Determine the independent (factor) and dependent variables of the data. These values should be represented numerically in two columns. Using the cholesterol difference among age group example, age would be the independent variable, or factor, with the cholesterol being the dependent variable. The ages must also be represented as groups, such as decades.
Click on the "Analyze" tab, "Compare means" list, and "One-way ANOVA" to bring up a dialog box.
Click the dependent variable column from the list on the right placing it in the "Dependent List" box.
Click the independent variable column from the list on the left placing it in the "Factor" box.
Click on the "Post Hoc" button to distinguish the specific type of procedural test needs to be performed based on whether there are equal or unequal variances (i.e. Tukey or Bonferri test). Consult a statistical reference to determine the specific type of test needed for the particular data set being tested.
Click "Continue," "Options" to bring up the "Options" menu to check which additional output tables will be produced. These include "Descriptives" and "Homogeneity of Variance."
Click "Continue," "OK" to produce the AVOVA results and the previously checked data output tables. The tables will appear in order followed by the ANOVA results table.
Interpret the values to report in the results or to identify additional tests that may be warranted. A guide to interpreting the results of the ANOVA analysis is provided in the Resources section.