According to the University of Connecticut, the criterion variable is the dependent variable, or Y hat, in a regression analysis. The criterion variable is the variable that the analysis predicts. The number given from the analysis fits into the regression line.
A dependent variable is a variable under manipulation. This is the variable that researchers want to see change. The measure of this change is indicative in the regression analysis. Independent variables correlate to a change in the dependent variable. The independent variables are the variables in the study that cannot be manipulated, such as gender or age. In a regression analysis, the independent variables are also the predictor variables. The analysis measures the change, or impact, of each predictor variable on the criterion variable. Researchers then input the information from each impact into a regression equation to determine the possible change in each criterion variable for any case that is not in the study. This allows researchers to determine the change in Y hat from the independent variables.
In a multiple regression analysis, the criterion variable is under the influence of more than one independent variable. Sometimes, in this type of analysis, an interaction between two independent variables correlates to a change in the criterion variable.