Equipping college students with statistical literacy under the massification of higher education

Abstract

Big data analysis has become an essential decision-making tool across various sectors, with statistics serving as the critical knowledge base. However, many college students, especially those with weaker mathematical skills, experience anxiety about statistics. This study explores strategies to improve statistical literacy among such students. Using a questionnaire survey, it focuses on an introductory statistics course at a university in Taiwan, applying the Partial Least Squares method to test the structural equation model. The findings suggest that integrating statistical software into instruction, conducting frequent formative assessments, and designing test content conducive to learning can enhance students' perceived learning outcomes and self-efficacy. Despite limitations like small sample size and representativeness, the study highlights that statistical software not only aids learning but also develops practical skills for real-world application. Open-book exams are recommended, but multiple assessment methods should be used to ensure fairness. The study concludes that while statistics anxiety is difficult to alleviate, enhancing self-efficacy can improve students' attitudes toward statistics. It proposes scaffolding strategies to support students with low mathematical skills, ultimately improving their statistical literacy and confidence.