ABSTRACT
This chapter analyzes the relationships between variables, but many of the decisions about what kind of analysis to use are driven by the types of variables being analyzed. Many of the more sophisticated techniques require that one or more of the variables are at the interval or ratio levels; when dealing with nominal- or ordinal-level variables, simpler techniques often must suffice. The most basic way to look at the relationship between two variables is through a crosstab: essentially a frequency table for two variables at the same time. To test the relationship between having a homosexual friend or relative and support for a particular issue, a researcher could run a crosstab. Based on this relationship and other ways of splitting up the data, he concluded that there was no significant relationship between the unemployment rate on Election Day and how many votes the incumbent party won in the Presidential election.
