Which statement correctly describes the null hypothesis in hypothesis testing?

Which statement correctly describes the null hypothesis in hypothesis testing?

• A. The null hypothesis posits no effect or difference; it is the default assumption tested by data.
• B. The null hypothesis posits no effect; the significance level describes the probability of obtaining results as extreme as observed if the null is false.
• C. The null hypothesis posits an effect; the p-value represents the probability of observing the data under the null.
• D. The null hypothesis posits no effect; it is tested and either rejected or not rejected based solely on the p-value.

Answer: (A)

The null hypothesis is a statement of no effect or no difference and serves as the baseline claim that data are used to test. In hypothesis testing, you collect data and assess how compatible those data are with that no-effect assumption. The p-value then tells you the probability of observing data as extreme as what you saw if the null is true; if that probability is very small, you reject the null and conclude there is evidence of an effect. The statement that captures this idea—no effect and the default assumption tested by data—is the best description. The other options mix up what the null asserts, what the p-value means, or how decisions are made (for example, treating the significance level as a probability about the null being true, or requiring decisions based solely on the p-value).

The null hypothesis is a statement of no effect or no difference and serves as the baseline claim that data are used to test. In hypothesis testing, you collect data and assess how compatible those data are with that no-effect assumption. The p-value then tells you the probability of observing data as extreme as what you saw if the null is true; if that probability is very small, you reject the null and conclude there is evidence of an effect. The statement that captures this idea—no effect and the default assumption tested by data—is the best description. The other options mix up what the null asserts, what the p-value means, or how decisions are made (for example, treating the significance level as a probability about the null being true, or requiring decisions based solely on the p-value).