%0 Book Section %1 Huang2015 %A Huang, Yun %A González-Brenes, José P. %A Brusilovsky, Peter %B Artificial Intelligence in Education: 17th International Conference, AIED 2015, Madrid, Spain, June 22-26, 2015. Proceedings %C Berlin %D 2015 %E Conati, Cristina %E Heffernan, Neil %E Mitrovic, Antonija %E Verdejo, M. Felisa %I Springer International Publishing %K %P 633--637 %R 10.1007/978-3-319-19773-9_79 %T Challenges of Using Observational Data to Determine the Importance of Example Usage %U https://doi.org/10.1007/978-3-319-19773-9_79 %X Educational interventions are often evaluated with randomized control trials, which can be very expensive to conduct. One of the promises of ``Big Data'' in education is to use non-experimental data to discover insights. We focus on studying the impact of example usage in a Java programming tutoring system using observational data. For this, we compare different formulations of a recently proposed generalized Knowledge Tracing framework called FAST. We discover that different formulations can have the same predictive performance; yet their coefficients may have opposite signs, which may lead researchers to contradictory conclusions. We discuss implications of using fully data-driven approaches to study non-experimental data. %@ 978-3-319-19773-9

@inbook{Huang2015, abstract = {Educational interventions are often evaluated with randomized control trials, which can be very expensive to conduct. One of the promises of ``Big Data'' in education is to use non-experimental data to discover insights. We focus on studying the impact of example usage in a Java programming tutoring system using observational data. For this, we compare different formulations of a recently proposed generalized Knowledge Tracing framework called FAST. We discover that different formulations can have the same predictive performance; yet their coefficients may have opposite signs, which may lead researchers to contradictory conclusions. We discuss implications of using fully data-driven approaches to study non-experimental data.}, added-at = {2023-12-12T19:03:58.000+0100}, address = {Berlin}, author = {Huang, Yun and Gonz{\'a}lez-Brenes, Jos{\'e} P. and Brusilovsky, Peter}, biburl = {https://www.bibsonomy.org/bibtex/28b3f6bd8b8e42cbb820714ea1c478067/admin}, booktitle = {Artificial Intelligence in Education: 17th International Conference, AIED 2015, Madrid, Spain, June 22-26, 2015. Proceedings}, doi = {10.1007/978-3-319-19773-9_79}, editor = {Conati, Cristina and Heffernan, Neil and Mitrovic, Antonija and Verdejo, M. Felisa}, interhash = {1b3dfbfe69a9b90d0f240ebb8098f3fb}, intrahash = {8b3f6bd8b8e42cbb820714ea1c478067}, isbn = {978-3-319-19773-9}, keywords = {}, pages = {633--637}, publisher = {Springer International Publishing}, timestamp = {2023-12-12T19:03:58.000+0100}, title = {Challenges of Using Observational Data to Determine the Importance of Example Usage}, url = {https://doi.org/10.1007/978-3-319-19773-9_79}, year = 2015 }