There are five documents included. The first PDF document contains the data analysis scripts and the analysis statement from the educational authority that regulates the research. Each page in this document is stamped for verification purposes. The second TXT document contains the data analysis scripts obtained from R program. The script is exactly the same as the one included in the first document. The third PDF document contains the official statement on the data access and confidentiality, with official stamp for verification. Due to ethical considerations and the regulation of local educational policies, the raw dataset for the study cannot be shared but can be accessed on-site with the educational authority. The last two documents are the speaking tasks used in the study, including a text presentation of task instructions and a test instruction recording.

The data analysis transcript is specifically for the study titled "What Shapes Automated Ratings in Computer-Based English Speaking Tests? Perspectives from Analytic Complexity, Accuracy, Fluency, and Pronunciation Indices". This study adopts a quantitative, correlational research design to investigate the extent to which various linguistic features—namely Complexity, Accuracy, Fluency, and Pronunciation (CAFP)—predict Automated Ratings (ARs) in China’s Computer-Based English Speaking Test (CBEST) administered during Zhongkao. The aim is to uncover how these linguistic indices influence machine-generated scores and to evaluate the validity and fairness of automated assessment systems in high-stakes educational contexts.<br>The CBEST format used in this study includes three task types: Reading-Aloud, Communicative Question &amp; Answer, and Response to a Topic. These tasks are scored using an integrated system developed by iFlytek, which combines automatic speech recognition (ASR), deep learning models, and benchmarked manual expert evaluation. The assessment model has been officially recognized and is widely adopted in Chinese provinces for junior secondary school students.