PMCC hypothesis testing help

Watch this thread
Stormragexox
Badges: 10
Rep:
? You'll earn badges for being active around the site. Rep gems come when your posts are rated by other community members.
#1
Report Thread starter 3 years ago
#1
Hey guys I have a problem deciding if the alternative hypothesis is not equal more than or less than ro

I looked at the answers and apparently ro is more than zero why can't we use not equal to or less than instead
0
reply
Stormragexox
Badges: 10
Rep:
? You'll earn badges for being active around the site. Rep gems come when your posts are rated by other community members.
#2
Report Thread starter 3 years ago
#2
(Original post by Stormragexox)
Hey guys I have a problem deciding if the alternative hypothesis is not equal more than or less than ro

I looked at the answers and apparently ro is more than zero why can't we use not equal to or less than instead
Name:  FullSizeRender (20).jpg
Views: 399
Size:  40.6 KB
0
reply
Notnek
Badges: 20
Rep:
? You'll earn badges for being active around the site. Rep gems come when your posts are rated by other community members.
#3
Report 3 years ago
#3
(Original post by Stormragexox)
Hey guys I have a problem deciding if the alternative hypothesis is not equal more than or less than ro

I looked at the answers and apparently ro is more than zero why can't we use not equal to or less than instead
I don't understand what you mean. Can you please try explaining again?
0
reply
Stormragexox
Badges: 10
Rep:
? You'll earn badges for being active around the site. Rep gems come when your posts are rated by other community members.
#4
Report Thread starter 3 years ago
#4
(Original post by Notnek)
I don't understand what you mean. Can you please try explaining again?
Yes sure what I mean is that I tried doing part b and I thought that the hypothesis test would be two tailed, as the question does not mention if the correlation is postive or negative. Howver in when I checked out the solutions the hypothesis testing was one tailed. So my main question is what part of the question hints that a one tailed hypothesis test has to be used, and as one tailed is used what part of the question hints that rho >0 or rho<0 has to be used (Sorry if this still doesn't make sense I am so confused about this question)
0
reply
Notnek
Badges: 20
Rep:
? You'll earn badges for being active around the site. Rep gems come when your posts are rated by other community members.
#5
Report 3 years ago
#5
(Original post by Stormragexox)
Yes sure what I mean is that I tried doing part b and I thought that the hypothesis test would be two tailed, as the question does not mention if the correlation is postive or negative. Howver in when I checked out the solutions the hypothesis testing was one tailed. So my main question is what part of the question hints that a one tailed hypothesis test has to be used, and as one tailed is used what part of the question hints that rho >0 or rho<0 has to be used (Sorry if this still doesn't make sense I am so confused about this question)
Ah I see what you mean now

"Test... whether these results show evidence of a linear relationship..."

This suggests to me that it should be a two-tailed test and I disagree with the answer. Tagging Gregorius though because I could be talking rubbish!
0
reply
Gregorius
Badges: 14
Rep:
? You'll earn badges for being active around the site. Rep gems come when your posts are rated by other community members.
#6
Report 3 years ago
#6
(Original post by Notnek)
Ah I see what you mean now

"Test... whether these results show evidence of a linear relationship..."

This suggests to me that it should be a two-tailed test and I disagree with the answer. Tagging Gregorius though because I could be talking rubbish!
No, I think you're right. It should be two-sided.

The question is problematic on a number of levels. Correlation is a measure of the strength of the linear relationship between two variables, but you simply can't use it to test for the linearity of the relationship itself (which is what the wording of the question seems to suggest). You can have a perfectly respectable non-linear relationship between two variables, for which a statistical test on the correlation coefficient would return a significant result. Clumsy all around...
0
reply
X

Quick Reply

Attached files
Write a reply...
Reply
new posts
Back
to top
Latest
My Feed

See more of what you like on
The Student Room

You can personalise what you see on TSR. Tell us a little about yourself to get started.

Personalise

Were exams easier or harder than you expected?

Easier (12)
37.5%
As I expected (7)
21.88%
Harder (12)
37.5%
Something else (tell us in the thread) (1)
3.13%

Watched Threads

View All