When I started my PhD, I had no experience in programming or working with educational data. Over the course of my PhD, I learned these skills and this chapter aims to help others in a similar position. The target audience for this chapter includes researchers working in educational fields, particularly in Learning Analytics (LA), who have little or no experience with student trace data.
Beginners can often provide better tutorials and explanations on a subject compared to experts. This is because beginners can offer fresh perspectives, use relatable language, and provide simplified explanations that resonate with other novices. They also tend to remember the challenges they faced while learning, which helps them anticipate and address common obstacles. Additionally, beginners are less likely to overlook basic concepts that experts might take for granted.
In this chapter, I will present all the steps I followed in small, digestible chunks, supplemented with dataset examples and code snippets. I will use data collected from students using a specific Learning Management System (LMS), Canvas, and employ R software for the analysis. However, the concepts and explanations can be applied to other LMSs or programming languages. While programming is not the main focus of this tutorial, it is necessary. Part 1 focuses more on understanding the basics of R (and programming in general) and can be skipped by experienced programmers; for this purpose it is not included in my dissertation. Part 2 focuses on cleaning and pre-processing the trace data in order to make it ready for the final analysis, thus, it is included in my dissertation.
This chapter reflects my personal journey, from the first time I opened an educational dataset to the publication of scientific papers based on it. If you are seeking specific analyses and expert explanations, I highly recommend this source, which I consider the best guide to Learning Analytics through R available today (https://lamethods.github.io/).
Another purpose of this chapter is to introduce the methodology employed in the subsequent two chapters, Chapter 4 (link to GitHub: https://github.com/Tudor-Cristea/Unobtrusive-COPES-paperand and paper: https://link.springer.com/article/10.1007/s10639-023-12372-6) and Chapter 5 (LINKS). Although each chapter includes its own methodology section, this chapter provides a comprehensive overview of the steps taken before reaching that point, detailing my personal process and understanding.