Question: Suppose you are the marketing director for a children's breakfast cereal, and you have traditionally allocated a significant portion of your advertising budget to commercials on late-afternoon TV shows typically viewed by kids, based on studies indicating that the average child in the age group of interest watches 90 minutes of late-afternoon TV each day. You suspect, however, that with the advent of the Internet, kids view less TV than they used to, so you commission a study of 100 children, and find that they view an average of 86 minutes per day, with a sample standard deviations of 20 minutes.

What are the null and alternative hypotheses?

Assuming a significance level of 0.05 -- that is, we will reject the null hypotheses if the probability of getting the sample results, assuming that μ does in fact equal 90, is less than 5%. – how do we proceed to test whether your suspicions of reduced TV viewing is true?