A New Model of Justice Evaluations: Using Graded Status Characteristics to Estimate Just Rewards

Abstract

In this dissertation I examine the link between status and perceptions of just rewards. Specifically I focus on how an individual's status-valued attributes shape their perceptions of just rewards, or the amount of a good that they deem fair. According to equity theorists, status-valued attributes constitute one 'input' that shapes perceptions of just rewards, but the precise nature of this relationship has been heretofore unspecified. Drawing from reward expectations theory, which is one of the equity theories, I develop a set of equations to estimate point predictions of just rewards based on individual's status-valued attributes. The model quantifies the commonly held belief that individuals with the more positively evaluated states of status-valued attributes expect to receive relatively more rewards from a distribution of valued goods. The model borrows the quantification of reward expectations states from reward expectations theory, which requires reducing all status differences to two states of relatively high and relatively low. This is an unnecessary simplifying assumption that requires throwing away the relative magnitude of status-valued attributes. In the interest of increasing the precision and realism of the formal model of just rewards, I also extend the mathematics of reward expectations theory to account for status-valued attributes with more than two states (e.g., occupational prestige or education). This extension not only increases the precision of the formal model of just rewards, but is also applicable to all of the expectation states theories, which account for a large body of scholarship and have a broad domain of applicability. To evaluate these ideas I use a variety of quantitative methodologies, including an experiment, a vignette study and the analysis of secondary data from thirteen countries. Across these methods, I find support for both the formal model of just rewards and the procedure for modeling status-valued attributes with more than two states. I conclude the dissertation with the implications of this research and future directions of the project.