Practice section. We then work through the paired exercise chapter in class and outside it. The stage closes with an in-class exercise review, and the milestone is due after it, usually a week later.

Monday meetings usually develop the reasoning or a worked example; Wednesday meetings emphasize practice and feedback. The schedule gives the arrangement for every meeting. The practice guide explains how to read a study, how to work an exercise, and how the reviews run.

You organize the project with the Rules/Memory/Knowledge structure from Chapter 2, which also serves as your record of AI use. Using AI →

Workload and project scope

This is a 3-credit course. Plan for up to 6 clock hours of required work outside class per teaching week, in addition to the scheduled meetings. The cap includes reading, discussion preparation, the exercises, coding and troubleshooting, writing, documentation, and reflections.

The weekly guides divide that budget among preparation, applied work, and writing. In milestone weeks, assembling the submission and writing the reflection take time otherwise allocated to reading or new exercises. Discussion leaders prepare their opening in place of the half-page Practice note.

Keep the project within the cap by keeping it narrow: one question, one concept to measure, a corpus of a few hundred to a few thousand records, and the required exercises. Preserve the quality of the evidence by narrowing the scope, not by skipping the checks. Optional exercises do not raise the grading expectations.

If required work is approaching the cap with important tasks unfinished, save the attempt and contact the instructor. We will prioritize tasks, reduce scope, or arrange support within the budget. Time spent is not graded, and faster completion does not create an obligation to add work. Spring break has no required work. Finals preparation also has a six-hour cap; the artifact and documentation are completed before that week.

This workload is consistent with IU’s credit-hour standard, which sets a three-credit benchmark of 150 instructional minutes and at least 300 outside-class minutes per week. Our ordinary meetings total 150 minutes, and the outside-work budget is 360 minutes. These are planning estimates; actual time varies by student and task.

Assessment

Your grade comes from project milestones (80 points) and individual reflections (20 points). Practice notes, discussion leadership, and the exercise reviews prepare you for those submissions and carry no separate points.

Submit graded work in Canvas. See the letter-grade scale and late-work policy: 3 late passes, each allowing 72 hours for an eligible interim submission.

What to submit for each milestone →

Before the first meeting

• Read Your project and bring one or two topics you would like to study.
• Read the setup guide and plan to try it before Wednesday’s workshop. Bring any setup difficulties; Monday’s introduction requires no technical preparation.
• Read Using AI. Chapters 1 and 2 will be in Canvas.

Office hours: by appointment; email the instructor to arrange a meeting. Contact Ji Ma with preparation or access questions. The Policies and support page links to university resources.