1.Here are the answers.

2.In this circumstance, a sample size of 24 subjects (23.7 rounded up to 24) is required for a one-sample test of mean cholesterol level against the general population assuming a clinically important difference of 20 mg/dl.