Alternative Hypothesis

The alternative hypothesis (H₁ or Hₐ) is the claim you actually suspect is true — the existence of a real effect or difference — which you adopt only if the data make the null implausible. It is the “something is going on here” counterpart to the null’s “nothing is going on.”

Decide whether it is one-sided (the effect goes in a specific direction, e.g. the new button is better) or two-sided (any difference counts) before you see the results — choosing afterward to get a smaller p-value is a form of cheating. A two-sided alternative is the honest default unless you have a strong prior reason for a direction.