Null Hypothesis

The null hypothesis states that there is no effect, no difference or no relationship in the population. It is the default position a statistical test is built to challenge, while the alternative hypothesis states what you actually expect to find.

Testing proceeds by assuming the null is true and asking how surprising the observed data would be under that assumption. The p-value quantifies that surprise, and if it falls below the chosen significance level — conventionally 0.05 — you reject the null. Rejecting it is not proof that the alternative is true; it is evidence inconsistent with the null.

Crucially, you never accept the null hypothesis. A non-significant result means you failed to reject it, which may reflect a genuinely absent effect or simply a study with too little power to detect one. Phrase such findings carefully in your thesis: no significant difference was found, rather than the groups were the same.