Learning statistics based on the compendium and reproducible computing

Abstract—This paper explains how the R Frame-work (wessa.net) and a newly developed Com-pendium Platform (freestatistics.org) allow us to cre-ate, use, and maintain documents that contain em-pirical research results which can be recomputed and reused in derived work. It is illustrated that this tech-nological innovation can be used to create educational applications that can be shown to support effective learning of statistics and associated analytical skills. It is explained how a Compendium can be created by anyone, without the need to understand the technical-ities of scientific word processing (LATEX) or statistical computing (R code). The proposed system allows ed-ucational researchers to objectively measure key as-pects of the actual learning process based on individ-ual and constructivist activities such as: peer review, collaboration in research, computational experimen-tation, etc. The system was implemented and tested in three statistics courses in which the use of Com-pendia was used to create an interactive e-learning environment that simulated the real-world process of empirical scientific research.