Course contentsShow
← Data Science
Lesson 308 of 2,1459. Hypothesis Testing FoundationsFree lesson

Defining the Alternative Hypothesis (H₁ or Hₐ)

Understand the alternative hypothesis as the research claim: what you hope to find evidence for in your data.

Defining the Alternative Hypothesis (H₁ or Hₐ)

You'll learn what the alternative hypothesis represents and why it's the claim you're trying to find evidence for.

What Is the Alternative Hypothesis?

The alternative hypothesis (written as H₁ or Hₐ) is your research claim—the statement you believe might be true and want to find evidence for in your data. It directly contradicts the null hypothesis (H₀) you learned in the previous lesson.

Think of it like a courtroom analogy: if the null hypothesis is "innocent until proven guilty," the alternative hypothesis is the prosecutor's claim that "the defendant is guilty." You gather evidence (data) hoping to support this claim.

Why It Matters

The alternative hypothesis represents what you're actually interested in discovering:

  • A medication works better than placebo (not that there's no difference)
  • Conversion rate A is higher than conversion rate B (not that they're equal)
  • Average customer spending has increased (not stayed the same)

Your entire hypothesis test is designed to determine whether your data provides sufficient evidence to support H₁ over H₀.

Types of Alternative Hypotheses

The alternative hypothesis can take different forms:

  • Two-sided: "The parameter is different from the null value" (≠)
  • One-sided (greater): "The parameter is greater than the null value" (>)
  • One-sided (less): "The parameter is less than the null value" (<)

The form you choose depends on your research question. If you only care about detecting an increase, use a one-sided "greater than" alternative.

Key Takeaway: The alternative hypothesis (H₁ or Hₐ) is your research claim—the effect or difference you hope to find evidence for in your data, and it always contradicts the null hypothesis.