Null Hypothesis

TL;DR

• Serves as the default starting assumption in statistical testing: that there is no meaningful difference between groups or variables.
• Must be stated so it can be rejected and is evaluated using statistical tests that measure the probability that observed differences are due to chance.
• Can never be proven true; absence of evidence for a difference does not confirm the null hypothesis.

Definition

The null hypothesis is a statistical hypothesis that suggests that there is no significant difference between two groups or variables. It is used as a starting point for statistical tests and is often referred to as the “default” hypothesis.

Explanation

• The null hypothesis functions as the initial assumption in an analysis: researchers test whether the data provide sufficient evidence to reject it.
• It should be formulated in a way that allows rejection—for example, stating that there is no significant difference in the outcome between the treatment group and the control group.
• Statistical tests quantify the likelihood that any observed difference would occur if the null hypothesis were true; this helps distinguish effects due to chance from true differences.
• Because statistical analysis can fail to detect an existing difference, the null hypothesis can never be proven true with certainty; results must be interpreted with caution.

Examples

Drug effectiveness (blood pressure)

Consider a study investigating whether a new drug reduces blood pressure. The null hypothesis might state that the new drug has no effect on blood pressure, meaning researchers expect no significant difference in blood pressure between people who take the drug and those who do not. If researchers find a significant difference between the two groups, they can reject the null hypothesis and conclude that the new drug is effective at reducing blood pressure.

School type and self-esteem

In a psychology study comparing self-esteem of children who attend single-sex schools versus co-ed schools, the null hypothesis might state that there is no significant difference in self-esteem between the two groups. If researchers find a significant difference in self-esteem between the groups, they can reject the null hypothesis and conclude that attending a single-sex school has an impact on self-esteem.

Notes or pitfalls

• The null hypothesis is not a statement of fact; it is the starting point for statistical analysis.
• It must be stated so it can be rejected (i.e., allow for the possibility of finding a significant difference).
• Statistical tests are required to assess the probability that observed differences arise by chance if the null hypothesis is true.
• The null hypothesis can never be proven true because an undetected significant difference may exist.

Related terms

• Statistical test
• Statistical analysis
• Default hypothesis