文本心理测量学：利用自然语言处理评估心理构念

Large language models (LLMs) have revolutionized natural language processing (NLP). Yet when used to assess psychological constructs in text, they are generally not evaluated for the types of validity, reliability, and standardization typically expected from traditional questionnaires with rating scales. This study bridges that gap by proposing a framework for evaluating the psychometric properties of NLP-based psychological assessments (including LLMs-based ones), an area of research we refer to as Text Psychometrics. We first review different NLP methods, compare their ability to address key