ABSTRACT
Once administrators go beyond studying a single variable, their interest lies in studying the relationships among variables. Examining bivariate relationships is typically only a first step in studying complex relationships. With a little practice, administrators can easily interpret contingency tables. Nevertheless, these tables quickly become cumbersome as more variables are added to the analysis. None of the contingency table examples in the chapter on nominal and ordinal variable relationships had more than three variables, and the control variables only had two values. Try to visualize what the tables would look like if you wanted to control for several variables at the same time or if you were studying a control variable that had eight or nine values. Interpretation would become elusive. Furthermore, contingency tables may not be the best way to analyze interval-level data.
