In the last three chapters, we have explored relationships between variables. We have asked whether sex could help explain a student’s choice of major, or whether choice of major could help explain a student’s GPA. We have tested whether sex or job type could help explain an employee’s salary. In each case, we started with a null hypothesis that the variables were independent; that one could not help explain the other. We calculated a statistic-Zc, tc, χ2c, or Fc-based on our sample values of these variables. And then, if our statistic was so large that it was unlikely enough, given our null hypothesis, we rejected our null hypothesis and concluded that the variables were not independent; that one could in fact help explain the other.