Integrating R Programming into Introductory Statistics: College Students’ Perceptions of and Attitudes Towards R

• In Event: AER Roundtable Session Friday 1:45 pm - Gold 4
  In Roundtable Session: Reimagining Introductory Statistics: A Research-Practice Partnership Model for Impactful Statistics Education

Abstract

Theoretical Framework and Objectives
This study draws on expectancy-value theory and research on culturally responsive computing to examine students’ motivational beliefs about R programming in statistics and data science education. The goals of this study are twofold: (a) to assess how students’ perceptions of and attitudes toward R programming (e.g., confidence, sentiment, enjoyment, anxiety, perceived difficulty, and openness to R-related careers) shift from the beginning to the end of the course, and (b) to identify which end-of-course beliefs predict students’ future interest in statistics and data science, including how these relationships vary by student background.

Methods
Participants included 412 college students enrolled in introductory statistics courses that integrate R programming into their curriculum using the CourseKata textbook. Students completed surveys at the beginning and end of the term measuring six dimensions of their perceptions and attitudes toward R programming: confidence, positive sentiment, enjoyment, anxiety, perceived difficulty, and openness to R-related careers. We conducted paired-sample t-tests to examine changes over time and multiple regression analyses to predict future interest in statistics and data science based on post-course attitudes. We also examined differences by racial group, focusing on comparisons between racially marginalized and non-marginalized students.

Results
Students reported significant increases in confidence, positive sentiment, and enjoyment with R programming, as well as significant reductions in anxiety and perceived difficulty over the course of the term (see Figure 1). No significant change was observed in students’ interest in R-related future careers. Racially marginalized students reported similar levels of anxiety towards R prior to the course, but then - despite experiencing a significant decline over the course - experienced higher levels of anxiety post-course compared to non-marginalized students. Regression analyses revealed that students’ pre-course anxiety and end-of-course confidence and enjoyment were strong predictors of their future interest in statistics and data science. These predictors held for non-racially marginalized groups, but not for racially marginalized groups where only post-course enjoyment predicted their future interest.

Significance
Findings suggest that integrating R programming into introductory statistics courses can foster more positive student attitudes, particularly in increasing confidence and reducing anxiety. These shifts are meaningful predictors of students’ sustained interest in the field. However, persistent disparities in anxiety - especially for racially marginalized students - highlight the need for psychologically supportive instructional practices. This study contributes to a reimagined vision of statistics education as a foundation for agency, equity, and access, not a barrier. By embedding real-world coding and data analysis into daily instruction, and attending to students’ psychological experiences with tools like R, we can better support diverse learners in developing confidence, curiosity, and future-oriented motivation in data science.

Authors

• Matthew Jackson

• Jisun Min, Yonsei University

• Claudia C. Sutter, CourseKata