Course contentsShow
← Data Science
Lesson 421 of 2,14511. Hypothesis Tests for Proportions and CategoriesFree lesson

Applications: Uniform, Genetic Ratios, and Distributions

Applying goodness of fit tests to real scenarios like testing uniformity, Mendelian ratios, and demographic models.

Applications: Uniform, Genetic Ratios, and Distributions

You'll learn how to apply chi-squared goodness of fit tests to real-world scenarios involving uniform distributions, genetic inheritance patterns, and demographic models.

Why This Matters

Now that you know how to perform a chi-squared goodness of fit test, it's time to see where it shines in practice. This test helps you answer questions like: "Is this die fair?", "Do my data follow Mendelian genetics?", or "Does voter turnout match population distribution?"

Common Real-World Applications

Testing for Uniformity

When you want to know if outcomes are equally likely (uniform distribution), you set all expected frequencies equal. Examples include:

  • Fair dice or spinners: Are all six faces equally likely?
  • Random number generators: Does each digit appear with equal frequency?
  • Website traffic: Are visitors distributed evenly across days of the week?

Genetic Ratios (Mendelian Inheritance)

Classical genetics predicts specific ratios from crosses. For a simple dominant/recessive trait, you expect a 3:1 ratio in offspring. The chi-squared test checks if observed offspring counts match theoretical predictions (like 75% dominant, 25% recessive).

Demographic or Distribution Models

You can test whether observed data match a theoretical population distribution:

  • Census data: Does your sample reflect regional population proportions?
  • Customer demographics: Do purchases align with market segment sizes?

The Process

For each application, you follow the same workflow: define your null hypothesis (data match the theoretical model), calculate expected frequencies based on that model, compute the chi-squared statistic, and decide whether deviations are due to chance or a real mismatch.

Key Takeaway: Chi-squared goodness of fit tests transform theoretical models (uniform, genetic, demographic) into testable hypotheses, letting you validate whether real-world data conform to expected patterns.