Comprehensive Content Analysis Dataset of the Grade 10 "History 1" Textbook: A Four-Stage Collaborative Adjudication Protocol (CAP) Based on the William Romey Method

Textbooks remain central to secondary education, yet conventional auditing methods frequently fail to capture multi-modal engagement or establish multi-rater analytical consensus. This empirical study presents a comprehensive content analysis dataset evaluating the Grade 10 "History 1" textbook (Humanities track) in Iranian secondary education. Grounded in the William Romey quantitative assessment methodology, the investigation extends beyond textual narratives to examine four core curricular dimensions: reading passages, visual illustrations, integrated learning activities, and end-of-chapter evaluation questions. To ensure methodological rigor and mitigate individual or algorithmic biases, the dataset was constructed using a novel Four-Stage Collaborative Adjudication Protocol (CAP). Implemented within a human-AI hybrid intelligence workflow, pre-service teacher research teams at Farhangian University operated across four institutionalized roles—First Auditor, Second Auditor, Prompt Engineer, and Documentation Manager—thereby positioning student-teachers as Intelligent Supervisors. Quantitative metrics, specifically the Romey Involvement Index, were systematically derived across all textbook components to measure the curriculum's capacity for fostering active learning, student agency, and inquiry-based thinking. Anchored in open-access methodological protocols, this standardized dataset offers an empirical benchmark for curriculum developers, educational auditors, and policy researchers. Ultimately, the study demonstrates a scalable framework for multi-agent curriculum auditing while enhancing professional identity and technological agency in pre-service teacher education.