Bias in LLMs as Annotators: The Effect of Party Cues on Labelling Decision by Large Language Models
This reveals a critical bias problem in LLM-based annotation for political analysis, which is incremental but important for researchers and practitioners using automated labeling.
The study investigated whether Large Language Models exhibit political bias when used as annotators, finding that LLMs use party cues to judge political statements and reflect human training data biases, with significant bias even for center-left and center-right parties unlike humans.
Human coders are biased. We test similar biases in Large Language Models (LLMs) as annotators. By replicating an experiment run by Ennser-Jedenastik and Meyer (2018), we find evidence that LLMs use political information, and specifically party cues, to judge political statements. Not only do LLMs use relevant information to contextualize whether a statement is positive, negative, or neutral based on the party cue, they also reflect the biases of the human-generated data upon which they have been trained. We also find that unlike humans, who are only biased when faced with statements from extreme parties, LLMs exhibit significant bias even when prompted with statements from center-left and center-right parties. The implications of our findings are discussed in the conclusion.