Mary Ann S. answered 07/24/22
Ph.D. Educational Measurement, Doctoral Minor in Statistics.
You are well on your way to the answer!
H0: p = .10 correct.
but H1 should be, H1: p ≠ p.10
1.) What quantities are being asked about: the sampling distribution of a proportion. A proportion is simply a special case of a mean, and you may use the statistical tests appropriate for the sampling distribution of a mean.
2.) You have a large N here, so you use a single sample Z test instead of a t test.
3.) The Z score is calculated as follows:
Z = (proportion observed in sample vs hypothesized population portion) / standard error of the proportion
4.) You need to go put a few pieces together to get the parts for your Z score:
- calculate your sample proportion, 39/359 = .1086
- calculate your standard error (same calculation as standard error of the mean)
- The variance of a proportion is p*(1-p), your case, (.1086)(.8914)
- The standard error is sqrt(variance/N)
5.) Look up your Z score in a z-table or run it through excels NORM.S.DIST() function.
- You are doing a two-tailed test because you have been asked whether or not there is a "difference" between two quantities, not whether one is greater than the other, or whether one is less than the other.
- You will need to find the p(Z <=) the negative value of your Z score + the p(Z >=) the positive value of your Z score. For instance, if you calculated a Z score of -1.58, you would want to find out the p(Z) <= 1.58 , and the p(Z) >= 1.58. The good news is you only have to do one of them and then multiply the probability by 2. This is your p-value.
- If you are using excel, enter =NORM.S.DIST(your z score, 1). (The "1" tells excel you are working with the cumulative distribution). This will give you p(Z <= Z). Do 1 - p(Z <= Z) to get the p(Z >= z) for a positive z value. Again, 2 * the probability you get as an answer will be your p-value
If your p value < .10, reject H0.