When using measures of association it is common to discuss the strength of the association. What is meant by the strength of an association?

What will be an ideal response?

The strength of an association refers to how different the observed values of the dependent variable are in the categories of the independent variable. In the case of cross-classified variables the strongest relationship possible between two variables is one in which the value of the dependent variable for every case in one category of the independent variable differs from that of every case in another category of the independent variable. We might call such a connection a perfect relationship, because the dependent variable is perfectly associated with the independent variable; that is, there are no exceptions to the pattern. If the results can be applied to future observations, a perfect relationship between the independent and dependent variables enables a researcher to predict accurately a case’s value on the dependent variable if the value on the independent variable is known. A weak relationship would be one in which the differences in the observed values of the dependent variable for different categories of the independent variable are slight. In fact, the weakest observed relationship is one in which the distribution is identical for all categories of the independent variable?in other words, one in which no relationship appears to exist. There may be a slight (but not nil) relationship, a strong (but not perfect) relationship, or a “moderate” relationship between two variables. Deciding which the case is requires the analyst to examine carefully the relative frequencies and determine if there is a substantively important pattern. When asked, “Is there a relationship between X and Y,” the answer will usually not be an unequivocal yes or no. Instead, the reply rests on judgment and interpretation.