A Crowdsourcing Approach to Collecting 399 Tutorial Videos on Logarithms

We present work on how to teachersource novel tutorial videos on topics related to logarithms. Specifically, we created a Human Intelligence Task (HIT) for the Amazon Mechanical Turk and collected 399 unique explanatory videos from 66 unique teachers in approximately 4 weeks. Coding of the videos is still ongoing, but initial analysis suggests significant variety of presentation format, pedagogical style, and language. In a follow-up experiment to assess the pedagogical effectiveness of the videos, we found that the best videos were statistically significantly more effective at increasing students' learning gains (posttest minus pretest) compared to a control video on a math topic unrelated to logarithms. The next step in the project is to create an intelligent decision-engine to assign tutorial videos to students based on joint properties of the video, the student, and the teacher.

[1]  Neil T. Heffernan,et al.  The Future of Adaptive Learning: Does the Crowd Hold the Key? , 2016, International Journal of Artificial Intelligence in Education.

[2]  Krzysztof Z. Gajos,et al.  Learnersourcing subgoal labeling to support learning from how-to videos , 2013, CHI Extended Abstracts.

[3]  Chih-Ming Chen,et al.  Intelligent web-based learning system with personalized learning path guidance , 2008, Comput. Educ..

[4]  Janet Wiles,et al.  Characterizing the temporal dynamics of student-teacher discourse , 2012, 2012 IEEE International Conference on Development and Learning and Epigenetic Robotics (ICDL).

[5]  Neil T. Heffernan,et al.  AXIS: Generating Explanations at Scale with Learnersourcing and Machine Learning , 2016, L@S.

[6]  Krzysztof Z. Gajos,et al.  Understanding in-video dropouts and interaction peaks inonline lecture videos , 2014, L@S.

[7]  T. Dee.  A Teacher Like Me: Does Race, Ethnicity, or Gender Matter? , 2005 .

[8]  Tiffany Barnes,et al.  Toward Automatic Hint Generation for Logic Proof Tutoring Using Historical Student Data , 2008, Intelligent Tutoring Systems.

[9]  Neil T. Heffernan,et al.  The MOOClet Framework: Improving Online Education through Experimentation and Personalization of Modules , 2014 .

[10]  Krzysztof Z. Gajos,et al.  Crowdsourcing step-by-step information extraction to enhance existing how-to videos , 2014, CHI.

[11]  Albert T. Corbett,et al.  Intelligent Tutoring Systems , 1985, Science.

[12]  Christoph Peylo,et al.  W2 - Adaptive and Intelligent Web-Based Education Systems , 2003, Intelligent Tutoring Systems.