3

I'm trying to understand correlation coefficients, and seem to be missing something.

For example,

Correlation coefficient: Indicates the direction, positively or negatively of the relationship, and how strongly the 2 variables are related. I understand that the strength can vary from 0-1 and I thought I understood that positive or negative simply had to do with the direction of the correlation.

This question is part of my review and I don't understand why my answer is incorrect...

Which correlation coefficient indicates the strongest relationship between two variables?

a) -0.85 - This was my answer which I thought was the closest to one (meaning strongest

b)1.94

c)0.58 - This is what the textbook says is the correct answer, but why?

d)0.19

Thanks for your help!

Anitra Parks
  • 31
  • seems like the "strongest" is meant to be the "most positive" - wich would an arguable definition in my opinion. – Max Oct 03 '15 at 13:52
  • By correlation, do you mean Pearson's coefficient or something else? If so, $1.94$ is not even valid value, so it's out of question. See interpretation on Wikipedia. – Ennar Oct 03 '15 at 13:53
  • Yes, I knew that 1.94 wasn't a valid choice so that immediately eliminated that question, but I thought the strongest correlation coefficient should be a) -0.85 – Anitra Parks Oct 03 '15 at 14:07
  • I'm guessing $1.94$ is a typo and $0.94$ was intended. ${}\qquad{}$ – Michael Hardy Oct 03 '15 at 14:10
  • If the textbook is consequent then it would say that a relationship characterized by correlation coëfficient $0.01$ (almost indepent) is stronger than one with coëfficient $-0.99$. In my view your answer is the correct one. – drhab Oct 03 '15 at 14:21
  • Thanks guys! I think I will email my prof and see if the answer is a typo? – Anitra Parks Oct 03 '15 at 14:34

0 Answers0