Mechanisms Influencing Learning Gains Under Information Security: Structural Equation Modeling with Mediating Effect

Abstract

With the expanding enrollments in higher education, the quality of college education and the learning gains of students have attracted much attention. It is important to study the influencing factors and mechanisms of individual students' acquisition of learning gains to improve the quality of talent cultivation in colleges. However, in the context of information security, the original data of learning situation surveys in various universities involve the security of educational evaluation data and daily privacy of teachers and students. To protect the original data, data feature mining and correlation analyses were performed at the model level. This study selected 12,181 pieces of data from X University, which participated in the Chinese College Student Survey (CCSS) from 2018 to 2021. A confirmatory factor analysis was conducted and a structural equation modeling was conducted using AMOS 24.0. Through hypothesis testing, this study explored the mechanisms that influence learning gains from the perspectives of student involvement, teacher involvement, and school support. The results indicated that the quality of student involvement has an important mediating effect on learning gains and that a supportive campus environment has the greatest influence on learning gains. Establishing positive emotional communications between teachers and students is a more direct and effective method than improving the teaching level to improve the quality of student involvement. This study discusses the implications of these results on the research and practice of connotative development in higher education.

A multidimensional lens of environmental consciousness: towards an environmentally conscious theory of planned behavior

“…The measurement model established the proposed model constructs' convergent and discriminant validity and reliability. The goodness of fit was examined using different indicators and thresholds: chi-square (x 2 ) with a degree of freedom below three, root mean residual (RMR) below 0.05, goodness of fit index (GFI) above 0.90, adjusted GFI (AGFI) above 0.80 and root mean square error of approximation (RMSEA) of 0.08 and below (Hair et al, 2015;Wang et al, 2019;Zong et al, 2023). Given that x 2 /df was 2.042, RMR was 0.047, GFI was 0.901, AGFI was 0.872 and RMSEA was 0.050, the measurement model was deemed to be acceptable.…”