Which of the following best describes the statistical concept of confounding in observational studies?

A variable that is caused by the explanatory variable and transmits its effect to the response variable.

A variable that is associated with both the explanatory variable and the response variable and can distort the estimated association between them.

Random variation in measurements that reduces precision but does not bias estimates.

A variable that is independent of both the explanatory and response variables and has no effect on estimates.