Remember that we test the null against an alternative. The null hypothesis is what is known or assumed from theory or previous research. In our example, previous research told us that college students spent about 6 hours per weekday in face-to-face contact (give or take 2 hours). Thus, our null hypothesis is that μ = 6.0. The null hypothesis sampling distribution and the alternative hypothesis sampling distribution are presented below.

Next, we draw our sample. We asked 30 college students to record in a log how much time they spend socializing face-to-face with friends for 1 week. The sample data are graphed below.

To test our hypothesis, we compare the sample mean to the null population (because we always test the null). We also set the level of error we will accept in our test (alpha level). If our sample mean is close to the null population mean, then we fail to reject the null hypothesis and conclude that our sample is like the null population. If our sample mean is far away from the null population mean, then we reject the null hypothesis and conclude that our sample came from the alternative distribution.