Open Up-Vote Assessment for Creative Coding: Model and Quality

Abstract

The assessment of creative coding is challenging in the credibility due to aesthetic reason. It also suffers from manual workload for the large number of enrollments in open online courses. We propose an online open up-vote assessment to tangle the above challenges. Our work has three contributions. First, we propose an approach to assessing creative coding by using online up-vote system in a crowd-sourcing way. Second, a scoring model has been established to convert the number of up-votes and views of creative works into the Wilson Score; the lines of creative codes were combined to evaluate the works on a scale from 0 to 10 points. Third, the scoring quality is evaluated by using real data set which is collected with 337 real-world creative codes and 9 sets of independent scores to them by teachers. The results show that the up-vote assessment has very tight tendency to the scores from teacher. Our work indicates that open online up-vote assessment can be trusted as a flexible and reliable way to evaluate creative works.