A correlation analysis differs from a regression analysis in the sense that a correlation analysis does not impose a causal relationship between two variables. Practically, there is no need to define the dependent variable and the independent variable in a correlation analysis. Very often a correlation analysis is used as a preceding step to regression analysis in order to discover the overall trend between two variables. Regression analysis goes a step further than correlation analysis because regression analysis aims at explaining the variation of the dependent variable by another variable called the independent variable.