In applied linguistics research, complexity, accuracy, and fluency (CAF) were based on the importance of task-based activities, task familiarity, communicative competence, or sociolinguistic factors. However, investigating CAF as an independent variable to assess language performance is controversial in research. Some studies found that language cannot be assessed in these dimensions alone since language is complex, changing, and dependent on many factors. Further research shed light on the importance of reconstructing these variables. The purpose aims to test the validity of a speaking assessment scheme based on previous research suggestions to reconstruct CAF measures by comparing it to the TOEFL speaking rubric. There were no statistical differences in the results of the participants in both assessment schemes. The implications are to further design valid assessment schemes based on reconstructed CAF measures, using an automated application that can transcribe and compute these measures instead of calculating them manually.
Read full abstract