Module 9: Hypothesis Testing

The "effect" of a particular factor on some health outcome can be described as a parameter. Statistical hypothesis testing begins with a probability model assuming there is no effect or a null hypothesis (H₀) and deciding whether there is sufficient evidence to reject the null hypothesis in favor of the alternative hypothesis (H₁).

One-sided Hypothesis:   H₀: θ = θ₀ vs H₁: θ > θ₀ (or θ < θ₀)

Two sided Hypothesis: H₀: θ = θ₀ vs H₁: θ != θ₀