Logistic regression analysis predicts the values on one dependent variable from one or more independent (predicting) variables when the dependent variable is dichotomous (meaning that it divides the respondents or cases into two exclusive groups such as having a particular illness and not having it). It can also be used, and this is its more common aim, to identify which predictor variables do predict the outcome and their comparative value in making the prediction. Logistic regression is used when the measure on the predicted variable is dichotomous. Logistic regression is the more flexible technique because it makes no assumptions about the nature of the relationship between the independent variables and the dependent variable and the predictors do not have to be normally distributed.
Different Types of Logistic Regression
There are three types of logistic regression which are most frequently used: direct, sequential and stepwise. They differ in the rules applied in deciding which independent variables should be included in the final model or regression equation.
In the direct or enter method of logistic regression all the independent variables are forced into the regression equation. This method is generally used when there are no specific hypotheses about the order or importance of the predictors. The direct method is not appropriate if the researcher is trying to test a hypothesis concerning the relative importance of the predictor variables; in this case, the sequential method should be used.
In sequential logistic regression the researcher determines the order in which the independent variables are entered into the model or equation. It allows one to determine the individual predictive value of the independent variables as well as the predictive value of the model as a whole. In the sequential procedure, the independent variable entered in the model first is given priority and all subsequent variables are assessed to determine whether they significantly add to the predictive value of the first variable.
The third procedure, stepwise logistic regression, is an exploratory method concerned with hypothesis generation. The inclusion or exclusion of independent variables in the regression equation is decided on statistical grounds as the model is run.