A medical statistician wants to estimate the average weight loss of people who are on a new diet plan.

Question

A medical statistician wants to estimate the average weight loss of people who are on a new diet plan. In a preliminary study, he guesses that the standard deviation of the population of weight loss is about 11 pounds. How large a sample should he take to estimate the mean weight loss to within 5 pounds, with 94% confidence?

Answer 1

When estimating the mean the error of estimation can be calculated by the formula zσ/√n.

 

We want this value to be 5 pounds or less. The z value for 94% confidence is 1.88 (we can use normal since we have a standard deviation to work with.)

 

That gives us (1.88)(11)/√n =5.

 

√n = (1.88)(11)/5=4.136

 

n=17.1  so choose n=18

 

Any questions comment back.

 

 

 

Answer 2

This question is about test power. You can express the minimal sample size for the desired power from a power inequality expression.