A lurking variable is a variable that is not included in a statistical analysis but can still affect the outcome of that analysis. These variables can create problems by falsely identifying a strong relationship between variables or hiding the true relationship. Lurking variables earned their name because they often go undetected and hide beneath the surface of the variables researchers are interested in studying. They can cause omitted variable bias only when they correlate with both the dependent variable and an independent variable and are not included in the model.
Lurking variables can either hide an existing relationship between variables or create a phantom correlation where none exists. They are extraneous variables that may make the relationship between dependent variables and independent variables seem other than it actually is. In statistical models, the error term explains lurking variables that affect the process. To discover lurking variables, researchers must take the time to understand their data and the important variables that can affect a process. They can also create a plot of the data to look for non-linear trends that can identify the presence of lurking variables.
Examples of lurking variables include smoking and stress levels in a study that examines the effect of diet and exercise on a persons blood pressure. Lurking variables can bias the results of a study, and the more severe the effect of the lurking variables on the dependent variable, the more misleading the analysis will be. For research results to be valid, lurking variables must be identified and then either eliminated, held constant, or included in the study. They can be identified with regression analysis: plot the residuals, and if you see a trend (either linear or non-linear), this is evidence that a particular variable is affecting the response variable.
