An Index System Construction Method Based on CCSS Data Analysis

College students’ learning harvest is a direct reflection of higher education level, and the construction of accurate and reliable index system is the basis and key of follow-up research. Conventional index system construction methods are mostly based on expert opinion consultation, and qualitative analysis is carried out through Delphi method and analytic hierarchy process, which lacks data collection and application of research objects. The construction of index system without the support of real data is not fully convincing. Based on the real data of X University's participation in the CCSS, this paper trimmed the correlation indicators affecting the learning harvest of undergraduates, screened the observed variables with confirmatory factor analysis, and comprehensively considered the correlation and difference between variables through data screening and packaging. Five indicators were obtained: “Quality of Student Involvement (QSI)”, “Emotional Support of Teachers and Students (EST-S)”, “Teaching Practice in Class (TPC)”, “Supportive Campus Environment (SCE)” and “Student Learning Harvest (SLH)”. The established index system has passed the reliability and validity tests of measurement tools, showing good explanatory power and laying a good foundation for follow-up related research.