This month, we’re thrilled to feature Paddy in a double spotlight on our blog! In this post, he delves into the findings of a qualitative study examining the learning experiences and attitudes of Psychology lecturers as they undertook an intensive course to develop R programming skills for data analysis.

Patrick (Paddy), a Senior Lecturer (Education) in the School of Psychology, Queen’s University Belfast, also leads the Research and Scholarship Special Interest Group (SIG) at RoSE. Learn more about his work and contributions here.

It is predicted that R will eventually replace SPSS as the most popular statistical software in science-based degrees and data science (Lindeløv, 2019; Muenchen, 2023). While studies have explored student attitudes towards learning R in psychology (e.g., Counsell & Cribbie, 2020), research on staff attitudes is limited. This gap is crucial as staff need the skills to teach and supervise students using R. In one study, Coetzee and Kagee (2021) investigated staff and student attitudes toward learning R through a six-week workshop. Interviews revealed three main themes; The first theme concerned the Advantages of the R Computer Language, whereby participants highlighted the benefits of learning R, particularly its value in research-related and academic positions. The second theme referred to Challenges Associated with Learning R: Some participants faced difficulties in learning R, citing the steep learning curve and the complexity of the language as major obstacles. These challenges were often compounded by limited prior exposure to programming languages and statistical software. Finally, participants reported Issues Related to the Workshop: Some participants appreciated the structure and content, while others felt that the pace was too fast or that there was insufficient support for beginners. The practical sessions and the hands-on approach were generally well-received, although some suggested improvements in the organization and delivery of the material. However, the study did not solely ask staff about their experiences of using R, making it difficult to extrapolate any firm conclusions on their attitudes.

The aim of the current study was to survey staff members, who embarked on an intensive two-week course on learning R, about their attitudes towards learning R. Qualitative data was obtained from seven participants (Psychology Lecturers from Queen’s University Belfast), using a Qualtrics survey, in which they were asked to indicate what has been the most challenging aspect of learning R over the duration of the course. Participants’ responses were analysed using thematic analysis (Braun & Clarke, 2019). Responses were coded inductively by a primary rater (Chandra & Shang, 2019), who generated an initial set of codes. The rater refined these codes through multiple iterations to minimize the number of distinct codes while preserving their meaningfulness. Inter-rater reliability was calculated (98% agreement). Disagreements were resolved through discussion. Thematic analysis was then applied to these finalised codes in order categorise these codes into meaningful themes.

The data yielded two themes, which are described in Table 2. The first theme, Memory Overload During Course, encapsulates the challenges faced by learners of the R statistical program, highlighting the difficulty of memorizing vast amounts of information in a short period. The second theme, The Cognitive Challenges of Mastering R, reflected the cognitive hurdles associated with learning and utilizing R. It acknowledges the difficulty in memorizing functions and code that are necessary for proficiency in R. It also reflects the struggle with relearning formulas and theory that are necessary to conduct the operations required by R.

These themes highlight the difficulties participants faced in quickly memorizing vast amounts of information and the cognitive challenges associated with learning and using R. Our findings align with Coetzee and Kagee (2021), who reported students struggled with R's steep learning curve and complexity. Extending these findings to academic staff, our study revealed significant difficulties with information retention and practical application, underscoring the cognitive demands of mastering R. Even those familiar with statistical software faced substantial challenges, suggesting that R's complexity and the need to relearn mathematical theories contribute to a high cognitive load. This indicates a need for R training programs to reduce cognitive overload and enhance practical application.

Despite valuable insights, this study has limitations. The sample size was small, and all participants were from a single institution, limiting generalizability. Future research should include a more diverse sample and compare experiences across different course durations and frequencies. Longitudinal studies assessing long-term retention and application of R skills would provide further insights into effective training aspects.

In conclusion, this study highlights the significant cognitive challenges and memory overload experienced by staff learning R. The findings emphasize the need for instructional approaches that reduce cognitive load and enhance practical application. Addressing these challenges is crucial for ensuring staff can effectively teach and supervise students using R, contributing to the goal of replacing SPSS with R in science-based degrees and data science.