Integrating the evaluation of out of the platform autoevaluated programming exercises with personalized answer in Open edX

摘要

This paper describes a procedure to integrate personalized self-evaluated programming exercises created in an external programming interactive environment using a standard problem type of a MOOC platform, making use of the anonymized identifier provided by the platform. We will explain how to integrate auto evaluated programming exercises with personalized answers created with Python notebooks, using the standard problem types that Open edX provides. We will review the alternatives we evaluated and why we discarded them and explain our final workflow with an example problem in the edx platform. In our workflow the autoevaluated programming exercises are created as if we were doing some test-driven development where the problem is our functionality and the unit tests are actually the verifications done to generate hints and evaluate the students. Once the problem is designed the unit tests create a code, based on the answer and the Anonymous userID, code that is obfuscated using an encryption technique. That code is used as the answer of a standard Open edX problem, creating a completely automated personalized environment and avoiding the use of Open Response Assessment tools that depend on the correction of other students.