Determine Sample Size

[ojvko]

I'm not sure what direction to go with this problem.

I'm given 2 random samples:
1. 120 men, mean age 26.7, standard deviation 5.2yrs
2. 150 women, mean age 25.1, standard deviation 4.6yrs

I'm required to 1st construct a 98% confidence interval for the difference in the true mean age between men and women.

I've done that... interval (0.19, 3.01)
Bound on the error = 1.41

Now I have to determine the minimum sample sizes necessary to estimate the difference in the population means to with 1 year with a 98% confidence level assuming equla sample sizes of both men and women and using the estimated standard deviations from above.

I can determine the sample size for 1 sample. I'm just not sure how to go about setting up for 2 different samples.

A point in the right direction would be greatly appreciated!!

[quark]

Hi ojvko,

If you've calculated Sp, the pooled standard deviation for the two samples, you can then get the sample size by solving:

2.33*Sp*sqrt(1/n+1/n) = margin of error

2.33 corresponds to 98% confidence.

[ojvko]

Thanks for the help!!