Question

The covariance of two variables is:

a) how they deviate from their means together

b) how they deviate from their standard deviations together

c) the total variance of both variables

d) the percent of variance in one variable explained by another

e) the squared differences from time 1 to time 2

If a researcher concludes that “decreases in self-esteem are strongly associated with decreases in social interaction” then what correlation coefficient describes her findings?

a) .08

b) .87

c) -.87

d) .32

e) -.32

The percent of variance in one variable explained by another is:

a) covariance

b) Pearson’s r

c) r-squared

d) partial r

e) sum of squares

A spurious relationship is one in which:

a) The relationship is strengthened by the addition of a control variable

b) The relationship disappears with the addition of a control variable

c) The relationship is unchanged by the addition of a control variable

d) The size of the relationship is halved by the addition of a control variable

e) None of the above

If we find a significant positive correlation between income and voting for Republican candidates we can conclude that:

a) A higher income causes people to vote Republican

b) Voting Republican leads to a higher income

c) In the population income and vote choice are unrelated

d) Knowing a person’s level of income will help us predict their vote choice

e) The relationship between income and vote choice is spurious

Answer 1

The covariance of two variables is (a) how they deviate from their means together.

This is the case of strongly positve correlation. So the value of the correlation coefficient is close to 1. Therefore, its value is (b) 0.87

The percent of variance in one variable explained by another is the coefficient of determination i.e. (c) r - squared.

A spurious relationship is one in which (a) the relationship is strengthened by addition of a control variable.

If we find a significant positive correlation between income and voting for republication candidates, we can conclude that (e) the relationship between income and vote choice is spurious.