The independent variable is the factor that an experimenter manipulates. It is the presumed cause whose effect on the dependent variable is under investigation.

In a study of the effect of sleep on memory, the independent variable is the amount of sleep allowed; participants are assigned to different levels of that variable. In a drug trial the independent variable is the dose or the presence of the active compound. Clear operational definitions specify exactly how the independent variable is manipulated so that the study can be replicated.

A true experiment requires that the experimenter control the independent variable and assign participants to its levels, preferably at random. When the variable of interest cannot be manipulated—age, sex, personality—the study becomes correlational or quasi-experimental, and causal conclusions are more limited. Confounding variables must be controlled so that changes in the dependent variable can be attributed to the independent variable rather than to unintended factors.

The independent variable is the question the experiment asks. The dependent variable is the answer it records.