The use of a control involves the Spss > Analyze > Descriptive Stats > CrossTabs after choosing your row and col variables, then consider what control variable to put in the "Layers" window. When you get your tables (with Statistics=Chi-Sq and Tau-b and with Cells=add row %s) you will see each row by column table is repeated for different values of the control (layer) variable. So are the statistics. This allows you to check whether (1) the row-col correlations replicate for different categories of the control or (2) the values control variable makes a difference to the row-col correlations Dont be surprised when you get less significance for tables with controls, since the sample sizes have dropped in each case ... read Chapter 5 carefully to understand why this is, and then read Chapter 7