Inter-rater reliability data of classroom observation: Fidelity in large-scale randomized research in education

This dataset belongs to a large-scale randomized controlled trial (RCT) in educational research targeting English learning students and their teachers' instructional capacity. The dataset includes ratings conducted through classroom observations of 45-minute English as a Second language (ESL) blocks. Each coder rated 60 recorded video segments collected from each teacher. During the 20-second segment, ratings of six domains of teachers' instruction (i.e., ESL Strategies, Group, Activity Structure, Mode, Language Content, Language of Teacher, Language of Student) were collected. The dataset is organized by teacher, by coder, and by domain, for researchers to analyze inter-rater reliability among coders by domain and/or cross-domain. This data article is related to the research article Tong et al. [3] on “The determination of appropriate coefficient indices for inter-rater reliability: using classroom observation instruments as fidelity measures in large-scale randomized research”.