In addition to assessing the strength and nature of a relationship, you may want to test hypotheses about the various measures of association. For example, you may want to test the null hypothesis that the value of a measure is zero in the population. This does not involve anything new. You just have to calculate the probability that you would obtain a value as large (in absolute value) as the one you observed if the value is zero in the population. All statistical software packages print as part of their output the observed significance levels for some of the measures of association that we have discussed in this chapter.