We wish to test whether the correlation between two variables is significantly different from 0 (i.e., no relationship between X and Y in the population). When we look at the scatterplot of our data, we find that values of Y are tightly clustered together at low values of X, but much more spread out at high values of X. What assumption have we violated?
A) the assumption of a linear relationship
B) the assumption of homoscedasticity
C) the assumption of interval-ratio measurement
D) the coefficient of determination assumption

Question

Quizplus · Accepted Answer

The assumption of homoscedasticity has been violated. Homoscedasticity assumes that the variance of the dependent variable is the same at all values of the independent variable. In this case, the variance of Y is smaller at low values of X than at high values of X.

We Wish to Test Whether the Correlation Between Two Variables