Learning Experiments Using AB Testing at Scale
Name
Iswp142-chudzick + posteri.pdf
Size
1.01 MB
Format
Adobe PDF
Checksum (MD5)
458f4023184796b530892fb4dcddb5f7
Author(s)
Chudzicki, Christopher
Pritchard, David E.
Chen, Zhongzhou
Date Issued
March 2015
Journal
Proceedings of the Second (2015) ACM Conference on Learning @ Scale (L@S '15)
Publisher
Association for Computing Machinery (ACM)
Citation
Christopher Chudzicki, David E. Pritchard, and Zhongzhou Chen. 2015. Learning Experiments Using AB Testing at Scale. In Proceedings of the Second (2015) ACM Conference on Learning @ Scale (L@S '15). ACM, New York, NY, USA, 405-408.
Version
Author's final manuscript
Abstract
We report the one of the first applications of treatment/control group learning experiments in MOOCs. We have compared the efficacy of deliberate practice-practicing a key procedure repetitively-with traditional practice on "whole problems". Evaluating the learning using traditional whole problems we find that traditional practice outperforms drag and drop, which in turn outperforms multiple choice. In addition, we measured the amount of learning that occurs during a pretest administered in a MOOC environment that transfers to the same question if placed on the posttest. We place a limit on the amount of such transfer, which suggests that this type of learning effect is very weak compared to the learning observed throughout the entire course.
MIT Department
Massachusetts Institute of Technology. Department of Physics
Terms of Use
Creative Commons Attribution-Noncommercial-Share Alike
Persistent DSpace Link
DOI of Published Version
https://doi.org/10.1145/2724660.2728703
