Evaluating the Impact of Code-Based Statistical Software Platforms on Undergraduate Learning Outcomes and Experiences in Courses Utilizing the Passion-Driven Statistics Curriculum

This study investigates whether undergraduate students in an introductory statistics course have learning outcomes and experiences that differ based on whether they used code-based or non-code-based statistical software. The sample includes 2,241 students enrolled in a course using the Passion-Driven Statistics curriculum from 61 post-secondary institutions. We compared the course experiences and learning outcomes for students between those that learned either code-based (R, SAS, Stata, Python) or non-code-based (SPSS, Excel, JMP, StatCrunch) platforms, controlling for student demographics and academic background. Students learning a code-based platform reported working harder in the course and finding the course more challenging than those learning a non-code-based platform. However, learning a code-based platform was also positively associated with perceived gains in analyzing data for patterns, feeling excited about learning new concepts, an increase in interest in conducting research, an increase in intentions to pursue advanced course in statistics or data analysis, and a greater interest in working on additional data projects in the future. Comparisons within code-based and non-code-based platforms revealed fewer differences. SPSS users reported more positive outcomes than Excel users, but R and SAS users did not differ in reported outcomes. This study highlights the educational advantages of incorporating code-based statistical software into undergraduate curricula, emphasizing their role in enhancing students' interest in data-driven opportunities and fostering a deeper engagement with statistical learning.