Introducing Copula as a Novel Statistical Method in Psychological Analysis.

During the past decades, the relationship between various psychological parameters had been studied in detail. However, the dependency structure of correlated parameters was rarely investigated. Knowing the dependence structure helps in finding the probability matrix of the interaction between the parameters. In this research, a novel approach was introduced in psychological analysis using copula functions. For this purpose, the self-esteem and anxiety of 141 university students in Iran were extracted using the Coopersmith Self-esteem Inventory and the Zang Anxiety Scale. Then the dependence structure of self-esteem and anxiety were established using copula functions. The Frank copula achieved the best fit for the joint variables of self-esteem and anxiety. Finally, the probability matrix of different classes of anxiety, taking into account self-esteem classes, was extracted. The results indicated that poor self-esteem leads to severe or very severe anxiety, with more than 98% probability, while strong self-esteem may lead to normal and mild anxiety, with about 80% probability. It can be concluded that the method was promising, and that copula functions can open a window to the dependence structure analysis of psychological parameters.