Statistics questions (just general, no calculations)

[thirtythree]

These questions are bugging me. I'm supposed to tell what is wrong with what these statements but I can't even figure out what they are trying to say.

A study reports statistically significant results (p=.001, df=0) after doing a one sample t-test for the proportion of the population.

First of all, I thought you were supposed to use a z-test for proportions. And can you have df=0? If so that would mean the sample size was 1 which wouldn't pass the np>=10 and n(1-p)>=10 tests for using proportions anyway. 😕

2. A student searched the Internet for data on a project. He found a file with the population of each of the states of the USA. He used a calculator to construct a 95% confidence interval for the means of the population.

I don't understand what he is doing here. Creating a C.I. for the mean population of a state? What would be the problem with that? It would be kinda pointless since he knows the pop. of all the states but is there anything wrong with doing it?

 

1) You're correct, as far as I know. Z-tests should be used for proportions, and T-tests should be used for populations.

2) I think that the 'wrongness' comes from constructing a C.I. for the mean population of a state - you already know the states' ACTUAL populations, but what the heck is a mean population of a state? According to the problem, the student already knows the real populations.

I might be wrong, but that's what it sounds like to me.

EDIT: either way, it sounds like these questions are worded pretty badly, even if you're supposed to be picking out inconsistencies.

 

[thirtythree]

Originally posted by: jumpr
2) I think what this 'student' is doing is construction a confidence interval for all the states in the U.S. (e.g., "If you were to pick a state at random, its population would 95/100 times be in this confidence interval."

So I guess I just say it would be pointless since he knows the populations of the entire population (hmm .. that sounds funny). These seem like stupid questions but oh well.

 

[thirtythree]

Originally posted by: jumpr
EDIT: either way, it sounds like these questions are worded pretty badly, even if you're supposed to be picking out inconsistencies.

Just what I was thinking ..

Thanks for the reply 😉