Introduction to Hypothesis Testing
You'll learn the fundamental framework for making statistical decisions based on data.
What is Hypothesis Testing?
Hypothesis testing is a formal procedure for using sample data to decide between two competing claims about a population. Think of it as a "courtroom for statistics"—you start by assuming someone is innocent (a default position), then examine evidence to see if it's strong enough to conclude they're guilty beyond reasonable doubt.
The Basic Framework
Null and Alternative Hypotheses
Every hypothesis test begins with two statements:
-
Null hypothesis (H₀): The default assumption or "status quo." It usually states that there's no effect, no difference, or that a parameter equals a specific value. Example: "The population mean is 100" or "This new drug has no effect."
-
Alternative hypothesis (H₁ or Hₐ): The claim you're trying to find evidence for. It contradicts the null. Example: "The population mean is not 100" or "This drug does have an effect."
Test Statistic
A test statistic is a single number computed from your sample data that measures how far your data deviates from what H₀ predicts. Common examples include z-scores or t-scores that standardize your sample mean.
Making a Decision
You compare your test statistic to a critical value (or examine its probability) to decide: either reject H₀ (conclude there's sufficient evidence for H₁) or fail to reject H₀ (insufficient evidence to abandon the null hypothesis).
The decision isn't about "proving" anything—it's about whether the data provide strong enough evidence against the null hypothesis.
Key Takeaway: Hypothesis testing provides a structured framework using null/alternative hypotheses, test statistics, and decision rules to make principled statistical inferences from sample data.