Independent Variable

In survey research, an independent variable is thought to influence, or at least be correlated with, another variable: the dependent variable. For example, researchers hypothesize that childhood exposure to violent television can lead to violent behavior in adulthood. In such a study, exposure to violent television programming as a child is an independent variable and violent behavior in adulthood is the dependent variable.

An independent variable is commonly denoted by an x and a dependent variable by y, with the implication that “x causes y” or, in the case of noncausal covariation, “x is related to y.“

Determining whether one variable influences another is of central importance in many surveys and studies, as making this determination helps researchers accept or reject hypotheses and thereby build social science knowledge. Relationships between variables help researchers to describe social phenomena.

In experimental studies, with random assignment of respondents to experimental conditions, a researcher can choose which variables are independent, because these are the variables controlled by the researcher. In population studies, patterns in data help researchers determine which variables are independent.

More than one independent variable may influence a dependent variable. Quantitative tools and approaches can assist researchers in accepting or rejecting their hypotheses about the relationships among independent variables and a dependent variable. In some analyses, researchers will “control for” the influence of certain independent variables in order to determine the strength of the relationship for other independent variables.

Using the example of childhood exposure to television violence again, another independent variable in the study could be parental control over television viewing. Yet another independent variable could be level of physical violence in the home. The complexity of a hypothesized causal model such as this increases with the number of independent variables and interaction effects among independent variables.