Open Social Student Modeling: Visualizing Student Models with - PowerPoint PPT Presentation

Open Social Student Modeling: Visualizing Student Models with Parallel IntrospectiveViews Sharon, I-Han Hsiao 1 , Fedor Bakalov 2 , Peter Brusilovsky 1 and Birgitta König-Ries 2 1 University of Pittsburgh & 2 University of Jena 2011.07.14 #UMAP2011

Outline This talk is about Open Social Student Modeling #umap2011 ¡ @shoha99 ¡

Introduction Open Student Modeling Provide support in social learning ¡ Allow students to feel in control

Visualization Representation of an individual student Communication and collaboration ¡

SQL-Tutor (Mitrovic , 2007); VisMod (Zapata-Rivera et al., 2004) …

iBlogVis (Vassileva et al.,2009); KeepUp (Webster et al.,2007) …

Framework to apply open user models in adaptive learning environments (Bull & Kay , 2007)

Open Social Student Modeling

Cognitive ~> Social

Progress to guide students to the most appropriate learning content

Implementation + ¡ QuizJET IntrospectiveViews

parallel IntrospectiveViews

QuizJET + IV A selected topic in detail

Study

QuizJET+IV ¡ QuizJET ¡portal ¡ JavaGuide ¡ Hsiao, I., Sosnovsky,S. and Brusilovsky, P. (2010) Guiding Students to the Right Questions: Adaptive Navigation Support in an E-learning System for Java Programming , Journal of Computer Assisted Learning, Volume 26 Issue 4, Pages 270 - 283.

Results Social Guidance mediates the motivating effect of progress visualization. Attempts 160 ¡ 120 ¡ 80 ¡ 25% ¡ ACempts ¡ decrease ¡ 40 ¡ 0 ¡ QuizJET+IV QuizJET+Portal JavaGuide

More conservative increase. Good? Bad?

knowledge-based + social guidance combined are more effective in guiding students to appropriate questions than knowledge-based guidance alone. Community wisdom does matter. Success Rate 100.00% 71.35% 80.00% F(1,32)= 11.303, p <.01 60.00% Success Rate 40.00% 58.31% 20.00% 42.63% 0.00% QuizJET+IV QuizJET+Portal JavaGuide

Peer Guidance peers' ¡models ¡ The ¡more ¡the ¡students ¡compared ¡ 8 ¡ 6.83 ¡ to ¡their ¡peers, ¡the ¡higher ¡post-­‑quiz ¡ scores ¡they ¡received ¡ 6 ¡ 4.67 ¡ ¡(r= ¡0.34 ¡ p=0.004) ¡ 4 ¡ 4 ¡ mean ¡ 2 ¡ 0 ¡ comparison ¡ topic ¡ aCempt ¡

Learning Effect QuizJET ¡+ ¡IV ¡ ¡ ¡ ¡ t1(17)= 7.203, p < .01 ¡ QuizJET ¡on ¡portal ¡ ¡ ¡ ¡ t2(15)= 6.108, p <.01 posttest " pretest NKG = max score " pretest Normalized Knowledge Gain (NKG) posttest " pretest NKG = max score " pretest

Subjective Evaluation A. Usefulness avg 80% é B. Ease of Use C. Ease of Learning 76.9% who ¡did ¡ not ¡ think ¡the ¡comparison ¡mode ¡allowed ¡ them ¡to ¡idenTfy ¡a ¡classmate ¡to ¡help ¡them ¡on ¡difficult ¡ topic D. Satisfaction E. Privacy and Data Sharing

Summary Open Social Student Modeling knowledge-based + social guidance combined are more effective in guiding students to appropriate questions that ¡they ¡are ¡ready ¡to ¡handle ¡ than knowledge-based guidance alone. ¡ Evidence of Peer Guidance & Learning Effects

Future Adaptive navigation support Personalized guidance Privacy management Visualizing multiple peers’ models Collaboration features

Progressor ROCKS!

Sneak peek … Attempts Success Rate 250 80.00% 71.35% 205.73 68.39% 58.31% 200 60.00% 113.05 42.63% 150 125.5 40.00% 100 Progressor Progressor 80.81 QuizJET+IV 20.00% QuizJET+IV 50 QuizJET+Portal QuizJET+Portal JavaGuide JavaGuide 0 0.00% Follow me on @shoha99

Q&A Sharon Hsiao says Thank You ~ ihh4@pitt.edu @shoha99 Personalized Adaptive Web Systems http://adapt2.sis.pitt.edu/wiki