Setting Up Hypotheses for Goodness of Fit
You'll learn how to formulate the null and alternative hypotheses when testing whether your observed data matches an expected distribution.
The Core Idea
When you perform a Chi-Squared Goodness of Fit test, you're essentially asking: "Does my data follow a particular distribution pattern I expected?" To answer this scientifically, you need to set up two competing hypotheses.
The null hypothesis (H₀) claims that your data does follow the specified distribution—that any differences between what you observed and what you expected are just due to random chance.
The alternative hypothesis (H₁ or Hₐ) claims that your data does not follow that distribution—that the deviations are real and meaningful.
Formulating the Hypotheses
Null Hypothesis (H₀)
"The observed data follows the specified distribution."
More formally: The population proportions in each category match the expected proportions.
Alternative Hypothesis (H₁)
"The observed data does not follow the specified distribution."
More formally: At least one category's population proportion differs from what was expected.
A Practical Example
Imagine testing whether a die is fair. You'd write:
- H₀: The die is fair (each face has a 1/6 probability)
- H₁: The die is not fair (at least one face has a probability ≠ 1/6)
Notice the alternative is non-directional—you're not specifying which face might be biased or how, just that something's off.
Key Takeaway: In goodness of fit tests, H₀ states the data follows a specific distribution, while H₁ asserts some deviation exists—without specifying the exact nature of that deviation.