EXAMPLE 7.3

One-sided test for watching traditional television. For the problem described in the previous example, we want to test whether the U.S. college student average is smaller than the overall U.S. population average. Here we test

H₀: μ = 18.5

versus

H_a: μ < 18.5

The t test statistic does not change: . As Figure 7.4 illustrates, however, the P-value is now , half of the value in the previous example. From Table D, we can determine that ; software gives the exact value as . Again, there is not enough evidence to reject the null hypothesis in favor of the alternative at the 0.05 significance level.