A Standardized Semantic Feature Norm for 954 Cantonese Concepts
摘要
This study established psycholinguistic norms on Cantonese conceptual semantic features (CanFeat) with 954 concepts and semantic features, including feature frequency, sharing degree, and type of each feature. We provided concept-level variables, including participant count, feature counts (total, distinct, concept-specific, concept-shared), feature proportions, coverage, category labels (superordinate category and subordinate category), familiarity, and concreteness. We validated the reliability, specificity, and categorical structural consistency of the dataset through multiple analyses, including intra-coder reliability checks, gender comparability assessments, semantic category distinctions, and feature-level comparisons with established norms. We also compared CanFeat with norms in other languages, highlighting both similarities and differences. The present norming offers a timely and valuable resource for researchers in psycholinguistics, particularly those focusing on Cantonese production and semantic representation.