"

8.3 Main Idea Behind Hypothesis Tests for μ

The main idea of a hypothesis test is to use the data as evidence to disprove the null H0 and thus prove that the alternative Ha is true. The idea behind a hypothesis test for the population mean is as follows:

Collect from the population a simple random sample: x1,x2,,xn and calculate the sample mean x¯=x1+x2++xnn. Our “evidence” stems from the discrepancy between the point estimate x¯ and the hypothesized population mean μ0.

Reject the null hypothesis H0 if the sample mean x¯ does not support the null H0. That is, we should reject H0 if x¯ is too extreme. The word “extreme” means contradicting the null H0 in favour of the alternative Ha.

Three density graphs illustrating the rejection areas for x-bar. Image description available.
Figure 8.2: Rejection Region Based on Sample Mean. [Image Description (See Appendix D Figure 8.2)]

In order to quantify how the data (our evidence) contradict the null hypothesis, we first assume the null hypothesis H0 is true and calculate the chance of observing a sample mean at least as extreme as the observed x¯. Reject the null H0 if the chance is small; otherwise, fail to reject H0. Recall that for a normal population or a large sample size, the sample mean X¯N(μ,σn)Z=X¯μσ/nN(0,1) and t=Xμ¯s/nt distribution with df=n1. We call the variables Z=X¯μσ/n or t=Xμ¯s/n the test statistics. We should reject the null hypothesis H0 if the observed test statistic zo=x¯μ0σ/n or t=x¯μ0s/n is too extreme.