How do you figure out sample size given a confidence interval and range of deviation?

[fstep]

Lets say you want to test 2 different banner ad campaigns.
1 is (control) and 2 is (test)
If the baseline click through ratio is 2% (for control) and the +- range of deviation from this response rate is 0.25%
What sample size would you need at 90% confidence interval? How would it differ at 90% if you were only willing to take a +-0.1% response rate?

Please explain how to do this.

 

[mmm4444bot]

Hi. In general, the tutors here do not provide step-by-step solutions or teach course material on these boards.

We answer specific questions about the exercise. We offer guidance, examples, or links to outside resources based on the information that you post about your efforts.

What have you thought about or tried thus far, in this exercise? Is there some terminology in the problem statement that you do not understand?

Cheers. :cool: