785 random subjects, 4years of college. 144 smoke. 641 do not. use the .01 level of significance to test claim that rate (proportion) of smoking among those w/ 4 years of college is less than 27 out of 100. why would rate of college grads who smoke be less than rate of general public. p-value, test statistic?
i have been using t-scale for confidence intervals. should i have a level of significance scale in my book as well. cant find anywhere. any help would be great. ty
2006-12-20 04:39:20 · 3 answers · asked by pluca 2 in Science & Mathematics ➔ Mathematics
I like that!!! If p is low the Ho must go. I hope I remeber that for next year......
So what is the test statistic? It is z= (phat-p)/ sqrt(p*q/n) so this equals (144/785-27/100)/ sqrt(1/785*27/100*73/100) = -5.46 approximately. So this is way in the reject H0 region: accept that smoking rate is less for college grads.
2006-12-20 04:49:15 · answer #1 · answered by a_math_guy 5 · 0⤊ 0⤋
Your Ho for this is that there is no difference or that college degree doesn't affect smoking, therefore your Ha is that a college degree does affect smoking. Run your test and get the p-value. if the p-value is higher than .05 then you "fail to reject the null (Ho)" if it is less than .05 then you "reject the null (Ho)"
The way I remember the p-value rule is:
"If the p is low, the Ho must go!"
Hope this helps!
2006-12-20 12:45:38 · answer #2 · answered by auequine 4 · 0⤊ 0⤋
Since you are using the .01 level of significance, that is your a (alpha) value... you will find that the t-table in your book will have the alpha values on it
2006-12-20 13:11:56 · answer #3 · answered by pinkpearls 3 · 0⤊ 1⤋