ABSTRACT
Measures of association give us an indication of how closely two variables are related in the sense of how the variables change together. Nominal data are frequently encountered in public and nonprofit management, and provide a bit of trouble in higher level analysis. This is simply because as these types of data have no rank order or set interval, they are very difficult to include in mathematical analysis. The Phi coefficient is the easiest of all the measures of association for nominal level data. It is simply equal to the square root of Chi-Square divided by the number of observations. The odds ratio expresses a ratio of the relative probability of the dependent variable equalling 1 for the two levels of the independent variable. Lambda is a statistic based on the idea of proportional reduction in error. Gamma uses the difference between concordant pairs and discordant pairs to estimate the strength and direction of a relationship.
