The rationality of choices is one of the most fundamental assumptions of traditional economic analysis. Yet, substantial evidence has documented that choices often cannot be rationalized by utility maximization. Several measures of rationality have been introduced in the literature to quantify the size of rationality violations. However, it is not clear which of these measures should be used in applications, and many measures are computationally very demanding, which has restricted their widespread use. First, we introduce novel variations of the measures that allow us to establish connections between the different measures. Second, we develop methods to compute the most-used measures of rationality. Exploiting this computational progress, we offer simulation-based comparisons of the accuracy of the measures. These simulations show that a new type of measure that combines the size of rationality violations with the number of rationality violations outperforms other measures. Finally, we offer a method to calculate statistical significance levels for rationality violations.
