The null hypothesis is a default assumption in statistical testing that states there’s no effect, no relationship, or no difference between groups or variables you’re studying.
We never “prove” the null hypothesis - we either reject it (if we find strong evidence against it) or fail to reject it (if the evidence isn’t strong enough). This asymmetry exists because it’s easier to find evidence that something exists than to prove something doesn’t exist.
The threshold for “strong enough evidence” is your significance level (often ), which represents the probability of rejecting the null hypothesis when it’s actually true (Type I error). If your p-value is less than , you reject the null hypothesis in favor of the alternative hypothesis.
See p-value for common misinterpretations and pitfalls.