Course contentsShow
← Data Science
Lesson 307 of 2,1459. Hypothesis Testing FoundationsFree lesson

Defining the Null Hypothesis (H₀)

Learn what the null hypothesis represents: a statement of no effect, no difference, or status quo to be tested.

Defining the Null Hypothesis (H₀)

What You'll Learn

You'll understand what the null hypothesis represents and why it serves as the starting point for statistical testing.

The Core Idea

The null hypothesis (H₀) is a formal statement that assumes nothing interesting is happening—no effect, no difference, no relationship between variables. It represents the status quo or default position that we test against.

Think of it like a courtroom trial: the defendant is presumed innocent (the null hypothesis) until evidence proves otherwise. We don't assume guilt from the start; we need strong evidence to reject innocence. Similarly, in statistics, we assume "no effect" until our data provides compelling evidence to the contrary.

Common Forms of H₀

The null hypothesis typically takes these shapes:

  • No difference: "The mean weight of Group A equals the mean weight of Group B" (μ₁ = μ₂)
  • No effect: "This new drug has no effect on blood pressure" (effect = 0)
  • No change: "Customer satisfaction hasn't changed after the redesign" (before = after)
  • No relationship: "There's no correlation between study hours and test scores" (correlation = 0)

Why Start with H₀?

We need a specific claim to test. The null hypothesis gives us a baseline to compare our data against. By assuming "nothing special," we can calculate how surprising our observed data would be if H₀ were actually true. If our data looks very unlikely under H₀, we have reason to doubt it.

Key Takeaway: The null hypothesis (H₀) is your starting assumption of "no effect" or "no difference"—it's what you're trying to find evidence against, not prove true.