Sociodemographic Prompting is Not Yet an Effective Approach for Simulating Subjective Judgments with LLMs (2311.09730v2)

Abstract: Human judgments are inherently subjective and are actively affected by personal traits such as gender and ethnicity. While LLMs are widely used to simulate human responses across diverse contexts, their ability to account for demographic differences in subjective tasks remains uncertain. In this study, leveraging the POPQUORN dataset, we evaluate nine popular LLMs on their ability to understand demographic differences in two subjective judgment tasks: politeness and offensiveness. We find that in zero-shot settings, most models' predictions for both tasks align more closely with labels from White participants than those from Asian or Black participants, while only a minor gender bias favoring women appears in the politeness task. Furthermore, sociodemographic prompting does not consistently improve and, in some cases, worsens LLMs' ability to perceive language from specific sub-populations. These findings highlight potential demographic biases in LLMs when performing subjective judgment tasks and underscore the limitations of sociodemographic prompting as a strategy to achieve pluralistic alignment. Code and data are available at: https://github.com/Jiaxin-Pei/LLM-as-Subjective-Judge.