Home Statistical Dictionary Extraneous Variable
Extraneous Variable
Any variable other than the independent variable under study that could potentially influence the dependent variable, a broad category that includes confounding variables as the specific subset that is also associated with the independent variable itself.
In Plain English
An extraneous variable is the broadest possible category: anything at all, besides the specific factor you're deliberately studying, that could plausibly affect your outcome. Room temperature during a cognitive test, a participant's mood that day, the time of day an experiment is run, all extraneous variables in a study of, say, memory performance. Not every extraneous variable is a problem for causal inference, only the subset that also happens to be associated with the independent variable you're studying, the confounders, actually distort your ability to draw a valid conclusion. A good experimental design tries to control extraneous variables broadly, whether by holding them constant, randomizing them away, or measuring and adjusting for them, precisely so that any of them that might otherwise become confounders are neutralized before they can distort the result.
Definition
An extraneous variable is any variable other than the independent variable (the exposure, treatment, or predictor under primary study) that could plausibly influence the dependent variable (outcome), a category considerably broader than confounding variables specifically. An extraneous variable becomes a confounding variable specifically when it is also statistically or causally associated with the independent variable itself, an extraneous variable that varies randomly and independently of the treatment or exposure being studied adds noise (increasing variance, reducing statistical power) but does not bias the estimated effect, whereas one correlated with the independent variable introduces genuine bias; good experimental and study design aims to control extraneous variables broadly, whether by holding them constant across conditions, randomizing their distribution across groups, statistically adjusting for them, or using a matched pairs design, precisely to prevent them from becoming confounders and to reduce the unexplained noise they would otherwise add to the analysis.
Properties
- Whether an extraneous variable actually threatens a study's validity depends entirely on whether it is correlated with the independent variable being studied, an extraneous variable that varies randomly and independently across treatment groups only adds statistical noise, reducing power but not biasing the estimated effect, it is only when an extraneous variable is also associated with the treatment that it graduates to being a genuine confounder.
- Good experimental design manages extraneous variables through several complementary strategies, holding them constant (testing everyone under identical conditions), randomization (spreading their influence evenly, in expectation, across groups), matching, or statistical adjustment, the choice among these depends on whether the extraneous variable is even measurable and controllable at the design stage.
- Every confounding variable is an extraneous variable, but not every extraneous variable is a confounding variable, this containment relationship (confounders are a strict subset of extraneous variables) is the key distinction between the two closely related but non-identical concepts, and is a common point of confusion in introductory research methods teaching.
At a Glance