95% confidence interval for mean of ad-areas on 20 pages

I am in my second 5 week course of statistis and am just not getting the equations to work for me. Can you help me with this one?

A sample of tweny pages taken without replacement from the 1,591-page directory Ameritech Pages Plus Yellow pages. On each page, the mean area devoted to display ads was measured. The data ( insquare millimeters) are shown:

0 260 356 403 536 0 268 369 428 536 268 396 469 536 162 338 403 536 536 130

a. Contruct a 95% confidence interval for the true mean ( which is 346,5 by the way) b. Why might normality be an issue here? c. What sample size would be needed to obtain an error of +/- 10 sqare millimeters with 99% confidence? d. If this is not a reasonable requirement, suggest one that is.

I need to show all the work and I have no idea how to set his up. :roll:

By "true mean" I assume you mean for the population, not for the simple random sample. There is no confidence interval for the mean of the SRS shown, since you can compute that exactly.

http://people.hofstra.edu/Stefan_Waner/realWorld/finitetopic1/confint.html

Normality refers to behavior like the normal curve, and in this sample, there seem to be a few pages with 0 mm ad space. What would that do to the graph? If at the mean, you have a maximum and then moving to the left, it decreases, but then all of a sudden, it goes back up at zero, the graph will not be a normal curve. You can't use the Central Limit Theorem unless you know the distribution to be normal. That could be a big problem here: We have reason to suspect that the distribution does not follow a normal curve. The t-distribution would be a better approach.

If you have a specific question, please ask, but be sure to show us how far you've gotten in the calculation of the confidence interval. Then, we can spot the problems you are having more clearly and help a lot better.

that was a great web page you gave me to check out. It actually made sense. thanks!