On the problem of nominal data correlation

This article suggests a new methodological approach to calculating the correlation coefficient between nominal data. In the course of the study, it was revealed that most of the work with nominal data uses two Yule coefficients and Pearson coefficient. Moreover, these coefficients are proposed based on the same assumption that there is no correlation between the variables. As a result of the study, this assumption is questioned by the authors, and a new coefficient of correlation between nominal data is proposed. A comparative analysis of the application of the three classical coefficients with the new coefficient is conducted. The advantages of the new coefficient are shown.