Final Project

Project Summary¶

Students will design an agent-based model of social norms—how they emerge, spread, and break down in a community. The project asks: How do shared norms of cooperation, fairness, or trust develop in societies, and what forces destabilize them?

This model invites students to blend technical ABM design with social theory concepts like Durkheim’s collective conscience, Goffman’s micro-interactions, or Elinor Ostrom’s work on commons governance.

Project Overview: Six Steps¶

Step 1: Brainstorming (Nov 6)¶

What Students Should Do¶

• Form teams of 3–4 students.

• Brainstorm social phenomena that involve norm emergence, spread, or breakdown.

• Consider phenomena from:

  • Everyday life (e.g., campus culture, study group dynamics, online communities)

  • Historical/social contexts (e.g., protest movements, cultural shifts, organizational behavior)

  • Theory from the course (e.g., Durkheim’s collective conscience, Ostrom’s commons governance, Goffman’s micro-interactions)

Expectations for Step 1¶

• Scope: Ideas should be specific enough to model but flexible enough to simplify where needed.

• Theory link: Teams should identify at least one social theorist whose framework will guide their model.

• Feasibility: The model should be buildable in 4–6 weeks using NetLogo.

Guidance Questions¶

• What question about norms are you trying to answer?

• Who are the agents in your system, and what are they trying to do?

• How do agents interact, and what causes norms to emerge or change?

• Under what conditions might norms become fragile or collapse?

Answer these questions and submit them as a one-page (500-words maximum) brainstorming document by Nov 9.

Step 2: Idea Pitch (Nov 11)¶

Presentation Content¶

Each team delivers a 5–10 minute presentation covering:

1. The Research Question (1–2 min)

   • What social phenomenon are you modeling?

   • Why does it matter?

2. Initial Concept (2–3 min)

   • What agents are in your system?

   • What are the key behaviors or interactions?

   • What could cause norms to stabilize or break down?

3. Theory Connection (1–2 min)

   • Which social theorist(s) inform your model?

   • How will their ideas guide your modeling choices?

4. First Prototype Sketch (1 min)

   • Pseudocode, flowchart, or diagram showing model logic

Expectations for Step 2¶

• Teams should demonstrate clear thinking about their problem space.

• Presentations should show connection to course material (readings, discussions, prior modules).

• Teams should be open to feedback and ready to refine their ideas.

Submit your slides presentation on Moodle .

Pitch Deliverables¶

• Slides by November 11 before class

• Peer feedback form (provided by instructors)

• Instructors feedback on feasibility and scope

Step 3: Design & Literature Review (Nov 13–20)¶

Work for Step 3¶

After receiving feedback, teams move into design and research:

1. Structure Your Model

   • Define agent properties (attributes, initial states)

   • Outline agent behaviors (rules, decision logic)

   • Describe interactions (how agents affect each other)

   • Specify parameters and environment (landscape, resources, etc.)

   • Identify possible “shocks” or perturbations to test norm fragility

2. Deep Dive into Social Theory

   • Re-read relevant theorist(s) with your model in mind

   • Extract key concepts and translate them into model rules

   • Document why each modeling choice connects to theory

   • Consider critiques: what does your model capture? What does it miss?

3. Literature Review

   • Find 2–3 empirical or computational papers related to your phenomenon

   • Summarize relevant findings and modeling approaches

   • Identify where your model builds on, differs from, or critiques prior work

Design & Literature Deliverables¶

• Pseudocode or flowchart showing model logic (pseudocode + conceptual diagram) - one page

• Theory memo (1–2 pages) mapping theoretical concepts to model rules

• Parameter table specifying all model inputs and their ranges

• Annotated bibliography (at least 3 sources: 1–2 theory + 1–2 empirical)

• Literature Review Summary (1–2 pages, shared with class in SRG-style discussion)

These deliverables are due by Nov 21 and will guide your coding in Step 4.

Step 4: Lab Sessions (~4 sessions, Nov 25–Dec 4)¶

Work for Step 4¶

Teams move into hands-on coding and model building:

Lab 1 (Nov 25): Model Setup & Basic Behavior

• Implement agent initialization and basic properties

• Code one or two core behaviors (e.g., agents observe neighbors, copy behavior)

• Test that agents move, interact, or update correctly

Lab 2 (Dec 2): Norm Emergence & Measurement

• Implement norm-tracking mechanisms (e.g., What % of agents follow a rule?)

• Add measurement/plotting of norm strength over time

• Run test simulations and check for expected patterns

Lab 3 (Nov 25–Dec 4): Shocks & Robustness

• Implement “shock” or perturbation (e.g., resource depletion, new agents entering)

• Run parameter sweeps to test sensitivity

• Analyze which conditions lead to norm stability vs. collapse

Lab 4 (Dec 2–4): Refinement & Final Tweaks

• Fix bugs, optimize code for clarity

• Run final simulation set

• Prepare data/visualizations for presentation

Expectations for Step 4¶

• Working code (NetLogo, Python Mesa, or agreed-upon platform)

• Clean, commented code with clear variable names and logic

• Simulation outputs (plots, statistics, example runs)

• Lab notes (brief write-up of what worked, what didn’t, what you learned)

In-Class Support¶

• Instructors and peer teams available for debugging and design discussion

• Time for peer code review and feedback

• Access to example models and reference code as needed

Lab Deliverables¶

• Working model (code repository or file)

• Sample output (plots, screenshots, video of simulation)

• Lab journal (brief notes from each session)

To be submitted on December 9 before class.

Step 5: Model Presentation (Dec 9–11)¶

What to Present¶

Each team gives a 10–15 minute presentation demonstrating:

1. Live Model Demonstration (3–5 min)

   • Show the model running with different parameter settings

   • Highlight key behaviors and norm dynamics

   • Explain what the visuals/outputs mean

2. Key Findings (2–3 min)

   • What surprised you?

   • Under what conditions do norms emerge or collapse?

   • How do your results connect to theory?

3. Reflection on Modeling (1–2 min)

   • What did you learn about the phenomenon by building this model?

   • What are the limits of your model? What did you have to simplify?

   • How might this model inform our understanding of real-world challenges?

4. Q&A (1–2 min)

   • Peers and instructors ask clarifying questions

Expectations for Step 5¶

• Clear, polished presentation (slides + live demo)

• Honest discussion of limitations (a hallmark of good science)

• Evidence of iterative refinement (you learned and improved along the way)

• Enthusiasm and engagement with the material

Presentation Deliverables¶

• Presentation slides

• Live or video demo of the model

• Handout or supplementary material (optional but encouraged)

Step 6: Final Report (Finals Week – Dec 13)¶

Report Structure¶

1. Title and Abstract (¼ page)

• Concise summary of research question, model, and key findings

2. Introduction & Motivation (1 page)

• Why does this social phenomenon matter?

• What question are you answering?

• How does your model contribute to understanding?

3. Social Theory & Model Design (2–3 pages)

• Which theorist(s) inform your model and why?

• How do theoretical concepts map to model rules?

• Pseudocode, diagrams, or detailed logic explaining agents, interactions, parameters

4. Implementation & Computational Methods (1–2 pages)

• Platform used (NetLogo, Python Mesa, etc.)

• Code structure (brief overview or appendix reference)

• How did you measure norms? (metrics, visualizations)

5. Results & Findings (2–3 pages)

• What patterns emerged?

• How sensitive is the model to parameter changes?

• Which conditions lead to norm stability vs. collapse?

• Plots, tables, or screenshots supporting your findings

6. Interpretation & Real-World Connections (1–2 pages)

• What do these results tell us about the real-world phenomenon?

• How do your findings relate to empirical research or theory?

• What are the limitations of your model? What did you simplify?

7. Reflection: Imperfect Models in a Fallen World (½–1 page)

• How does Smedes’ concept of “imperfect models” apply to your work?

• What ethical or philosophical questions does modeling norms raise?

• What would a “better” model look like, and why is it hard to build?

8. References & Appendices (as needed)

• Bibliography (APA or Chicago format)

• Full code (if not already submitted)

• Additional plots or data tables

Expectations for Step 6¶

• 7–10 pages (single-spaced, including figures)

• Academic writing style (clear, organized, evidence-based)

• Strong connection between theory and modeling

• Honest discussion of limitations and uncertainty

• Proper citations for all sources (code, data, theory, prior models)

Report Deliverables¶

• Final Report (PDF or Word document)

• Code (documented and runnable)

• Data/Results (plots, statistics, or supplementary files)

Why Doing This Work?¶

• Hands-on: Requires real modeling work (parameter design, simulation runs, debugging).

• Interdisciplinary: Merges social theory with computation in a concrete way.

• Theological/Philosophical Depth: Students reflect on the “imperfect models” Smedes highlights—what are the limits of modeling moral/social order in a fallen world?

• Scalable: Strong students can push into advanced territory (network science, machine learning for calibration), while everyone can complete a meaningful baseline project.

Evaluation Criteria¶

See Course Policies → Grading Rubric for the breakdown of project components and point values. Each milestone will be evaluated on:

• Clarity & Communication: Is your idea, design, and results clearly explained?

• Technical Depth: Does your model work? Is the code well-organized?

• Theory Integration: How well do you connect social theory to modeling choices?

• Critical Reflection: Do you honestly discuss what your model reveals and what it misses?

• Collaboration & Teamwork: (If applicable) How well did your team work together?