Measles County Simulation and Tipping Points

ISEF Category: Computational Biology and Bioinformatics

Subcategory: Computational Epidemiology  ·  Difficulty: Advanced  ·  Setup: Home Setup  ·  Time: Full Year

The Hook

One missed vaccination can matter more than you think. Measles spreads so fast that a small gap in coverage can act like a hole in a dam. Your model can test where that dam starts to fail in real U.S. counties. You can turn public data into a map of risk, not just a guess.

What Is It?

This project asks a simple question with a big answer, when do measles cases stop staying isolated and start spreading through a county? You will build an agent-based simulation, which means you model people as individuals with simple rules instead of treating a county like one big blur. Think of it like a crowd scene in a video game, where each person moves, meets others, and passes disease based on local contact.

Measles is a good disease for this kind of model because it is very contagious. That means small changes in vaccination coverage can have large effects. You can use public kindergarten MMR exemption data from CDC SchoolVaxView, then test how outbreak risk changes across counties or demographic clusters. The goal is to find tipping points, the coverage level where outbreak spread changes from rare to likely.

Why This Is a Good Topic

This is a strong science fair topic because you can test a real public health question with real data and clear outputs. You do not need a wet lab, but you still get to do original research by building your own model, choosing your own assumptions, and testing your own scenarios. The project connects to vaccine policy, outbreak prevention, and county-level health differences. You can learn simulation design, parameter testing, clustering, and how to turn messy public data into a clear result.

Research Questions

• How does county kindergarten MMR exemption rate change the probability of a simulated measles outbreak?
• What is the effect of school clustering on the tipping point for measles spread?
• Does adding demographic clustering change outbreak size compared with a random mixing model?
• To what extent do different contact network structures shift the vaccination threshold for outbreak control?
• Which county-level exemption patterns produce the largest simulated outbreak sizes?
• How does the assumed number of imported cases affect the estimated re-emergence threshold?

Basic Materials

• Laptop or desktop computer with enough memory to run repeated simulations.
• CDC SchoolVaxView county or state kindergarten vaccination and exemption data.
• Spreadsheet software for cleaning and organizing data.
• Python installed with Jupyter Notebook.
• Internet access for downloading public datasets and documentation.
• Basic statistics reference for probability, distributions, and confidence intervals.

Advanced Materials

• High-performance laptop or desktop computer for large simulation batches.
• County-level demographic and school enrollment datasets from Census or NCES.
• GIS software or mapping tool for visualizing spatial patterns.
• Python packages for network modeling, simulation, and statistical analysis.
• Replication of SchoolVaxView data over multiple years for trend analysis.
• PubMed review articles on measles transmission and herd immunity.

Software & Tools

• Python: Runs the agent-based simulation, data cleaning, and repeated trials.
• Jupyter Notebook: Keeps code, notes, figures, and results in one place.
• pandas: Organizes county data and helps you merge datasets.
• NetworkX: Builds contact networks for people, schools, or counties.
• matplotlib: Makes plots of outbreak size, risk curves, and threshold charts.

Experiment Steps

1. Define the exact outcome you want to predict, such as outbreak probability, final outbreak size, or the coverage tipping point.
2. Choose your unit of simulation, such as people, schools, or counties, and decide how they connect.
3. Gather public vaccination and exemption data, then clean it into a format your model can read.
4. Build a baseline model with simple transmission rules and check that it behaves sensibly in extreme cases.
5. Add one realistic feature at a time, such as demographic clustering, school structure, or imported cases, then compare results.
6. Plan how you will summarize uncertainty, compare scenarios, and show which assumptions change the tipping point most.

Common Pitfalls

• Treating county averages like all households are evenly mixed, which can hide school-level clustering effects.
• Using one simulation run per county, which makes random noise look like a real pattern.
• Mixing data from different years without checking whether exemption rates and demographics match the same time period.
• Choosing transmission rules that are too simple, which can make measles look less contagious than it is.
• Reporting a tipping point without uncertainty bounds, which makes the threshold look more exact than the model supports.

What Makes This Competitive

A class-level project might only report one risk curve. A stronger project compares multiple contact structures, then shows which one best matches known outbreak behavior. You can raise the level again by testing sensitivity to assumptions, like importation rate, household mixing, and school clustering. Clear uncertainty analysis and careful validation against known measles patterns can make the work feel much more like real epidemiology.

Project Variations

• Replace county-level data with school-level exemption patterns and compare risk across districts.
• Model another vaccine-preventable disease, such as pertussis, and test whether its tipping point differs from measles.
• Add spatial spread between neighboring counties to see whether local clusters can trigger regional outbreaks.

Learn More

• CDC SchoolVaxView: Find vaccination and exemption data for kindergarten and other school levels on the CDC website.
• CDC Measles Data and Statistics: Read background on U.S. measles trends and outbreak patterns on the CDC measles page.
• CDC Pink Book, Measles chapter: Review transmission, immunity, and vaccination basics in the CDC’s vaccine handbook.
• PubMed: Search for review articles on measles transmission, herd immunity, and agent-based modeling.
• MIT OpenCourseWare, Introduction to Modeling and Simulation: Use free course materials to learn how to build and test simulation models.