You are Here: Home >< Maths

# Testing if Spearmans Rank is zero watch

1. Hi guys.

Can anyone explain to me why they used a two tailed test in part b as opposed to a one tailed test to test for positive correlation.

2. (Original post by Shaanv)
Hi guys.

Can anyone explain to me why they used a two tailed test in part b as opposed to a one tailed test to test for positive correlation.

Eek!, that's a poorly thought out question. The point is that a-priori you have no knowledge of whether the association is positive, negative or null. This means that an appropriate approach is to test the H0 and H1 that they give in the answers to see whether there is any association, which would rightly use a two-tailed test. You can then look at the sign of the correlation to conclude as they wish...

However, if you really are only interested in whether there's a positive association, then a one-tailed test seems perfectly legitimate - but you would have to alter your null and alternative hypotheses appropriately.
3. (Original post by Gregorius)
Eek!, that's a poorly thought out question. The point is that a-priori you have no knowledge of whether the association is positive, negative or null. This means that an appropriate approach is to test the H0 and H1 that they give in the answers to see whether there is any association, which would rightly use a two-tailed test. You can then look at the sign of the correlation to conclude as they wish...

However, if you really are only interested in whether there's a positive association, then a one-tailed test seems perfectly legitimate - but you would have to alter your null and alternative hypotheses appropriately.
Ahh ok makes sense so if the test statistic was significant causing them to reject the H0 and conclude that rho is non zero, then i would state that the correlation is positive as my test statistic is above zero. Is this correct?

Thanks for the help
4. (Original post by Shaanv)
Hi guys.

Can anyone explain to me why they used a two tailed test in part b as opposed to a one tailed test to test for positive correlation.

I'm pretty sure that a one-tailed test should have been used, since the question specifically says to test for a "positive association" (not just any association). I definitely would have just done exactly the same thing as you did. I feel like this is just another oversight in the solution bank (which aren't exactly infrequent!).

Even still, I think that some of the Edexcel mark schemes still let you get marks, even if you've used a one-tailed test when they wanted a two-tailed test (and vice-versa), so you shouldn't be completely ruined if you did so.
5. (Original post by Shaanv)
Ahh ok makes sense so if the test statistic was significant causing them to reject the H0 and conclude that rho is non zero, then i would state that the correlation is positive as my test statistic is above zero. Is this correct?
Yes, that's right. I hope you'd get the marks whichever way around you did it, provided that you explained the reasoning correctly!
6. (Original post by mupsman2312)
...
(Original post by Gregorius)
...
U guys are awesome thanks a bunch for ur help.

Have a great day.

TSR Support Team

We have a brilliant team of more than 60 Support Team members looking after discussions on The Student Room, helping to make it a fun, safe and useful place to hang out.

This forum is supported by:
Updated: March 2, 2018
Today on TSR

### Uni league tables

Do they actually matter?

### University open days

• Staffordshire University
Everything except: Midwifery, Operating Department Practice, Paramedic Undergraduate
Sun, 21 Oct '18
• University of Exeter
Wed, 24 Oct '18
Wed, 24 Oct '18
Poll
Useful resources

### Maths Forum posting guidelines

Not sure where to post? Read the updated guidelines here

### How to use LaTex

Writing equations the easy way

### Study habits of A* students

Top tips from students who have already aced their exams