r Testing

Question 1

What is an assumption of testing differences between to r values?

Answer 1

The correlations are independent and come from different people.

Question 2

Under the null hypothesis, what happens to the sampling distribution when rho1 = rho2?

Answer 2

The sampling distribution will be skewed. Because rho is bound by 1, so if rho does not equal 0 then some values will become possible but less likely.

Question 3

Why convert r to r’?

Answer 3

To account for skewed distributions when population correlation does not equal 0

Question 4

As an r value moves further from 0 how will the r’ transformation be affected?

Answer 4

Greater transformation from original value the further from 0. (Close to 0 requires less transformation)

Question 5

What is a curvilinear relationship?

Answer 5

Where the direction of relationship eventually reverses

Question 6

What factors affect r? 4

Answer 6

1. Nonlinear relationships
2. Restriction of range
3. Extreme scores 
4. Heterogenous sub samples

Question 7

What are heterogenous sub samples?

Answer 7

Data which can be divided into distinct sets based on some other variable (eg, men and women vs. all people)