Evaluating objective and subjective frequency measures in L2 lexical processing
-
Xiaocong Chen [1] ; Yanping Dong [2]
-
[1] Guangdong University of Foreign Studies
Guangdong University of Foreign Studies
China
-
[2] Zhejiang University
Zhejiang University
China
-
- Localización: Lingua: International review of general linguistics, ISSN 0024-3841, Nº 230, 2019
- Idioma: inglés
- Texto completo no disponible (Saber más ...)
-
Resumen
The superiority of the subjective frequency measure over objective corpus frequency measures in L1 lexical research has long been debated, but a systermatic investigation into this issue in the L2 context is still lacking. To bridge the gap, the present study explored the relations between six typical English corpus frequency norms and subjective frequency ratings collected from L2 English learners, and assessed their predictive power on L2 English lexical-decision data. The results revealed that despite strong correlations between the two types of frequency measures, larger discrepancies were detected in the lower frequency range, and that in general the subjective measure explained extra variance of L2 lexical processing data, especially for low-frequency words. But for high-frequency words, some corpus frequency norms (e.g., SUBTLEX-US) also showed equally good predictive power. Methodological implications for the selection of frequency measures in L2 research are discussed.
