ENVIRONMENTAL EDUCATION ASSESSMENT WITH SEMANTIC NETWORK ANALYSIS: EXAMPLE FROM A UNIVERSITY IN THE GUANGDONG PROVINCE

This study explores how to effectively evaluate the efficiency of environmental education (EE) among college students in Guangdong Province by comparing two complementary methods: (1) traditional statistical analysis of structured questionnaires using chi-square tests, and (2) semantic network analysis (SNA) of open-ended responses. The structured questionnaire assessed students' environmental knowledge, attitudes, behaviors, and education needs using multiple-choice and Likert-scale items. An open-ended question was included to capture students’ understanding of ""environmental protection"" in their own words, which was then analyzed through semantic network modeling.
The study collected 225 valid responses from students at Guangdong University of Science and Technology. Chi-square tests revealed significant differences between first-year students and those in later years in their knowledge of environmental issues (e.g., χ², p < 0.05 across 35 indicators), but no significant differences were observed based on gender or place of residence. Semantic network analysis, using centrality, density, and clustering coefficients, showed similar cognitive divergences between academic years but confirmed no meaningful structural differences in conceptual associations across gender or residence groups. For instance, the semantic network of first-year students had lower density (0.031) and fewer central nodes than that of senior students, indicating a simpler understanding of environmental concepts.
The convergence of results from both methods validates the use of semantic network analysis as a complementary, quantitative tool for evaluating students’ cognitive structures and discourse in environmental education. This integrated approach offers a richer and more nuanced perspective than traditional surveys alone.
The findings suggest that environmental education efforts should focus more on advancing conceptual depth as students progress through university. Curriculum designers can use semantic analysis as an innovative diagnostic tool to improve the cognitive and discourse-based effectiveness of environmental education programs.