On the Credibility of Evaluating LLMs using Survey Questions
This work addresses the problem of unreliable evaluation methods for LLMs in social science research, which is incremental by refining existing approaches.
The paper identifies methodological limitations in evaluating large language models (LLMs) using survey questions, showing that prompting and decoding strategies significantly affect results and that high average agreement with human data does not guarantee structural alignment in responses, as demonstrated with the World Value Survey across languages and countries.
Recent studies evaluate the value orientation of large language models (LLMs) using adapted social surveys, typically by prompting models with survey questions and comparing their responses to average human responses. This paper identifies limitations in this methodology that, depending on the exact setup, can lead to both underestimating and overestimating the similarity of value orientation. Using the World Value Survey in three languages across five countries, we demonstrate that prompting methods (direct vs. chain-of-thought) and decoding strategies (greedy vs. sampling) significantly affect results. To assess the interaction between answers, we introduce a novel metric, self-correlation distance. This metric measures whether LLMs maintain consistent relationships between answers across different questions, as humans do. This indicates that even a high average agreement with human data, when considering LLM responses independently, does not guarantee structural alignment in responses. Additionally, we reveal a weak correlation between two common evaluation metrics, mean-squared distance and KL divergence, which assume that survey answers are independent of each other. For future research, we recommend CoT prompting, sampling-based decoding with dozens of samples, and robust analysis using multiple metrics, including self-correlation distance.