Measurement in the Ultimatum and Dictator Games: Construct validity, standardization and the problem of coordination

Abstract The ultimatum game (UG) and the dictator game (DG) are two of the most prominent tools in experimental and behavioural social sciences. Originally designed to study bargaining behaviour and decision-making in economics, these simple yet versatile scripts have come to be viewed by many as reliable instruments for measuring prosocial preferences, such as altruism, generosity, or sensitivity to social norms. Proponents of the measurement view have argued that the UG and DG serve as standardized “social thermometers,” capable of revealing cultural and contextual differences in normative behaviour across diverse populations. For instance, the UG has been used to demonstrate the influence of institutional contexts and cultural norms on fairness expectations, while the DG has been described as a straightforward measure of altruism. In this paper, we critically examine the idea that the UG and DG function as robust measuring devices, raising key questions about their validity as tools of measurement. We argue that the constructs these games are purported to measure—ranging from prosociality to norm compliance—are often imprecisely defined and vary across studies, and this severely limits the sense in which the DG or the UG can be interpreted as measuring devices. By analysing the ontological and methodological challenges surrounding these games, we contend that the UG and DG are better understood as tools for exploratory experimentation rather than as reliable instruments for systematic measurement.