Empirical Support for a new Taxonomy of Sadness: a Factor Analysis

Sadness is a complex emotion and has undergone several reclassifications. The structure and composition of sadness has been broken down into several differential sad states - differing in valence, arousal, outcomes, and situational context. This classification of sad states sought to explain the differential sad states experienced by individuals and how this impacts experiences of depression. This classification, however, was completely theoretical. In this study, a sad states measure (SSM) was designed and underwent exploratory and confirmatory factor analyses, with the aim of providing empirical evidence for the existence of these sad states. Data from 748 participants was collected. The exploratory factor analysis used a random split half of the sample, while a confirmatory factor analysis used the other split half. Ultimately, a seven-factor solution was extracted comprising latent constructs labelled: grief, melancholia, bleak, empathic sadness, sweet sorrow, comforting sorrow, and sublime sorrow. Strong goodness-of-fit measures supported the factor model and the observed factors aligned closely with the sad states proposed in previous literature. The individual states were then entered into in a regression analysis with depression as the outcome variable. Three states were found to be significant predictors of depression. These findings provide empirical evidence of differential sad states and suggest a close link between specific sad states and depression – suggesting a potential clinical impact of this understanding.