Popularity of massive online open courses (MOOCs) allowed educational researchers to address problems which were not accessible few years ago. Although classical statistical techniques still apply, large datasets allow us to discover deeper patterns and to provide more accu-rate predictions of student’s behaviors and outcomes. The goal of this tutorial is to disseminate knowledge on elementary data analysis tools as well as facilitating simple practical data-analysis activities with the purpose of stimulate reflection on the great potential of large datasets. In particular, during the tutorial we introduce elementary tools for using machine learning models in education. Although, the methodology presented here applies in any programming environment, we choose R and CARET package due to simplicity and access to the most recent machine learning methods.