Discriminant function analysis is just one type of multivariate statistical analysis. In discriminant function analysis a linear combination of independent variables are used to predict a categorical variable. This type of analysis is basically the reverse of a multivariate analysis of variance (MANOVA), where categorical variables are used to predict continuous variables.

To conduct a discriminant function analysis in SPSS click on the "Analyze" button at the top of the SPSS window. Then select "Classify", followed by "Discriminant".

The discriminant function analysis module will then pop up. Take note of the entry boxes labelled "Grouping Variable" and "Independents". Variables will have to be entered into these boxes before SPSS will all you to run the discriminant function analysis.

To run the analysis, the first step is to identify the categorical variable that you would like to like to predict. Enter this variable into the box labelled as "Grouping Variable". Then click on the "Define Range" button that is just below the "Grouping Variable" entry box and define the range of values for the variables. You can double check the range of values for the categorical variable that you selected by going back to your data file in SPSS. Go to the tabs at the bottom of the SPSS window and select the "Variable View" tab. Then find to the row containing the categorical variable that you entered into the "Grouping Variable" box and click on the cell in that row that matches up with the "Values" column. Click on that cell, and then you will see "…" appear in the cell. Click on the "…" and a window will pop up showing the codes used for each level of the variable. Now you can confirm whether you defined the range of this variable correctly.

Next enter the predictor independent variables into the entry box labelled "Independents". At this point you can run the analysis by pressing the "OK" at the bottom of the module. However, before you run the analysis it is a good idea to explore the buttons off to the right side of the module. For example, to obtain descriptive statistics click on the "Statistics" button, which will prompt a new window to open that includes an option for descriptive statistics.