To achieve internal validity, a researcher must design and conduct experiments where only the _____ variable can be the cause of the results.

A) dependent
B) independent
C) confounding
D) criterion

To achieve internal validity, researchers must carefully control all variables except for the independent variable, which is manipulated to examine its effects on the dependent variable. Confounding variables, which are extraneous variables that vary systematically with the independent variable, can threaten internal validity by providing alternative explanations for the observed results. Therefore, they must be eliminated or controlled to ensure that only the independent variable can be the cause of the results. The criterion variable is a variable that measures the outcome or effect of interest, but it is not relevant to the question of internal validity.