Project Activities
The research team will extend an existing intelligent tutor to deliver practice of pre-algebra skills. The computer tutor will be designed to deliver practice sessions that target discrete prerequisite math skills. The system will use a personalized model of each student's learning to determine when and how much practice is needed for each prerequisite skill in order to maximize progress. Using personalized models of instruction allows the computer tutor to target needed prerequisite skills for each student while avoiding unnecessary review of skills the student has already mastered. The new intervention will be tested using a randomized within-subjects experimental design.
Structured Abstract
Setting
The research setting is middle schools in a large, metropolitan school district in California.
Sample
The study population includes primarily urban, low-income, Hispanic students in grades 6-8.
The project extends the Bridge to Algebra cognitive tutor, currently produced by Carnegie Learning Inc., by creating a highly controlled system to deliver practice of prerequisite pre-algebra skills. The computer tutor will be designed to deliver practice sessions that target discrete prerequisite math skills. The system will use a personalized model of each student's learning to determine when and how much practice is needed for each prerequisite skill in order to maximize progress in the Bridge to Algebra tutor. Using personalized models of instruction allows the computer tutor to target needed prerequisite skills for each student while avoiding unnecessary review of skills the student has already mastered.
Research design and methods
The new intervention will be tested using a randomized within-subjects experimental design. To compare the effects of the intervention with the effects of existing instructional practice, half of the students, chosen randomly, use the new version of the Bridge to Algebra tutor in the fall semester, and the other half use the new version of the Bridge to Algebra tutor in the spring semester. This design compares the same student's progress during experimental treatment versus during existing instructional practice. Math performance will be measured at the beginning of the year, at the end of the fall semester just prior to the crossover from experimental to control condition (or vice versa), and at the end of the academic year.
Control condition
During the control periods, students use the existing Bridge to Algebra tutor.
Key measures
The primary measure consists of a random sample of math assessment items taken from the Trends in International Mathematics and Science Survey (TIMSS) grade 8 questions. Primary measures computed from data logged by the Bridge to Algebra tutor and intervention software include individual problem performance accuracy and latency.
Data analytic strategy
For the analysis of crossover efficacy studies, the research team is conducting repeated measures analyses of variance of TIMSS test scores and Bridge to Algebra dependent measures.
People and institutions involved
IES program contact(s)
Products and publications
Products from this project include software designed to aid middle school students in the learning of prerequisite pre-algebra skills, and published reports on the initial impact of the software on student achievement in pre-algebra and algebra.
Publications:
Book chapter
Pavlik Jr., P.I., Yudelson, M., and Koedinger, K. (2011). Using Contextual Factors Analysis to Explain Transfer of Least Common Multiple Skills. In G. Biswas, S. Bull, J. Kay, and A. Mitrovic (Eds.), Artificial Intelligence in Education (pp. 256-263). Berlin, Heidelberg: Springer.
Yudelson, M., Pavlik Jr., P.I., and Koedinger, K.R. (2011). User Modeling: A Notoriously Black Art. In J. Konstan, R. Conejo, J. Marzo, and N. Oliver (Eds.), User Modeling, Adaption and Personalization (pp. 317-328). Berlin, Heidelberg: Springer.
Journal article, monograph, or newsletter
Pavlik Jr., P.I., and Anderson, J.R. (2008). Using a Model to Compute the Optimal Schedule of Practice. Journal of Experimental Psychology: Applied, 14(2): 101-117.
Proceedings
Frishkoff, G., Levin, L., Pavlik, P., Idemaru, K., and de Jong, N. (2008). A Model-Based Approach to Second-Language Learning of Grammatical Constructions. In V. Sloutsky, B. Love and K. McRae (Eds.), Proceedings of the 30th Conference of the Cognitive Science Society (pp. 916-921). Washington, DC: Cognitive Science Society.
Koedinger, K., Pavlik Jr., P.I., Stamper, J., Nixon, T., and Ritter, S. (2011). Fair Blame Assignment in Student Modeling. In Proceedings of the 3rd International Conference on Educational Data Mining (pp. 91-100). Eindhoven, Netherlands: Educational Data Mining.
Pavlik Jr., P.I., and Toth, J. (2010). How to Build Bridges Between Intelligent Tutoring System Subfields of Research. In Proceedings of the 10th International Conference on Intelligent Tutoring Systems, Part II (pp. 103-112). Pittsburgh: Springer.
Pavlik Jr., P.I., Bolster, T., Wu, S., Koedinger, K.R., and MacWhinney, B. (2008). Using Optimally Selected Drill Practice to Train Basic Facts. In Proceedings of the 9th International Conference on Intelligent Tutoring Systems (pp. 593-602). Berlin, Germany: Springer.
Pavlik Jr., P.I., Cen, H., and Koedinger, K.R. (2009). Learning Factors Transfer Analysis: Using Learning Curve Analysis to Automatically Generate Domain Models. In Proceedings of the 2nd International Conference on Educational Data Mining (pp. 121-130). Cordoba, Spain: Educational Data Mining.
Pavlik Jr., P.I., Cen, H., and Koedinger, K.R. (2009). Performance Factors Analysis: A New Alternative to Knowledge Tracing. In Proceedings of the 14th International Conference on Artificial Intelligence in Education, Brighton, England (pp. 531-538). Amsterdam: IOS Press.
Pavlik, P.I., Cen, H., Wu, L., and Keodinger, K.R. (2008). Using Item-Type Performance Covariance to Improve the Skill Model of an Existing Tutor. In Proceedings of the 1st International Conference on Educational Data Mining (pp. 77-86). Montreal, Canada: UQAM.
Yudelson, M., Pavlik Jr., P.I., and Koedinger, K.R. (2011). Towards Better Understanding of Transfer in Cognitive Models of Practice. In Proceedings of the 3rd International Conference on Educational Data Mining (pp. 373-374). Eindhoven, Netherlands: Educational Data Mining.
Related projects
Questions about this project?
To answer additional questions about this project or provide feedback, please contact the program officer.
